Clinical Validation across 89 Radiographic Condition Classifiers

AI Condition Classifier Performance

Veterinary AI Classifier Performance Metrics

Name: Vetology AI Classifier Performance Metrics
Creator: Vetology AI
Published: 2026-03-27
License: https://vetology.net/terms/

89+ Condition Classifiers Built on 300,000 Expert-Reviewed Veterinary Imaging Cases

Vetology’s AI classifiers have been validated using a foundation of over 300,000 multi-image patient cases from real-world veterinary practices. Below you’ll find transparent performance data including sensitivity, specificity, Radiologist Agreement Rates and case counts for every classifier. Use the dropdown below each condition classifier to see even more in-depth data.

This deep, real-world dataset helps our AI support veterinarians with reliable screening results at the point of care.

Vetology is the ONLY veterinary AI imaging platform that offers this level of transparency. We publish complete performance metrics to help you make informed decisions about diagnostic accuracy and clinical implementation.

While these results offer meaningful insight into expected performance, real-world factors such as image quality, positioning, and patient variability can influence accuracy in clinical settings.

Jump to Classifier Metrics

What Makes Our Data Different

Full Transparency

Vetology is the ONLY veterinary AI company that publishes classifier-level performance metrics. We believe you deserve to see exactly how our technology performs on every condition we detect.

Species-Specific Classifier Development

Canine and feline conditions present differently. Vetology develops separate, species-specific classifiers rather than one shared model, ensuring more accurate and context-appropriate results for each patient type.

Real-World Testing

All metrics are based on actual clinical cases from veterinary practices, not synthetic datasets or cherry-picked examples. This reflects real-world diagnostic performance.

Understanding Our AI Virtual Radiologist Report Screening Metrics

Why Radiologist Benchmarking Matters

Instead of comparing AI to a theoretical “perfect” standard, Vetology benchmarks performance against the Radiologist Agreement Rates: how often multiple boarded radiologists reach the same interpretation. This provides realistic context for conditions where even experts may disagree.

When sensitivity or specificity approaches or exceeds this rate, it indicates performance comparable to specialist-level interpretation for that finding.

These measures help you understand when AI can reinforce diagnostic confidence and when a case may benefit from further review.

Making Sense of the Metrics

Radiologist Agreement Rate

The percentage of cases where multiple board-certified veterinary radiologists independently arrive at the same diagnosis. This serves as a benchmark for inherent diagnostic difficulty.

Conditions with lower Radiologist Agreement Rates are more subjective or challenging, even for specialists. When AI performance approaches or exceeds the Radiologist Agreement Rate, it demonstrates specialist-comparable accuracy.

The percentage of cases where multiple board-certified veterinary radiologists independently arrive at the same diagnosis. This serves as a benchmark for inherent diagnostic difficulty.Conditions with lower Radiologist Agreement Rates are more subjective or challenging, even for specialists. When AI performance approaches or exceeds the Radiologist Agreement Rate, it demonstrates specialist-comparable accuracy.

Total Test Cases

The number of independent radiographic studies used to validate AI performance for each condition. Larger test sets provide more reliable performance estimates. All cases were diagnosed by United States board-certified veterinary radiologists and represent diverse presentations across multiple clinical settings.

Sensitivity

How reliably the AI detects a condition when it is present.
(True Positive Rate)

The proportion of actual positive cases (condition present) that the AI correctly identifies. For example, 89% sensitivity means the AI correctly detects the condition in 89 out of 100 cases where it is actually present. Higher sensitivity reduces false negatives—cases where a condition exists but goes undetected.

Specificity

How accurately the AI rules out a condition when it is absent.
(True Negative Rate)

The proportion of actual negative cases (condition absent) that the AI correctly identifies. For example, 92% specificity means the AI correctly rules out the condition in 92 out of 100 cases where it is actually absent. Higher specificity reduces false positives—cases where the AI flags a condition that is not present.

Confusion Matrix (TP, FN, TN, FP)

Clinical meaning: The confusion matrix lets you see the actual case counts behind every percentage. A classifier with 90% sensitivity and 10 positive test cases missed 1 case. A classifier with 90% sensitivity and 1,000 positive test cases missed 100 cases. The percentages are identical, but the confusion matrix tells you the scale. It also lets you see the absolute number of false positives — which directly affects how often the AI sends you down a path that turns out to be nothing.

The raw counts of True Positives (correctly flagged), False Negatives (missed), True Negatives (correctly ruled out), and False Positives (incorrectly flagged). These four numbers are the foundation from which sensitivity, specificity, and every other metric is calculated.Clinical meaning: The confusion matrix lets you see the actual case counts behind every percentage. A classifier with 90% sensitivity and 10 positive test cases missed 1 case. A classifier with 90% sensitivity and 1,000 positive test cases missed 100 cases. The percentages are identical, but the confusion matrix tells you the scale. It also lets you see the absolute number of false positives — which directly affects how often the AI sends you down a path that turns out to be nothing.

95% Confidence Intervals

(for Sensitivity and Specificity)

Clinical meaning: If a classifier shows 89.51% sensitivity with a tight confidence interval, you can trust that number. If the interval is wide, you may want to weight your own clinical findings more heavily.

A single sensitivity number is a point estimate. The confidence interval tells you the range within which the true sensitivity likely falls. Tighter intervals mean more certainty; wider intervals mean the classifier was validated on fewer cases for that condition.Clinical meaning: If a classifier shows 89.51% sensitivity with a tight confidence interval, you can trust that number. If the interval is wide, you may want to weight your own clinical findings more heavily.

Area Under Curve (AUC)

Clinical meaning: AUC gives you a single number that captures overall discriminative ability. A classifier with an AUC of 0.94 (like Heart Failure Left in canine thorax) is performing at a high level – it reliably separates patients who have the condition from those who do not.

AUC values between 0.80 and 0.90 are generally considered “excellent” discrimination, and values above 0.90 are considered “outstanding.”^[1]

AUC measures how well the classifier distinguishes between positive and negative cases across all possible decision thresholds. It is reported on a scale from 0 to 1, where 1 represents a classifier that never makes a mistake and 0.5 represents random chance.Clinical meaning: AUC gives you a single number that captures overall discriminative ability. A classifier with an AUC of 0.94 (like Heart Failure Left in canine thorax) is performing at a high level – it reliably separates patients who have the condition from those who do not.AUC values between 0.80 and 0.90 are generally considered “excellent” discrimination, and values above 0.90 are considered “outstanding.”^[1]

Positive Predictive Value (PPV)

Clinical meaning: A high PPV means that when the AI flags something, it is very likely real. A lower PPV in a rare condition is expected – it does not mean the classifier is unreliable, it means the condition is uncommon and you should correlate with clinical signs.

When the classifier flags a finding as positive, PPV tells you the probability that the patient actually has the condition. This metric is directly affected by how common the condition is in the test population.Clinical meaning: A high PPV means that when the AI flags something, it is very likely real. A lower PPV in a rare condition is expected – it does not mean the classifier is unreliable, it means the condition is uncommon and you should correlate with clinical signs.

Negative Predictive Value (NPV)

Clinical meaning: A high NPV gives you confidence that a negative AI result genuinely means “nothing concerning here.” This is especially valuable for conditions where missing a diagnosis carries serious consequences.

When the classifier reports no finding, NPV tells you the probability that the patient truly does not have the condition. For screening purposes, this is one of the most important metrics available.Clinical meaning: A high NPV gives you confidence that a negative AI result genuinely means “nothing concerning here.” This is especially valuable for conditions where missing a diagnosis carries serious consequences.

Overall Accuracy

Clinical meaning: Use accuracy as a starting point, then look at the metrics that tell you how it performs. Check sensitivity to see how well it catches true positives, and specificity to see how well it avoids false alarms. Then check prevalence in the test set: if a condition only appears in 2% of cases, a classifier that always says “negative” would still show 98% accuracy while missing every positive case.

When accuracy is high but prevalence is low, sensitivity and NPV are the metrics that tell you whether the classifier is actually doing its job.

The percentage of all cases (both positive and negative) that the classifier categorized correctly. While useful as a general performance summary, accuracy alone can be misleading when a condition is rare or common.Clinical meaning: Use accuracy as a starting point, then look at the metrics that tell you how it performs. Check sensitivity to see how well it catches true positives, and specificity to see how well it avoids false alarms. Then check prevalence in the test set: if a condition only appears in 2% of cases, a classifier that always says “negative” would still show 98% accuracy while missing every positive case.When accuracy is high but prevalence is low, sensitivity and NPV are the metrics that tell you whether the classifier is actually doing its job.

Balanced Accuracy

Clinical meaning: When a condition is rare (say it appears in only 3% of cases) overall accuracy can look impressive while the classifier misses most positive cases. Balanced accuracy corrects for this by weighting positive and negative performance equally. If you see a large gap between overall accuracy and balanced accuracy, it usually means the condition is rare and the classifier leans on “negative” predictions.

Balanced accuracy gives you the fairer picture.

The average of sensitivity and specificity, giving equal weight to how well the classifier catches true positives and how well it avoids false positives. Unlike overall accuracy, balanced accuracy is not skewed by class imbalance.Clinical meaning: When a condition is rare (say it appears in only 3% of cases) overall accuracy can look impressive while the classifier misses most positive cases. Balanced accuracy corrects for this by weighting positive and negative performance equally. If you see a large gap between overall accuracy and balanced accuracy, it usually means the condition is rare and the classifier leans on “negative” predictions.Balanced accuracy gives you the fairer picture.

Matthews Correlation Coefficient (MCC)

Clinical meaning: MCC answers the question: “Taking everything into account (true positives, true negatives, false positives, and false negatives) how well is this classifier actually performing?”

An MCC above 0.7 indicates strong agreement between the classifier and ground truth. Unlike accuracy, MCC cannot be inflated by class imbalance, which makes it especially valuable for rare conditions.

A single number that summarizes classifier quality using all four confusion matrix values (TP, FP, TN, FN). MCC ranges from −1 (total disagreement) through 0 (no better than random) to +1 (perfect prediction). It is widely regarded as one of the most informative single metrics for imbalanced datasets.Clinical meaning: MCC answers the question: “Taking everything into account (true positives, true negatives, false positives, and false negatives) how well is this classifier actually performing?”An MCC above 0.7 indicates strong agreement between the classifier and ground truth. Unlike accuracy, MCC cannot be inflated by class imbalance, which makes it especially valuable for rare conditions.

Prevalence in Test Set

The proportion of positive cases in the validation dataset. This is the denominator that drives PPV and NPV calculations.

Clinical meaning: Knowing the test set prevalence helps you calibrate expectations. If your practice population has a higher prevalence of a condition than the test set (e.g., you are a cardiology referral center), the real-world PPV for your caseload will be higher than the reported number.

The proportion of positive cases in the validation dataset. This is the denominator that drives PPV and NPV calculations.Clinical meaning: Knowing the test set prevalence helps you calibrate expectations. If your practice population has a higher prevalence of a condition than the test set (e.g., you are a cardiology referral center), the real-world PPV for your caseload will be higher than the reported number.

Clinical Urgency Rating

Clinical meaning: Urgency ratings help you prioritize your review queue. When multiple AI findings appear on a single study, urgency tells you where to look first.

Each classifier is tagged with a clinical urgency level: Critical, High, Moderate, or Low. These ratings are defined through consensus by Vetology’s in-house board-certified veterinary radiologists based on the clinical consequence of a delayed or missed detection – not the AI’s confidence level.Clinical meaning: Urgency ratings help you prioritize your review queue. When multiple AI findings appear on a single study, urgency tells you where to look first.

Ground Truth Case Counts

(Positive + Negative)

The exact number of board-certified veterinary radiologist-reviewed cases used to validate each classifier, broken into positive and negative counts.

Clinical meaning: More cases mean more reliable statistics. When you see a classifier validated on 10,951 cases, you know those performance numbers are robust. This also helps you evaluate whether the validation dataset was balanced enough to produce meaningful results.

The exact number of board-certified veterinary radiologist-reviewed cases used to validate each classifier, broken into positive and negative counts.Clinical meaning: More cases mean more reliable statistics. When you see a classifier validated on 10,951 cases, you know those performance numbers are robust. This also helps you evaluate whether the validation dataset was balanced enough to produce meaningful results.

Model Release Date

The date each classifier model was initially released or, in most cases, when it was most recently retrained and re-released.

Clinical meaning: Knowing the date the model was released tells you how current the model is. At Vetology we regularly retrain classifiers to ensure we are incorporating the latest validation data and performance improvements. Artificial Intelligence is changing fast! Our models are too.

The date each classifier model was initially released or, in most cases, when it was most recently retrained and re-released.Clinical meaning: Knowing the date the model was released tells you how current the model is. At Vetology we regularly retrain classifiers to ensure we are incorporating the latest validation data and performance improvements. Artificial Intelligence is changing fast! Our models are too.

Vetology’s Published Condition Classifier Metrics

Our industry-leading transparency allows you to evaluate our technology objectively.

Real-World Validation: Tested on a foundation of 300,000+ multi-image patient cases from veterinary practices.
Radiologist Benchmarking: Performance compared directly to board-certified veterinary radiologists
Comprehensive Coverage: Canine and feline imaging across thorax, abdomen, spine, and musculoskeletal studies

Vetology AI Classifier Performance Metrics

Veterinary Radiology AI — Clinical Validation Data by Physiological Region

Built on a foundation of 300,000 board certified veterinary radiologist-reviewed multi-image patient cases

Understanding These Metrics ▼

Sensitivity

How often the AI correctly detects a condition when it is actually present. Think of it as the "catch rate." A sensitivity of 89% means the AI catches 89 out of every 100 cases where the condition exists. Higher is better -- you want fewer missed findings.

Specificity

How often the AI correctly identifies a normal case as normal. A specificity of 92% means that when there is no finding, the AI agrees 92 out of 100 times. Higher specificity means fewer false alarms that could lead to unnecessary follow-up.

95% Confidence Interval

A range showing how precise the sensitivity or specificity measurement is. A narrower range means we tested more cases and can be more certain of the result. For example, "85% - 93%" means the true performance most likely falls within that range.

Radiologist Agreement Rate

How often board-certified veterinary radiologists agree with each other when reading the same images. Some findings are straightforward and radiologists almost always agree; others are more subjective. This number gives you context -- if even specialists disagree 30% of the time on a finding, an AI performing in that range is working within the natural variability of expert interpretation.

Area Under Curve (AUC)

A single number that summarizes overall classifier quality. An AUC of 1.0 would be a perfect classifier; 0.5 would be no better than flipping a coin. In practice, values above 0.85 indicate strong performance. This is a metric primarily used by data scientists to compare classifier models.

Positive Predictive Value (PPV)

When the AI flags a finding, how often is it actually there? A PPV of 80% means that 8 out of 10 times the AI says "positive," the condition is truly present. PPV depends heavily on how common the condition is in practice, so we calculate it using real-world condition frequency from our clinical case database (a Bayesian adjustment) rather than the test set, giving you a number that better reflects what you would see in your clinic.

Negative Predictive Value (NPV)

When the AI says a finding is not present, how often is it right? An NPV of 99% means you can be highly confident in a negative result. Like PPV, we calculate NPV using real-world condition frequency from our clinical case database (a Bayesian adjustment) so the value reflects actual clinical practice rather than test set composition.

Prevalence

How common this condition is in real-world veterinary practice. Rather than using the test set (which may over- or under-represent certain conditions), we calculate prevalence from our clinical case database of real patient reports. This gives you a more accurate picture of how often you are likely to encounter each condition. Prevalence matters because it directly affects how much you should trust a positive or negative AI result.

Overall Accuracy

The percentage of all cases -- both positive and negative -- that the classifier got right. While it sounds like the most important number, accuracy can be misleading for rare conditions. For example, a condition that appears in only 1% of cases could show 99% accuracy simply by always saying "not present." That is why we report sensitivity and specificity alongside accuracy to give the full picture.

Balanced Accuracy

The average of sensitivity and specificity, giving equal weight to detecting positive cases and correctly ruling out negative ones. This is a more reliable measure than overall accuracy, especially for conditions that are rare in practice, because it is not skewed by how common or uncommon the condition is.

Matthews Correlation Coefficient (MCC)

A single number that captures overall classifier quality, accounting for true positives, false positives, true negatives, and false negatives all at once. MCC ranges from -1 (the classifier gets everything wrong) through 0 (no better than random) to +1 (perfect). Data scientists consider MCC one of the most balanced and informative single metrics, particularly when conditions are rare.

Confusion Matrix

The four raw counts behind all the other metrics. True Positives: the AI said yes and was right. False Negatives: the AI missed it. True Negatives: the AI said no and was right. False Positives: the AI flagged something that was not there. Every metric on this page is derived from these four numbers.

Clinical Urgency

How time-sensitive the condition is in clinical practice, rated by our board-certified veterinary radiologists. Critical conditions (such as heart failure or pericardial effusion) require immediate attention. High-urgency findings should be addressed promptly. This helps you prioritize which AI findings to act on first.

Ground Truth Cases

The number of cases used to evaluate this classifier, reviewed and labeled by board-certified veterinary radiologists. More cases generally means more reliable metrics. The count is split into positive (condition present) and negative (condition absent) so you can see how balanced the test was.

Canine Thorax (26 conditions)

Heart Failure

Canine

Clinical Urgency: Critical

Sensitivity

89.51%

95% Confidence Interval: 84.79% - 94.23%

Specificity

92.07%

95% Confidence Interval: 91.56% - 92.58%

Radiologist Agreement Rate: 90%

Total Cases Evaluated: 10,951

▼ View Full Metrics

Area Under Curve (AUC) 0.9522

Balanced Accuracy 90.79%

Overall Accuracy 92.03%

Matthews Correlation Coefficient (MCC) 0.3417

Positive Predictive Value (PPV) 11.88%

Negative Predictive Value (NPV) 99.86%

Prevalence 1.18%

Confusion Matrix

True Positives (TP) 145

False Negatives (FN) 17

True Negatives (TN) 9,933

False Positives (FP) 856

Ground Truth Cases

Positive Cases 162

Negative Cases 10,789

Release Date September 2025

Cardiomegaly

Canine

Clinical Urgency: Moderate

Sensitivity

75.64%

95% Confidence Interval: 74.22% - 77.05%

Specificity

86.28%

95% Confidence Interval: 85.69% - 86.86%

Radiologist Agreement Rate: 93%

Total Cases Evaluated: 16,899

▼ View Full Metrics

Area Under Curve (AUC) 0.8848

Balanced Accuracy 80.96%

Overall Accuracy 84.05%

Matthews Correlation Coefficient (MCC) 0.5696

Positive Predictive Value (PPV) 16.79%

Negative Predictive Value (NPV) 98.98%

Prevalence 3.53%

Confusion Matrix

True Positives (TP) 2,679

False Negatives (FN) 863

True Negatives (TN) 11,524

False Positives (FP) 1,833

Ground Truth Cases

Positive Cases 3,542

Negative Cases 13,357

Release Date September 2025

Left-Sided Cardiomegaly

Canine

Clinical Urgency: Moderate

Sensitivity

85.4%

95% Confidence Interval: 84.51% - 86.28%

Specificity

78.71%

95% Confidence Interval: 77.86% - 79.57%

Radiologist Agreement Rate: 77%

Total Cases Evaluated: 14,936

▼ View Full Metrics

Area Under Curve (AUC) 0.8691

Balanced Accuracy 82.05%

Overall Accuracy 81.43%

Matthews Correlation Coefficient (MCC) 0.6308

Positive Predictive Value (PPV) 10.8%

Negative Predictive Value (NPV) 99.44%

Prevalence 2.93%

Confusion Matrix

True Positives (TP) 5,193

False Negatives (FN) 888

True Negatives (TN) 6,970

False Positives (FP) 1,885

Ground Truth Cases

Positive Cases 6,081

Negative Cases 8,855

Release Date September 2025

Thorax Left Atrial Enlargement

Canine

Clinical Urgency: Moderate

Sensitivity

84.44%

95% Confidence Interval: 82.17% - 86.7%

Specificity

83.48%

95% Confidence Interval: 82.72% - 84.23%

Radiologist Agreement Rate: 54%

Total Cases Evaluated: 10,291

▼ View Full Metrics

Area Under Curve (AUC) 0.9048

Balanced Accuracy 83.96%

Overall Accuracy 83.57%

Matthews Correlation Coefficient (MCC) 0.4743

Positive Predictive Value (PPV) 11.17%

Negative Predictive Value (NPV) 99.54%

Prevalence 2.4%

Confusion Matrix

True Positives (TP) 830

False Negatives (FN) 153

True Negatives (TN) 7,770

False Positives (FP) 1,538

Ground Truth Cases

Positive Cases 983

Negative Cases 9,308

Release Date December 2025

Left Auricular Appendage Enlargement

Canine

Clinical Urgency: Moderate

Sensitivity

84.65%

95% Confidence Interval: 83.64% - 85.66%

Specificity

84.51%

95% Confidence Interval: 84.13% - 84.88%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 40,272

▼ View Full Metrics

Area Under Curve (AUC) 0.9044

Balanced Accuracy 84.58%

Overall Accuracy 84.53%

Matthews Correlation Coefficient (MCC) 0.5296

Positive Predictive Value (PPV) 1.94%

Negative Predictive Value (NPV) 99.93%

Prevalence 0.36%

Confusion Matrix

True Positives (TP) 4,136

False Negatives (FN) 750

True Negatives (TN) 29,904

False Positives (FP) 5,482

Ground Truth Cases

Positive Cases 4,886

Negative Cases 35,386

Release Date December 2025

Thorax Heart Based Mass

Canine

Clinical Urgency: Moderate

Sensitivity

71.7%

95% Confidence Interval: 64.7% - 78.7%

Specificity

76.04%

95% Confidence Interval: 75.27% - 76.8%

Radiologist Agreement Rate: 89%

Total Cases Evaluated: 12,064

▼ View Full Metrics

Area Under Curve (AUC) 0.7975

Balanced Accuracy 73.87%

Overall Accuracy 75.98%

Matthews Correlation Coefficient (MCC) 0.1264

Positive Predictive Value (PPV) 0.63%

Negative Predictive Value (NPV) 99.92%

Prevalence 0.21%

Confusion Matrix

True Positives (TP) 114

False Negatives (FN) 45

True Negatives (TN) 9,052

False Positives (FP) 2,853

Ground Truth Cases

Positive Cases 159

Negative Cases 11,905

Release Date December 2025

Pulmonary Venous Enlargement

Canine

Clinical Urgency: Moderate

Sensitivity

82.41%

95% Confidence Interval: 75.23% - 89.59%

Specificity

83.76%

95% Confidence Interval: 83.19% - 84.34%

Radiologist Agreement Rate: 69%

Total Cases Evaluated: 15,962

▼ View Full Metrics

Area Under Curve (AUC) 0.8772

Balanced Accuracy 83.09%

Overall Accuracy 83.76%

Matthews Correlation Coefficient (MCC) 0.1455

Positive Predictive Value (PPV) 3.74%

Negative Predictive Value (NPV) 99.84%

Prevalence 0.76%

Confusion Matrix

True Positives (TP) 89

False Negatives (FN) 19

True Negatives (TN) 13,280

False Positives (FP) 2,574

Ground Truth Cases

Positive Cases 108

Negative Cases 15,854

Release Date September 2025

Pulmonary Arterial Enlargement

Canine

Clinical Urgency: Moderate

Sensitivity

66.74%

95% Confidence Interval: 62.49% - 70.99%

Specificity

77.34%

95% Confidence Interval: 76.93% - 77.75%

Radiologist Agreement Rate: 78%

Total Cases Evaluated: 40,535

▼ View Full Metrics

Area Under Curve (AUC) 0.7739

Balanced Accuracy 72.04%

Overall Accuracy 77.22%

Matthews Correlation Coefficient (MCC) 0.1121

Positive Predictive Value (PPV) 2.35%

Negative Predictive Value (NPV) 99.65%

Prevalence 0.81%

Confusion Matrix

True Positives (TP) 315

False Negatives (FN) 157

True Negatives (TN) 30,985

False Positives (FP) 9,078

Ground Truth Cases

Positive Cases 472

Negative Cases 40,063

Release Date December 2025

Pulmonary Arterial Enlargement

Canine

Clinical Urgency: Moderate

Sensitivity

68.15%

95% Confidence Interval: 62.35% - 73.94%

Specificity

64.66%

95% Confidence Interval: 63.89% - 65.43%

Radiologist Agreement Rate: 80%

Total Cases Evaluated: 14,920

▼ View Full Metrics

Area Under Curve (AUC) 0.6965

Balanced Accuracy 66.4%

Overall Accuracy 64.72%

Matthews Correlation Coefficient (MCC) 0.0874

Positive Predictive Value (PPV) 1.55%

Negative Predictive Value (NPV) 99.6%

Prevalence 0.81%

Confusion Matrix

True Positives (TP) 169

False Negatives (FN) 79

True Negatives (TN) 9,487

False Positives (FP) 5,185

Ground Truth Cases

Positive Cases 248

Negative Cases 14,672

Release Date September 2025

Caudodorsal Alveolar Pattern

Canine

Clinical Urgency: High

Sensitivity

82.06%

95% Confidence Interval: 78.92% - 85.19%

Specificity

88.05%

95% Confidence Interval: 87.51% - 88.59%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 14,504

▼ View Full Metrics

Area Under Curve (AUC) 0.9124

Balanced Accuracy 85.05%

Overall Accuracy 87.81%

Matthews Correlation Coefficient (MCC) 0.3857

Positive Predictive Value (PPV) 21.98%

Negative Predictive Value (NPV) 99.17%

Prevalence 3.94%

Confusion Matrix

True Positives (TP) 471

False Negatives (FN) 103

True Negatives (TN) 12,265

False Positives (FP) 1,665

Ground Truth Cases

Positive Cases 574

Negative Cases 13,930

Release Date September 2025

Cranioventral Alveolar Pattern

Canine

Clinical Urgency: High

Sensitivity

83.67%

95% Confidence Interval: 81.98% - 85.36%

Specificity

85.09%

95% Confidence Interval: 84.47% - 85.72%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 14,334

▼ View Full Metrics

Area Under Curve (AUC) 0.8949

Balanced Accuracy 84.38%

Overall Accuracy 84.91%

Matthews Correlation Coefficient (MCC) 0.5403

Positive Predictive Value (PPV) 18.71%

Negative Predictive Value (NPV) 99.22%

Prevalence 3.94%

Confusion Matrix

True Positives (TP) 1,537

False Negatives (FN) 300

True Negatives (TN) 10,634

False Positives (FP) 1,863

Ground Truth Cases

Positive Cases 1,837

Negative Cases 12,497

Release Date September 2025

Interstitial Pattern

Canine

Clinical Urgency: Moderate

Sensitivity

76.46%

95% Confidence Interval: 72.59% - 80.32%

Specificity

71.33%

95% Confidence Interval: 70.59% - 72.07%

Radiologist Agreement Rate: 61%

Total Cases Evaluated: 14,794

▼ View Full Metrics

Area Under Curve (AUC) 0.8031

Balanced Accuracy 73.89%

Overall Accuracy 71.49%

Matthews Correlation Coefficient (MCC) 0.1813

Positive Predictive Value (PPV) 14.66%

Negative Predictive Value (NPV) 97.92%

Prevalence 6.05%

Confusion Matrix

True Positives (TP) 354

False Negatives (FN) 109

True Negatives (TN) 10,222

False Positives (FP) 4,109

Ground Truth Cases

Positive Cases 463

Negative Cases 14,331

Release Date September 2025

Miliary Pulmonary Pattern

Canine

Clinical Urgency: High

Sensitivity

84.72%

95% Confidence Interval: 81.01% - 88.44%

Specificity

84.27%

95% Confidence Interval: 83.87% - 84.67%

Radiologist Agreement Rate: 89%

Total Cases Evaluated: 31,904

▼ View Full Metrics

Area Under Curve (AUC) 0.9091

Balanced Accuracy 84.5%

Overall Accuracy 84.28%

Matthews Correlation Coefficient (MCC) 0.1963

Positive Predictive Value (PPV) 0.64%

Negative Predictive Value (NPV) 99.98%

Prevalence 0.12%

Confusion Matrix

True Positives (TP) 305

False Negatives (FN) 55

True Negatives (TN) 26,583

False Positives (FP) 4,961

Ground Truth Cases

Positive Cases 360

Negative Cases 31,544

Release Date December 2025

Pulmonary Mass(es)

Canine

Clinical Urgency: Moderate

Sensitivity

76.35%

95% Confidence Interval: 69.51% - 83.2%

Specificity

74.2%

95% Confidence Interval: 73.49% - 74.9%

Radiologist Agreement Rate: 74%

Total Cases Evaluated: 15,118

▼ View Full Metrics

Area Under Curve (AUC) 0.8093

Balanced Accuracy 75.27%

Overall Accuracy 74.22%

Matthews Correlation Coefficient (MCC) 0.1130

Positive Predictive Value (PPV) 4.09%

Negative Predictive Value (NPV) 99.54%

Prevalence 1.42%

Confusion Matrix

True Positives (TP) 113

False Negatives (FN) 35

True Negatives (TN) 11,107

False Positives (FP) 3,863

Ground Truth Cases

Positive Cases 148

Negative Cases 14,970

Release Date September 2025

Pulmonary Nodules Cranioventral

Canine

Clinical Urgency: Moderate

Sensitivity

80.09%

95% Confidence Interval: 74.94% - 85.24%

Specificity

88.43%

95% Confidence Interval: 88.03% - 88.83%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 25,115

▼ View Full Metrics

Area Under Curve (AUC) 0.9158

Balanced Accuracy 84.26%

Overall Accuracy 88.35%

Matthews Correlation Coefficient (MCC) 0.1999

Positive Predictive Value (PPV) 4.33%

Negative Predictive Value (NPV) 99.85%

Prevalence 0.65%

Confusion Matrix

True Positives (TP) 185

False Negatives (FN) 46

True Negatives (TN) 22,005

False Positives (FP) 2,879

Ground Truth Cases

Positive Cases 231

Negative Cases 24,884

Release Date February 2026

Pulmonary Nodules Caudodorsal

Canine

Clinical Urgency: Moderate

Sensitivity

74.81%

95% Confidence Interval: 69.64% - 79.99%

Specificity

91.09%

95% Confidence Interval: 90.78% - 91.39%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 33,303

▼ View Full Metrics

Area Under Curve (AUC) 0.8961

Balanced Accuracy 82.95%

Overall Accuracy 90.96%

Matthews Correlation Coefficient (MCC) 0.2021

Positive Predictive Value (PPV) 2.86%

Negative Predictive Value (NPV) 99.9%

Prevalence 0.35%

Confusion Matrix

True Positives (TP) 202

False Negatives (FN) 68

True Negatives (TN) 30,089

False Positives (FP) 2,944

Ground Truth Cases

Positive Cases 270

Negative Cases 33,033

Release Date February 2026

Hypoinflation

Canine

Clinical Urgency: Low

Sensitivity

63.34%

95% Confidence Interval: 60.67% - 66.01%

Specificity

58.25%

95% Confidence Interval: 57.54% - 58.96%

Radiologist Agreement Rate: 77%

Total Cases Evaluated: 19,609

▼ View Full Metrics

Area Under Curve (AUC) 0.5994

Balanced Accuracy 60.79%

Overall Accuracy 58.58%

Matthews Correlation Coefficient (MCC) 0.1066

Positive Predictive Value (PPV) 2.5%

Negative Predictive Value (NPV) 98.95%

Prevalence 1.66%

Confusion Matrix

True Positives (TP) 793

False Negatives (FN) 459

True Negatives (TN) 10,693

False Positives (FP) 7,664

Ground Truth Cases

Positive Cases 1,252

Negative Cases 18,357

Release Date September 2025

Bronchial Pattern

Canine

Clinical Urgency: Moderate

Sensitivity

52.24%

95% Confidence Interval: 49.4% - 55.09%

Specificity

60.62%

95% Confidence Interval: 59.94% - 61.3%

Radiologist Agreement Rate: 76%

Total Cases Evaluated: 21,146

▼ View Full Metrics

Area Under Curve (AUC) 0.5873

Balanced Accuracy 56.43%

Overall Accuracy 60.15%

Matthews Correlation Coefficient (MCC) 0.0603

Positive Predictive Value (PPV) 8.38%

Negative Predictive Value (NPV) 94.85%

Prevalence 6.45%

Confusion Matrix

True Positives (TP) 617

False Negatives (FN) 564

True Negatives (TN) 12,103

False Positives (FP) 7,862

Ground Truth Cases

Positive Cases 1,181

Negative Cases 19,965

Release Date September 2025

Bronchiectasis

Canine

Clinical Urgency: Moderate

Sensitivity

54.5%

95% Confidence Interval: 47.4% - 61.6%

Specificity

75.09%

95% Confidence Interval: 74.29% - 75.89%

Radiologist Agreement Rate: 81%

Total Cases Evaluated: 11,433

▼ View Full Metrics

Area Under Curve (AUC) 0.7335

Balanced Accuracy 64.79%

Overall Accuracy 74.75%

Matthews Correlation Coefficient (MCC) 0.0867

Positive Predictive Value (PPV) 0.78%

Negative Predictive Value (NPV) 99.78%

Prevalence 0.36%

Confusion Matrix

True Positives (TP) 103

False Negatives (FN) 86

True Negatives (TN) 8,443

False Positives (FP) 2,801

Ground Truth Cases

Positive Cases 189

Negative Cases 11,244

Release Date September 2025

Intrathoracic Tracheal Narrowing

Canine

Clinical Urgency: High

Sensitivity

82.26%

95% Confidence Interval: 79.35% - 85.16%

Specificity

83.3%

95% Confidence Interval: 82.94% - 83.66%

Radiologist Agreement Rate: 80%

Total Cases Evaluated: 42,269

▼ View Full Metrics

Area Under Curve (AUC) 0.8504

Balanced Accuracy 82.78%

Overall Accuracy 83.28%

Matthews Correlation Coefficient (MCC) 0.2136

Positive Predictive Value (PPV) 4.1%

Negative Predictive Value (NPV) 99.82%

Prevalence 0.86%

Confusion Matrix

True Positives (TP) 547

False Negatives (FN) 118

True Negatives (TN) 34,655

False Positives (FP) 6,949

Ground Truth Cases

Positive Cases 665

Negative Cases 41,604

Release Date December 2025

Pleural Fluid

Canine

Clinical Urgency: High

Sensitivity

75.15%

95% Confidence Interval: 71.85% - 78.45%

Specificity

75.34%

95% Confidence Interval: 74.47% - 76.22%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 10,045

▼ View Full Metrics

Area Under Curve (AUC) 0.7962

Balanced Accuracy 75.25%

Overall Accuracy 75.33%

Matthews Correlation Coefficient (MCC) 0.2787

Positive Predictive Value (PPV) 17.65%

Negative Predictive Value (NPV) 97.73%

Prevalence 6.57%

Confusion Matrix

True Positives (TP) 496

False Negatives (FN) 164

True Negatives (TN) 7,071

False Positives (FP) 2,314

Ground Truth Cases

Positive Cases 660

Negative Cases 9,385

Release Date September 2025

Obscuring Pleural Effusion

Canine

Clinical Urgency: Critical

Sensitivity

87.23%

95% Confidence Interval: 82.97% - 91.5%

Specificity

96.73%

95% Confidence Interval: 96.46% - 97.0%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 16,588

▼ View Full Metrics

Area Under Curve (AUC) 0.9715

Balanced Accuracy 91.98%

Overall Accuracy 96.59%

Matthews Correlation Coefficient (MCC) 0.4806

Positive Predictive Value (PPV) 21.92%

Negative Predictive Value (NPV) 99.86%

Prevalence 1.04%

Confusion Matrix

True Positives (TP) 205

False Negatives (FN) 30

True Negatives (TN) 15,818

False Positives (FP) 535

Ground Truth Cases

Positive Cases 235

Negative Cases 16,353

Release Date February 2026

Esophageal Enlargement

Canine

Clinical Urgency: Moderate

Sensitivity

87.58%

95% Confidence Interval: 86.1% - 89.06%

Specificity

50%

95% Confidence Interval: 49.07% - 50.94%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 12,919

▼ View Full Metrics

Area Under Curve (AUC) 0.7732

Balanced Accuracy 68.79%

Overall Accuracy 55.58%

Matthews Correlation Coefficient (MCC) 0.2688

Positive Predictive Value (PPV) 23.37%

Negative Predictive Value (NPV) 95.85%

Prevalence 14.83%

Confusion Matrix

True Positives (TP) 1,678

False Negatives (FN) 238

True Negatives (TN) 5,502

False Positives (FP) 5,501

Ground Truth Cases

Positive Cases 1,916

Negative Cases 11,003

Release Date September 2025

Cranial Mediastinal Mass

Canine

Clinical Urgency: Moderate

Sensitivity

71.33%

95% Confidence Interval: 67.57% - 75.08%

Specificity

82.94%

95% Confidence Interval: 82.46% - 83.43%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 23,975

▼ View Full Metrics

Area Under Curve (AUC) 0.8346

Balanced Accuracy 77.14%

Overall Accuracy 82.67%

Matthews Correlation Coefficient (MCC) 0.2115

Positive Predictive Value (PPV) 2.98%

Negative Predictive Value (NPV) 99.75%

Prevalence 0.73%

Confusion Matrix

True Positives (TP) 398

False Negatives (FN) 160

True Negatives (TN) 19,423

False Positives (FP) 3,994

Ground Truth Cases

Positive Cases 558

Negative Cases 23,417

Release Date February 2026

Mediastinal Shift Right

Canine

Clinical Urgency: Moderate

Sensitivity

80.97%

95% Confidence Interval: 76.08% - 85.87%

Specificity

78.06%

95% Confidence Interval: 77.66% - 78.45%

Radiologist Agreement Rate: 33%

Total Cases Evaluated: 42,495

▼ View Full Metrics

Area Under Curve (AUC) 0.8517

Balanced Accuracy 79.51%

Overall Accuracy 78.08%

Matthews Correlation Coefficient (MCC) 0.1078

Positive Predictive Value (PPV) 1.1%

Negative Predictive Value (NPV) 99.93%

Prevalence 0.3%

Confusion Matrix

True Positives (TP) 200

False Negatives (FN) 47

True Negatives (TN) 32,978

False Positives (FP) 9,270

Ground Truth Cases

Positive Cases 247

Negative Cases 42,248

Release Date December 2025

Mediastinal Shift Left

Canine

Clinical Urgency: Moderate

Sensitivity

74.16%

95% Confidence Interval: 69.19% - 79.13%

Specificity

74.41%

95% Confidence Interval: 74.0% - 74.82%

Radiologist Agreement Rate: 80%

Total Cases Evaluated: 43,910

▼ View Full Metrics

Area Under Curve (AUC) 0.8033

Balanced Accuracy 74.29%

Overall Accuracy 74.41%

Matthews Correlation Coefficient (MCC) 0.0910

Positive Predictive Value (PPV) 0.86%

Negative Predictive Value (NPV) 99.9%

Prevalence 0.3%

Confusion Matrix

True Positives (TP) 221

False Negatives (FN) 77

True Negatives (TN) 32,452

False Positives (FP) 11,160

Ground Truth Cases

Positive Cases 298

Negative Cases 43,612

Release Date December 2025

Canine Abdomen (22 conditions)

Hepatomegaly

Canine

Clinical Urgency: Moderate

Sensitivity

87.11%

95% Confidence Interval: 85.49% - 88.73%

Specificity

88.47%

95% Confidence Interval: 88.09% - 88.85%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 28,440

▼ View Full Metrics

Area Under Curve (AUC) 0.9402

Balanced Accuracy 87.79%

Overall Accuracy 88.39%

Matthews Correlation Coefficient (MCC) 0.4817

Positive Predictive Value (PPV) 25.66%

Negative Predictive Value (NPV) 99.34%

Prevalence 4.37%

Confusion Matrix

True Positives (TP) 1,426

False Negatives (FN) 211

True Negatives (TN) 23,713

False Positives (FP) 3,090

Ground Truth Cases

Positive Cases 1,637

Negative Cases 26,803

Release Date February 2026

Hepatic Mass

Canine

Clinical Urgency: Moderate

Sensitivity

77.65%

95% Confidence Interval: 75.13% - 80.17%

Specificity

84.48%

95% Confidence Interval: 83.8% - 85.15%

Radiologist Agreement Rate: 83%

Total Cases Evaluated: 12,140

▼ View Full Metrics

Area Under Curve (AUC) 0.8826

Balanced Accuracy 81.06%

Overall Accuracy 83.89%

Matthews Correlation Coefficient (MCC) 0.4291

Positive Predictive Value (PPV) 1.92%

Negative Predictive Value (NPV) 99.9%

Prevalence 0.39%

Confusion Matrix

True Positives (TP) 813

False Negatives (FN) 234

True Negatives (TN) 9,371

False Positives (FP) 1,722

Ground Truth Cases

Positive Cases 1,047

Negative Cases 11,093

Release Date November 2024

Microhepatia

Canine

Clinical Urgency: Moderate

Sensitivity

80.42%

95% Confidence Interval: 78.01% - 82.84%

Specificity

78.68%

95% Confidence Interval: 77.92% - 79.43%

Radiologist Agreement Rate: 84%

Total Cases Evaluated: 12,302

▼ View Full Metrics

Area Under Curve (AUC) 0.8624

Balanced Accuracy 79.55%

Overall Accuracy 78.82%

Matthews Correlation Coefficient (MCC) 0.3729

Positive Predictive Value (PPV) 0.11%

Negative Predictive Value (NPV) 99.99%

Prevalence 0.03%

Confusion Matrix

True Positives (TP) 834

False Negatives (FN) 203

True Negatives (TN) 8,863

False Positives (FP) 2,402

Ground Truth Cases

Positive Cases 1,037

Negative Cases 11,265

Release Date November 2024

Gallbladder/Intra-hepatic Biliary Mineral

Canine

Clinical Urgency: Moderate

Sensitivity

65.74%

95% Confidence Interval: 61.43% - 70.04%

Specificity

77.96%

95% Confidence Interval: 77.52% - 78.4%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 33,964

▼ View Full Metrics

Area Under Curve (AUC) 0.7685

Balanced Accuracy 71.85%

Overall Accuracy 77.79%

Matthews Correlation Coefficient (MCC) 0.1216

Positive Predictive Value (PPV) 1.68%

Negative Predictive Value (NPV) 99.75%

Prevalence 0.57%

Confusion Matrix

True Positives (TP) 307

False Negatives (FN) 160

True Negatives (TN) 26,114

False Positives (FP) 7,383

Ground Truth Cases

Positive Cases 467

Negative Cases 33,497

Release Date February 2026

Splenomegaly

Canine

Clinical Urgency: Moderate

Sensitivity

70%

95% Confidence Interval: 63.31% - 76.69%

Specificity

87.04%

95% Confidence Interval: 86.67% - 87.4%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 32,713

▼ View Full Metrics

Area Under Curve (AUC) 0.8437

Balanced Accuracy 78.52%

Overall Accuracy 86.94%

Matthews Correlation Coefficient (MCC) 0.1243

Positive Predictive Value (PPV) 3.05%

Negative Predictive Value (NPV) 99.8%

Prevalence 0.58%

Confusion Matrix

True Positives (TP) 126

False Negatives (FN) 54

True Negatives (TN) 28,316

False Positives (FP) 4,217

Ground Truth Cases

Positive Cases 180

Negative Cases 32,533

Release Date February 2026

Splenic Mass

Canine

Clinical Urgency: Moderate

Sensitivity

69.44%

95% Confidence Interval: 64.69% - 74.2%

Specificity

76.19%

95% Confidence Interval: 75.35% - 77.02%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 10,317

▼ View Full Metrics

Area Under Curve (AUC) 0.7835

Balanced Accuracy 72.82%

Overall Accuracy 75.95%

Matthews Correlation Coefficient (MCC) 0.1924

Positive Predictive Value (PPV) 3.5%

Negative Predictive Value (NPV) 99.5%

Prevalence 1.23%

Confusion Matrix

True Positives (TP) 250

False Negatives (FN) 110

True Negatives (TN) 7,586

False Positives (FP) 2,371

Ground Truth Cases

Positive Cases 360

Negative Cases 9,957

Release Date November 2024

Multi Label Gastric Contents

Canine

Clinical Urgency: Low

Sensitivity

70.94%

95% Confidence Interval: 69.21% - 72.67%

Specificity

66.21%

95% Confidence Interval: 64.49% - 67.93%

Radiologist Agreement Rate: 63%

Total Cases Evaluated: 5,562

▼ View Full Metrics

Area Under Curve (AUC) 0.7410

Balanced Accuracy 68.58%

Overall Accuracy 68.46%

Matthews Correlation Coefficient (MCC) 0.3713

Positive Predictive Value (PPV) 12.64%

Negative Predictive Value (NPV) 97.06%

Prevalence 6.45%

Confusion Matrix

True Positives (TP) 1,880

False Negatives (FN) 770

True Negatives (TN) 1,928

False Positives (FP) 984

Ground Truth Cases

Positive Cases 2,650

Negative Cases 2,912

Release Date November 2024

Mineral/Metal Gastric Material

Canine

Clinical Urgency: Moderate

Sensitivity

60.29%

95% Confidence Interval: 56.66% - 63.91%

Specificity

71.61%

95% Confidence Interval: 70.78% - 72.44%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 11,978

▼ View Full Metrics

Area Under Curve (AUC) 0.7253

Balanced Accuracy 65.95%

Overall Accuracy 70.95%

Matthews Correlation Coefficient (MCC) 0.1629

Positive Predictive Value (PPV) 0.97%

Negative Predictive Value (NPV) 99.74%

Prevalence 0.46%

Confusion Matrix

True Positives (TP) 422

False Negatives (FN) 278

True Negatives (TN) 8,076

False Positives (FP) 3,202

Ground Truth Cases

Positive Cases 700

Negative Cases 11,278

Release Date November 2024

Mineral/Metal SI Material

Canine

Clinical Urgency: Moderate

Sensitivity

58.37%

95% Confidence Interval: 54.1% - 62.63%

Specificity

70.86%

95% Confidence Interval: 69.85% - 71.88%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 8,264

▼ View Full Metrics

Area Under Curve (AUC) 0.6922

Balanced Accuracy 64.62%

Overall Accuracy 70.09%

Matthews Correlation Coefficient (MCC) 0.1527

Positive Predictive Value (PPV) 0.18%

Negative Predictive Value (NPV) 99.95%

Prevalence 0.09%

Confusion Matrix

True Positives (TP) 300

False Negatives (FN) 214

True Negatives (TN) 5,492

False Positives (FP) 2,258

Ground Truth Cases

Positive Cases 514

Negative Cases 7,750

Release Date November 2024

Colon Diffuse Distension

Canine

Clinical Urgency: Moderate

Sensitivity

85.92%

95% Confidence Interval: 77.82% - 94.01%

Specificity

79.24%

95% Confidence Interval: 78.62% - 79.85%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 16,705

▼ View Full Metrics

Area Under Curve (AUC) 0.8879

Balanced Accuracy 82.58%

Overall Accuracy 79.26%

Matthews Correlation Coefficient (MCC) 0.1040

Positive Predictive Value (PPV) 7.93%

Negative Predictive Value (NPV) 99.63%

Prevalence 2.04%

Confusion Matrix

True Positives (TP) 61

False Negatives (FN) 10

True Negatives (TN) 13,180

False Positives (FP) 3,454

Ground Truth Cases

Positive Cases 71

Negative Cases 16,634

Release Date February 2026

Segmental SI Distension/Mechanical Ileus

Canine

Clinical Urgency: High

Sensitivity

69.56%

95% Confidence Interval: 67.68% - 71.44%

Specificity

76.91%

95% Confidence Interval: 76.47% - 77.35%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 37,275

▼ View Full Metrics

Area Under Curve (AUC) 0.7808

Balanced Accuracy 73.24%

Overall Accuracy 76.46%

Matthews Correlation Coefficient (MCC) 0.2547

Positive Predictive Value (PPV) 2.75%

Negative Predictive Value (NPV) 99.63%

Prevalence 0.93%

Confusion Matrix

True Positives (TP) 1,595

False Negatives (FN) 698

True Negatives (TN) 26,905

False Positives (FP) 8,077

Ground Truth Cases

Positive Cases 2,293

Negative Cases 34,982

Release Date February 2026

Kidney Size L

Canine

Clinical Urgency: Low

Sensitivity

66.18%

95% Confidence Interval: 61.17% - 71.19%

Specificity

66.61%

95% Confidence Interval: 66.05% - 67.18%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 27,277

▼ View Full Metrics

Area Under Curve (AUC) 0.7260

Balanced Accuracy 66.4%

Overall Accuracy 66.61%

Matthews Correlation Coefficient (MCC) 0.0773

Positive Predictive Value (PPV) 0.79%

Negative Predictive Value (NPV) 99.8%

Prevalence 0.4%

Confusion Matrix

True Positives (TP) 227

False Negatives (FN) 116

True Negatives (TN) 17,942

False Positives (FP) 8,992

Ground Truth Cases

Positive Cases 343

Negative Cases 26,934

Release Date February 2026

Kidney Size R

Canine

Clinical Urgency: Low

Sensitivity

79.65%

95% Confidence Interval: 75.72% - 83.58%

Specificity

70.55%

95% Confidence Interval: 69.65% - 71.46%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 10,133

▼ View Full Metrics

Area Under Curve (AUC) 0.7932

Balanced Accuracy 75.1%

Overall Accuracy 70.92%

Matthews Correlation Coefficient (MCC) 0.2113

Positive Predictive Value (PPV) 0.99%

Negative Predictive Value (NPV) 99.89%

Prevalence 0.37%

Confusion Matrix

True Positives (TP) 321

False Negatives (FN) 82

True Negatives (TN) 6,865

False Positives (FP) 2,865

Ground Truth Cases

Positive Cases 403

Negative Cases 9,730

Release Date February 2026

L Renal Margin Change

Canine

Clinical Urgency: Moderate

Sensitivity

57.63%

95% Confidence Interval: 52.48% - 62.77%

Specificity

76.46%

95% Confidence Interval: 75.99% - 76.93%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 31,572

▼ View Full Metrics

Area Under Curve (AUC) 0.7126

Balanced Accuracy 67.04%

Overall Accuracy 76.25%

Matthews Correlation Coefficient (MCC) 0.0841

Positive Predictive Value (PPV) 0.12%

Negative Predictive Value (NPV) 99.97%

Prevalence 0.05%

Confusion Matrix

True Positives (TP) 204

False Negatives (FN) 150

True Negatives (TN) 23,869

False Positives (FP) 7,349

Ground Truth Cases

Positive Cases 354

Negative Cases 31,218

Release Date February 2026

R Renal Margin Change

Canine

Clinical Urgency: Moderate

Sensitivity

61.8%

95% Confidence Interval: 54.66% - 68.94%

Specificity

75.43%

95% Confidence Interval: 74.98% - 75.87%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 36,386

▼ View Full Metrics

Area Under Curve (AUC) 0.7389

Balanced Accuracy 68.61%

Overall Accuracy 75.36%

Matthews Correlation Coefficient (MCC) 0.0602

Positive Predictive Value (PPV) 0.13%

Negative Predictive Value (NPV) 99.97%

Prevalence 0.05%

Confusion Matrix

True Positives (TP) 110

False Negatives (FN) 68

True Negatives (TN) 27,311

False Positives (FP) 8,897

Ground Truth Cases

Positive Cases 178

Negative Cases 36,208

Release Date February 2026

Nephroliths/Renal Mineral L

Canine

Clinical Urgency: Moderate

Sensitivity

83.15%

95% Confidence Interval: 79.97% - 86.32%

Specificity

71.78%

95% Confidence Interval: 70.97% - 72.58%

Radiologist Agreement Rate: 85%

Total Cases Evaluated: 12,634

▼ View Full Metrics

Area Under Curve (AUC) 0.8470

Balanced Accuracy 77.46%

Overall Accuracy 72.26%

Matthews Correlation Coefficient (MCC) 0.2399

Positive Predictive Value (PPV) 2.55%

Negative Predictive Value (NPV) 99.79%

Prevalence 0.88%

Confusion Matrix

True Positives (TP) 444

False Negatives (FN) 90

True Negatives (TN) 8,685

False Positives (FP) 3,415

Ground Truth Cases

Positive Cases 534

Negative Cases 12,100

Release Date November 2024

Nephroliths/Renal Mineral R

Canine

Clinical Urgency: Moderate

Sensitivity

76.99%

95% Confidence Interval: 73.16% - 80.81%

Specificity

74.55%

95% Confidence Interval: 73.82% - 75.28%

Radiologist Agreement Rate: 84%

Total Cases Evaluated: 14,240

▼ View Full Metrics

Area Under Curve (AUC) 0.7902

Balanced Accuracy 75.77%

Overall Accuracy 74.63%

Matthews Correlation Coefficient (MCC) 0.2060

Positive Predictive Value (PPV) 2.47%

Negative Predictive Value (NPV) 99.74%

Prevalence 0.83%

Confusion Matrix

True Positives (TP) 358

False Negatives (FN) 107

True Negatives (TN) 10,269

False Positives (FP) 3,506

Ground Truth Cases

Positive Cases 465

Negative Cases 13,775

Release Date November 2024

Urocystoliths/Urethroliths

Canine

Clinical Urgency: High

Sensitivity

73.1%

95% Confidence Interval: 68.57% - 77.63%

Specificity

85.99%

95% Confidence Interval: 85.59% - 86.38%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 29,775

▼ View Full Metrics

Area Under Curve (AUC) 0.8402

Balanced Accuracy 79.54%

Overall Accuracy 85.83%

Matthews Correlation Coefficient (MCC) 0.1841

Positive Predictive Value (PPV) 5.25%

Negative Predictive Value (NPV) 99.67%

Prevalence 1.05%

Confusion Matrix

True Positives (TP) 269

False Negatives (FN) 99

True Negatives (TN) 25,286

False Positives (FP) 4,121

Ground Truth Cases

Positive Cases 368

Negative Cases 29,407

Release Date February 2026

Decreased Serosal Detail

Canine

Clinical Urgency: High

Sensitivity

71.7%

95% Confidence Interval: 68.65% - 74.76%

Specificity

53.61%

95% Confidence Interval: 52.43% - 54.79%

Radiologist Agreement Rate: 80%

Total Cases Evaluated: 7,710

▼ View Full Metrics

Area Under Curve (AUC) 0.6779

Balanced Accuracy 62.65%

Overall Accuracy 55.56%

Matthews Correlation Coefficient (MCC) 0.1572

Positive Predictive Value (PPV) 3.83%

Negative Predictive Value (NPV) 98.66%

Prevalence 2.51%

Confusion Matrix

True Positives (TP) 598

False Negatives (FN) 236

True Negatives (TN) 3,686

False Positives (FP) 3,190

Ground Truth Cases

Positive Cases 834

Negative Cases 6,876

Release Date November 2024

Peritoneal Gas

Canine

Clinical Urgency: Critical

Sensitivity

74.31%

95% Confidence Interval: 66.11% - 82.51%

Specificity

82.47%

95% Confidence Interval: 82.06% - 82.89%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 32,624

▼ View Full Metrics

Area Under Curve (AUC) 0.8338

Balanced Accuracy 78.39%

Overall Accuracy 82.45%

Matthews Correlation Coefficient (MCC) 0.0858

Positive Predictive Value (PPV) 1.92%

Negative Predictive Value (NPV) 99.86%

Prevalence 0.46%

Confusion Matrix

True Positives (TP) 81

False Negatives (FN) 28

True Negatives (TN) 26,816

False Positives (FP) 5,699

Ground Truth Cases

Positive Cases 109

Negative Cases 32,515

Release Date February 2026

Mid Abdominal Mass

Canine

Clinical Urgency: Moderate

Sensitivity

68.06%

95% Confidence Interval: 61.84% - 74.27%

Specificity

73.22%

95% Confidence Interval: 72.78% - 73.66%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 39,499

▼ View Full Metrics

Area Under Curve (AUC) 0.7525

Balanced Accuracy 70.64%

Overall Accuracy 73.19%

Matthews Correlation Coefficient (MCC) 0.0686

Positive Predictive Value (PPV) 2.5%

Negative Predictive Value (NPV) 99.56%

Prevalence 1%

Confusion Matrix

True Positives (TP) 147

False Negatives (FN) 69

True Negatives (TN) 28,764

False Positives (FP) 10,519

Ground Truth Cases

Positive Cases 216

Negative Cases 39,283

Release Date February 2026

Pregnancy

Canine

Clinical Urgency: Moderate

Sensitivity

61.11%

95% Confidence Interval: 38.59% - 83.63%

Specificity

97.23%

95% Confidence Interval: 96.9% - 97.55%

Radiologist Agreement Rate: 98%

Total Cases Evaluated: 10,109

▼ View Full Metrics

Area Under Curve (AUC) 0.8969

Balanced Accuracy 79.17%

Overall Accuracy 97.16%

Matthews Correlation Coefficient (MCC) 0.1471

Positive Predictive Value (PPV) 3.62%

Negative Predictive Value (NPV) 99.93%

Prevalence 0.17%

Confusion Matrix

True Positives (TP) 11

False Negatives (FN) 7

True Negatives (TN) 9,811

False Positives (FP) 280

Ground Truth Cases

Positive Cases 18

Negative Cases 10,091

Release Date November 2024

Feline Thorax (13 conditions)

Heart Failure Left

Feline

Clinical Urgency: Critical

Sensitivity

82.3%

95% Confidence Interval: 75.26% - 89.34%

Specificity

97.63%

95% Confidence Interval: 97.42% - 97.84%

Radiologist Agreement Rate: 83%

Total Cases Evaluated: 19,855

▼ View Full Metrics

Area Under Curve (AUC) 0.9481

Balanced Accuracy 89.97%

Overall Accuracy 97.54%

Matthews Correlation Coefficient (MCC) 0.3629

Positive Predictive Value (PPV) 20.56%

Negative Predictive Value (NPV) 99.87%

Prevalence 0.74%

Confusion Matrix

True Positives (TP) 93

False Negatives (FN) 20

True Negatives (TN) 19,274

False Positives (FP) 468

Ground Truth Cases

Positive Cases 113

Negative Cases 19,742

Release Date November 2025

Left-Sided Cardiomegaly

Feline

Clinical Urgency: Moderate

Sensitivity

68.82%

95% Confidence Interval: 61.86% - 75.79%

Specificity

77.55%

95% Confidence Interval: 76.77% - 78.32%

Radiologist Agreement Rate: 98%

Total Cases Evaluated: 11,415

▼ View Full Metrics

Area Under Curve (AUC) 0.7957

Balanced Accuracy 73.18%

Overall Accuracy 77.42%

Matthews Correlation Coefficient (MCC) 0.1332

Positive Predictive Value (PPV) 14.12%

Negative Predictive Value (NPV) 97.89%

Prevalence 5.09%

Confusion Matrix

True Positives (TP) 117

False Negatives (FN) 53

True Negatives (TN) 8,720

False Positives (FP) 2,525

Ground Truth Cases

Positive Cases 170

Negative Cases 11,245

Release Date November 2025

Pulmonary Vasculature Enlargement

Feline

Clinical Urgency: Moderate

Sensitivity

65.73%

95% Confidence Interval: 61.02% - 70.43%

Specificity

80%

95% Confidence Interval: 79.3% - 80.71%

Radiologist Agreement Rate: 64%

Total Cases Evaluated: 12,729

▼ View Full Metrics

Area Under Curve (AUC) 0.7715

Balanced Accuracy 72.87%

Overall Accuracy 79.57%

Matthews Correlation Coefficient (MCC) 0.1924

Positive Predictive Value (PPV) 20.68%

Negative Predictive Value (NPV) 96.71%

Prevalence 7.35%

Confusion Matrix

True Positives (TP) 257

False Negatives (FN) 134

True Negatives (TN) 9,871

False Positives (FP) 2,467

Ground Truth Cases

Positive Cases 391

Negative Cases 12,338

Release Date November 2025

Caudodorsal Alveolar Pattern

Feline

Clinical Urgency: High

Sensitivity

76.84%

95% Confidence Interval: 70.62% - 83.05%

Specificity

81.56%

95% Confidence Interval: 80.92% - 82.2%

Radiologist Agreement Rate: 83%

Total Cases Evaluated: 14,450

▼ View Full Metrics

Area Under Curve (AUC) 0.8748

Balanced Accuracy 79.2%

Overall Accuracy 81.5%

Matthews Correlation Coefficient (MCC) 0.1632

Positive Predictive Value (PPV) 15.74%

Negative Predictive Value (NPV) 98.74%

Prevalence 4.29%

Confusion Matrix

True Positives (TP) 136

False Negatives (FN) 41

True Negatives (TN) 11,641

False Positives (FP) 2,632

Ground Truth Cases

Positive Cases 177

Negative Cases 14,273

Release Date November 2025

Cranioventral Alveolar Pattern

Feline

Clinical Urgency: High

Sensitivity

72.42%

95% Confidence Interval: 68.82% - 76.02%

Specificity

74.94%

95% Confidence Interval: 74.19% - 75.69%

Radiologist Agreement Rate: 83%

Total Cases Evaluated: 13,398

▼ View Full Metrics

Area Under Curve (AUC) 0.7901

Balanced Accuracy 73.68%

Overall Accuracy 74.83%

Matthews Correlation Coefficient (MCC) 0.2187

Positive Predictive Value (PPV) 11.47%

Negative Predictive Value (NPV) 98.38%

Prevalence 4.29%

Confusion Matrix

True Positives (TP) 428

False Negatives (FN) 163

True Negatives (TN) 9,598

False Positives (FP) 3,209

Ground Truth Cases

Positive Cases 591

Negative Cases 12,807

Release Date November 2025

Interstitial Pattern

Feline

Clinical Urgency: Moderate

Sensitivity

65.33%

95% Confidence Interval: 60.13% - 70.52%

Specificity

78.25%

95% Confidence Interval: 77.58% - 78.92%

Radiologist Agreement Rate: 94.29%

Total Cases Evaluated: 15,054

▼ View Full Metrics

Area Under Curve (AUC) 0.7828

Balanced Accuracy 71.79%

Overall Accuracy 77.97%

Matthews Correlation Coefficient (MCC) 0.1508

Positive Predictive Value (PPV) 19.9%

Negative Predictive Value (NPV) 96.46%

Prevalence 7.64%

Confusion Matrix

True Positives (TP) 211

False Negatives (FN) 112

True Negatives (TN) 11,527

False Positives (FP) 3,204

Ground Truth Cases

Positive Cases 323

Negative Cases 14,731

Release Date November 2025

Bronchial Pattern

Feline

Clinical Urgency: Moderate

Sensitivity

76.37%

95% Confidence Interval: 72.3% - 80.44%

Specificity

79.57%

95% Confidence Interval: 78.81% - 80.33%

Radiologist Agreement Rate: 62%

Total Cases Evaluated: 11,228

▼ View Full Metrics

Area Under Curve (AUC) 0.8469

Balanced Accuracy 77.97%

Overall Accuracy 79.45%

Matthews Correlation Coefficient (MCC) 0.2539

Positive Predictive Value (PPV) 36.1%

Negative Predictive Value (NPV) 95.7%

Prevalence 13.13%

Confusion Matrix

True Positives (TP) 320

False Negatives (FN) 99

True Negatives (TN) 8,601

False Positives (FP) 2,208

Ground Truth Cases

Positive Cases 419

Negative Cases 10,809

Release Date November 2025

Pulmonary Masses

Feline

Clinical Urgency: Moderate

Sensitivity

78.95%

95% Confidence Interval: 71.46% - 86.43%

Specificity

86.46%

95% Confidence Interval: 85.75% - 87.17%

Radiologist Agreement Rate: 85.96%

Total Cases Evaluated: 9,154

▼ View Full Metrics

Area Under Curve (AUC) 0.8746

Balanced Accuracy 82.7%

Overall Accuracy 86.37%

Matthews Correlation Coefficient (MCC) 0.2069

Positive Predictive Value (PPV) 12.45%

Negative Predictive Value (NPV) 99.41%

Prevalence 2.38%

Confusion Matrix

True Positives (TP) 90

False Negatives (FN) 24

True Negatives (TN) 7,816

False Positives (FP) 1,224

Ground Truth Cases

Positive Cases 114

Negative Cases 9,040

Release Date December 2025

Perihilar Infiltrate

Feline

Clinical Urgency: Moderate

Sensitivity

73.04%

95% Confidence Interval: 64.93% - 81.15%

Specificity

79.48%

95% Confidence Interval: 79.07% - 79.88%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 38,009

▼ View Full Metrics

Area Under Curve (AUC) 0.8651

Balanced Accuracy 76.26%

Overall Accuracy 79.46%

Matthews Correlation Coefficient (MCC) 0.0712

Positive Predictive Value (PPV) 0.74%

Negative Predictive Value (NPV) 99.93%

Prevalence 0.21%

Confusion Matrix

True Positives (TP) 84

False Negatives (FN) 31

True Negatives (TN) 30,117

False Positives (FP) 7,777

Ground Truth Cases

Positive Cases 115

Negative Cases 37,894

Release Date December 2025

Hypoinflation

Feline

Clinical Urgency: Moderate

Sensitivity

63.31%

95% Confidence Interval: 56.05% - 70.58%

Specificity

79.95%

95% Confidence Interval: 79.25% - 80.64%

Radiologist Agreement Rate: 89%

Total Cases Evaluated: 12,886

▼ View Full Metrics

Area Under Curve (AUC) 0.7986

Balanced Accuracy 71.63%

Overall Accuracy 79.73%

Matthews Correlation Coefficient (MCC) 0.1217

Positive Predictive Value (PPV) 3.99%

Negative Predictive Value (NPV) 99.4%

Prevalence 1.3%

Confusion Matrix

True Positives (TP) 107

False Negatives (FN) 62

True Negatives (TN) 10,167

False Positives (FP) 2,550

Ground Truth Cases

Positive Cases 169

Negative Cases 12,717

Release Date November 2025

Pleural Fluid

Feline

Clinical Urgency: High

Sensitivity

82.44%

95% Confidence Interval: 79.71% - 85.17%

Specificity

92.46%

95% Confidence Interval: 91.99% - 92.92%

Radiologist Agreement Rate: 75%

Total Cases Evaluated: 13,101

▼ View Full Metrics

Area Under Curve (AUC) 0.9351

Balanced Accuracy 87.45%

Overall Accuracy 91.89%

Matthews Correlation Coefficient (MCC) 0.5378

Positive Predictive Value (PPV) 21.05%

Negative Predictive Value (NPV) 99.54%

Prevalence 2.38%

Confusion Matrix

True Positives (TP) 615

False Negatives (FN) 131

True Negatives (TN) 11,423

False Positives (FP) 932

Ground Truth Cases

Positive Cases 746

Negative Cases 12,355

Release Date November 2025

Cranial Mediastinal Mass

Feline

Clinical Urgency: Moderate

Sensitivity

70.3%

95% Confidence Interval: 65.37% - 75.23%

Specificity

83.01%

95% Confidence Interval: 82.63% - 83.39%

Radiologist Agreement Rate: 63%

Total Cases Evaluated: 37,942

▼ View Full Metrics

Area Under Curve (AUC) 0.8312

Balanced Accuracy 76.66%

Overall Accuracy 82.9%

Matthews Correlation Coefficient (MCC) 0.1304

Positive Predictive Value (PPV) 3.9%

Negative Predictive Value (NPV) 99.65%

Prevalence 0.97%

Confusion Matrix

True Positives (TP) 232

False Negatives (FN) 98

True Negatives (TN) 31,221

False Positives (FP) 6,391

Ground Truth Cases

Positive Cases 330

Negative Cases 37,612

Release Date December 2025

Thoracic Lymphadenopathy

Feline

Clinical Urgency: High

Sensitivity

67.97%

95% Confidence Interval: 60.58% - 75.37%

Specificity

79.19%

95% Confidence Interval: 78.67% - 79.72%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 23,058

▼ View Full Metrics

Area Under Curve (AUC) 0.7528

Balanced Accuracy 73.58%

Overall Accuracy 79.12%

Matthews Correlation Coefficient (MCC) 0.0938

Positive Predictive Value (PPV) 0.94%

Negative Predictive Value (NPV) 99.88%

Prevalence 0.29%

Confusion Matrix

True Positives (TP) 104

False Negatives (FN) 49

True Negatives (TN) 18,139

False Positives (FP) 4,766

Ground Truth Cases

Positive Cases 153

Negative Cases 22,905

Release Date February 2026

Feline Abdomen (19 conditions)

Hepatomegaly

Feline

Clinical Urgency: Moderate

Sensitivity

82.04%

95% Confidence Interval: 78.09% - 86.0%

Specificity

80.86%

95% Confidence Interval: 80.42% - 81.3%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 31,134

▼ View Full Metrics

Area Under Curve (AUC) 0.8867

Balanced Accuracy 81.45%

Overall Accuracy 80.87%

Matthews Correlation Coefficient (MCC) 0.1690

Positive Predictive Value (PPV) 6.36%

Negative Predictive Value (NPV) 99.65%

Prevalence 1.56%

Confusion Matrix

True Positives (TP) 297

False Negatives (FN) 65

True Negatives (TN) 24,881

False Positives (FP) 5,891

Ground Truth Cases

Positive Cases 362

Negative Cases 30,772

Release Date February 2026

Hepatic Mass

Feline

Clinical Urgency: Moderate

Sensitivity

66.67%

95% Confidence Interval: 56.93% - 76.41%

Specificity

79%

95% Confidence Interval: 78.54% - 79.45%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 30,880

▼ View Full Metrics

Area Under Curve (AUC) 0.8240

Balanced Accuracy 72.83%

Overall Accuracy 78.96%

Matthews Correlation Coefficient (MCC) 0.0603

Positive Predictive Value (PPV) 1.07%

Negative Predictive Value (NPV) 99.86%

Prevalence 0.34%

Confusion Matrix

True Positives (TP) 60

False Negatives (FN) 30

True Negatives (TN) 24,324

False Positives (FP) 6,466

Ground Truth Cases

Positive Cases 90

Negative Cases 30,790

Release Date February 2026

Splenomegaly

Feline

Clinical Urgency: Moderate

Sensitivity

63.93%

95% Confidence Interval: 58.55% - 69.32%

Specificity

85.94%

95% Confidence Interval: 85.5% - 86.38%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 24,195

▼ View Full Metrics

Area Under Curve (AUC) 0.8238

Balanced Accuracy 74.94%

Overall Accuracy 85.66%

Matthews Correlation Coefficient (MCC) 0.1572

Positive Predictive Value (PPV) 5.73%

Negative Predictive Value (NPV) 99.44%

Prevalence 1.32%

Confusion Matrix

True Positives (TP) 195

False Negatives (FN) 110

True Negatives (TN) 20,531

False Positives (FP) 3,359

Ground Truth Cases

Positive Cases 305

Negative Cases 23,890

Release Date February 2026

Colon Enlargement

Feline

Clinical Urgency: Moderate

Sensitivity

67.82%

95% Confidence Interval: 60.87% - 74.76%

Specificity

94.39%

95% Confidence Interval: 94.18% - 94.6%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 47,369

▼ View Full Metrics

Area Under Curve (AUC) 0.8719

Balanced Accuracy 81.1%

Overall Accuracy 94.29%

Matthews Correlation Coefficient (MCC) 0.1605

Positive Predictive Value (PPV) 3.85%

Negative Predictive Value (NPV) 99.89%

Prevalence 0.33%

Confusion Matrix

True Positives (TP) 118

False Negatives (FN) 56

True Negatives (TN) 44,548

False Positives (FP) 2,647

Ground Truth Cases

Positive Cases 174

Negative Cases 47,195

Release Date December 2025

Colon Diffuse Distension

Feline

Clinical Urgency: Moderate

Sensitivity

63.77%

95% Confidence Interval: 59.14% - 68.4%

Specificity

77.07%

95% Confidence Interval: 76.45% - 77.68%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 18,376

▼ View Full Metrics

Area Under Curve (AUC) 0.7765

Balanced Accuracy 70.42%

Overall Accuracy 76.77%

Matthews Correlation Coefficient (MCC) 0.1422

Positive Predictive Value (PPV) 5.37%

Negative Predictive Value (NPV) 99.05%

Prevalence 2%

Confusion Matrix

True Positives (TP) 264

False Negatives (FN) 150

True Negatives (TN) 13,843

False Positives (FP) 4,119

Ground Truth Cases

Positive Cases 414

Negative Cases 17,962

Release Date February 2026

Small Intestine Enlargement

Feline

Clinical Urgency: Moderate

Sensitivity

61.09%

95% Confidence Interval: 58.44% - 63.74%

Specificity

78.76%

95% Confidence Interval: 78.35% - 79.16%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 40,486

▼ View Full Metrics

Area Under Curve (AUC) 0.7566

Balanced Accuracy 69.92%

Overall Accuracy 78.19%

Matthews Correlation Coefficient (MCC) 0.1683

Positive Predictive Value (PPV) 16.66%

Negative Predictive Value (NPV) 96.68%

Prevalence 6.5%

Confusion Matrix

True Positives (TP) 796

False Negatives (FN) 507

True Negatives (TN) 30,859

False Positives (FP) 8,324

Ground Truth Cases

Positive Cases 1,303

Negative Cases 39,183

Release Date February 2026

Segmental SI Distension/Mechanical Ileus

Feline

Clinical Urgency: High

Sensitivity

69.69%

95% Confidence Interval: 66.8% - 72.58%

Specificity

73.75%

95% Confidence Interval: 73.21% - 74.29%

Radiologist Agreement Rate: 93%

Total Cases Evaluated: 26,834

▼ View Full Metrics

Area Under Curve (AUC) 0.7751

Balanced Accuracy 71.72%

Overall Accuracy 73.6%

Matthews Correlation Coefficient (MCC) 0.1810

Positive Predictive Value (PPV) 10.1%

Negative Predictive Value (NPV) 98.29%

Prevalence 4.06%

Confusion Matrix

True Positives (TP) 676

False Negatives (FN) 294

True Negatives (TN) 19,075

False Positives (FP) 6,789

Ground Truth Cases

Positive Cases 970

Negative Cases 25,864

Release Date February 2026

Mineral/Metal Gastric Material

Feline

Clinical Urgency: Moderate

Sensitivity

66.88%

95% Confidence Interval: 64.28% - 69.47%

Specificity

81.94%

95% Confidence Interval: 81.58% - 82.31%

Radiologist Agreement Rate: 76%

Total Cases Evaluated: 44,561

▼ View Full Metrics

Area Under Curve (AUC) 0.7704

Balanced Accuracy 74.41%

Overall Accuracy 81.52%

Matthews Correlation Coefficient (MCC) 0.2049

Positive Predictive Value (PPV) 0.07%

Negative Predictive Value (NPV) 99.99%

Prevalence 0.02%

Confusion Matrix

True Positives (TP) 846

False Negatives (FN) 419

True Negatives (TN) 35,478

False Positives (FP) 7,818

Ground Truth Cases

Positive Cases 1,265

Negative Cases 43,296

Release Date December 2025

Radiopaque Gastric Foreign Body

Feline

Clinical Urgency: High

Sensitivity

72.63%

95% Confidence Interval: 63.67% - 81.6%

Specificity

62.6%

95% Confidence Interval: 61.79% - 63.41%

Radiologist Agreement Rate: 77%

Total Cases Evaluated: 13,886

▼ View Full Metrics

Area Under Curve (AUC) 0.7127

Balanced Accuracy 67.62%

Overall Accuracy 62.67%

Matthews Correlation Coefficient (MCC) 0.0599

Positive Predictive Value (PPV) 72.84%

Negative Predictive Value (NPV) 62.35%

Prevalence 58%

Confusion Matrix

True Positives (TP) 69

False Negatives (FN) 26

True Negatives (TN) 8,633

False Positives (FP) 5,158

Ground Truth Cases

Positive Cases 95

Negative Cases 13,791

Release Date August 2024

Kidney Size L

Feline

Clinical Urgency: Low

Sensitivity

68.44%

95% Confidence Interval: 66.27% - 70.61%

Specificity

63.53%

95% Confidence Interval: 62.62% - 64.44%

Radiologist Agreement Rate: 53%

Total Cases Evaluated: 12,519

▼ View Full Metrics

Area Under Curve (AUC) 0.6957

Balanced Accuracy 65.99%

Overall Accuracy 64.22%

Matthews Correlation Coefficient (MCC) 0.2261

Positive Predictive Value (PPV) 13.88%

Negative Predictive Value (NPV) 95.91%

Prevalence 7.91%

Confusion Matrix

True Positives (TP) 1,206

False Negatives (FN) 556

True Negatives (TN) 6,834

False Positives (FP) 3,923

Ground Truth Cases

Positive Cases 1,762

Negative Cases 10,757

Release Date August 2024

Kidney Size R

Feline

Clinical Urgency: Low

Sensitivity

72.77%

95% Confidence Interval: 70.49% - 75.05%

Specificity

65.02%

95% Confidence Interval: 64.09% - 65.95%

Radiologist Agreement Rate: 53%

Total Cases Evaluated: 11,571

▼ View Full Metrics

Area Under Curve (AUC) 0.7380

Balanced Accuracy 68.89%

Overall Accuracy 66%

Matthews Correlation Coefficient (MCC) 0.2570

Positive Predictive Value (PPV) 15.16%

Negative Predictive Value (NPV) 96.53%

Prevalence 7.91%

Confusion Matrix

True Positives (TP) 1,069

False Negatives (FN) 400

True Negatives (TN) 6,568

False Positives (FP) 3,534

Ground Truth Cases

Positive Cases 1,469

Negative Cases 10,102

Release Date August 2024

L Renal Margin Change

Feline

Clinical Urgency: Moderate

Sensitivity

67.26%

95% Confidence Interval: 61.11% - 73.42%

Specificity

74.73%

95% Confidence Interval: 73.88% - 75.58%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 10,227

▼ View Full Metrics

Area Under Curve (AUC) 0.7315

Balanced Accuracy 71%

Overall Accuracy 74.57%

Matthews Correlation Coefficient (MCC) 0.1395

Positive Predictive Value (PPV) 4.07%

Negative Predictive Value (NPV) 99.31%

Prevalence 1.57%

Confusion Matrix

True Positives (TP) 150

False Negatives (FN) 73

True Negatives (TN) 7,476

False Positives (FP) 2,528

Ground Truth Cases

Positive Cases 223

Negative Cases 10,004

Release Date February 2026

R Renal Margin Change

Feline

Clinical Urgency: Moderate

Sensitivity

70.27%

95% Confidence Interval: 66.91% - 73.62%

Specificity

71.82%

95% Confidence Interval: 70.91% - 72.72%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 10,226

▼ View Full Metrics

Area Under Curve (AUC) 0.7687

Balanced Accuracy 71.04%

Overall Accuracy 71.71%

Matthews Correlation Coefficient (MCC) 0.2315

Positive Predictive Value (PPV) 3.4%

Negative Predictive Value (NPV) 99.42%

Prevalence 1.39%

Confusion Matrix

True Positives (TP) 501

False Negatives (FN) 212

True Negatives (TN) 6,832

False Positives (FP) 2,681

Ground Truth Cases

Positive Cases 713

Negative Cases 9,513

Release Date February 2026

Nephroliths/Renal Mineral L

Feline

Clinical Urgency: Moderate

Sensitivity

69.71%

95% Confidence Interval: 65.11% - 74.31%

Specificity

70.09%

95% Confidence Interval: 69.27% - 70.9%

Radiologist Agreement Rate: 88%

Total Cases Evaluated: 12,519

▼ View Full Metrics

Area Under Curve (AUC) 0.6990

Balanced Accuracy 69.9%

Overall Accuracy 70.08%

Matthews Correlation Coefficient (MCC) 0.1480

Positive Predictive Value (PPV) 10.74%

Negative Predictive Value (NPV) 97.82%

Prevalence 4.91%

Confusion Matrix

True Positives (TP) 267

False Negatives (FN) 116

True Negatives (TN) 8,506

False Positives (FP) 3,630

Ground Truth Cases

Positive Cases 383

Negative Cases 12,136

Release Date August 2024

Nephroliths/Renal Mineral R

Feline

Clinical Urgency: Moderate

Sensitivity

74.74%

95% Confidence Interval: 69.69% - 79.78%

Specificity

71.31%

95% Confidence Interval: 70.51% - 72.11%

Radiologist Agreement Rate: 85%

Total Cases Evaluated: 12,519

▼ View Full Metrics

Area Under Curve (AUC) 0.7302

Balanced Accuracy 73.02%

Overall Accuracy 71.39%

Matthews Correlation Coefficient (MCC) 0.1502

Positive Predictive Value (PPV) 11.86%

Negative Predictive Value (NPV) 98.2%

Prevalence 4.91%

Confusion Matrix

True Positives (TP) 213

False Negatives (FN) 72

True Negatives (TN) 8,724

False Positives (FP) 3,510

Ground Truth Cases

Positive Cases 285

Negative Cases 12,234

Release Date August 2024

Urocystoliths/Urethroliths

Feline

Clinical Urgency: High

Sensitivity

75.26%

95% Confidence Interval: 66.67% - 83.85%

Specificity

68.21%

95% Confidence Interval: 67.71% - 68.72%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 32,505

▼ View Full Metrics

Area Under Curve (AUC) 0.7837

Balanced Accuracy 71.74%

Overall Accuracy 68.24%

Matthews Correlation Coefficient (MCC) 0.0509

Positive Predictive Value (PPV) 1.6%

Negative Predictive Value (NPV) 99.75%

Prevalence 0.68%

Confusion Matrix

True Positives (TP) 73

False Negatives (FN) 24

True Negatives (TN) 22,107

False Positives (FP) 10,301

Ground Truth Cases

Positive Cases 97

Negative Cases 32,408

Release Date February 2026

Decreased Serosal Detail

Feline

Clinical Urgency: High

Sensitivity

70.75%

95% Confidence Interval: 68.78% - 72.72%

Specificity

73.06%

95% Confidence Interval: 72.26% - 73.86%

Radiologist Agreement Rate: 76%

Total Cases Evaluated: 13,886

▼ View Full Metrics

Area Under Curve (AUC) 0.7602

Balanced Accuracy 71.91%

Overall Accuracy 72.72%

Matthews Correlation Coefficient (MCC) 0.3294

Positive Predictive Value (PPV) 12.08%

Negative Predictive Value (NPV) 97.95%

Prevalence 4.97%

Confusion Matrix

True Positives (TP) 1,449

False Negatives (FN) 599

True Negatives (TN) 8,649

False Positives (FP) 3,189

Ground Truth Cases

Positive Cases 2,048

Negative Cases 11,838

Release Date August 2024

Mid Abdominal Mass

Feline

Clinical Urgency: Moderate

Sensitivity

81.68%

95% Confidence Interval: 75.05% - 88.3%

Specificity

73.43%

95% Confidence Interval: 72.68% - 74.18%

Radiologist Agreement Rate: 72%

Total Cases Evaluated: 13,493

▼ View Full Metrics

Area Under Curve (AUC) 0.8246

Balanced Accuracy 77.56%

Overall Accuracy 73.51%

Matthews Correlation Coefficient (MCC) 0.1216

Positive Predictive Value (PPV) 4.73%

Negative Predictive Value (NPV) 99.6%

Prevalence 1.59%

Confusion Matrix

True Positives (TP) 107

False Negatives (FN) 24

True Negatives (TN) 9,812

False Positives (FP) 3,550

Ground Truth Cases

Positive Cases 131

Negative Cases 13,362

Release Date December 2025

Mineralized Fetal Skeletons/Pregnancy

Feline

Clinical Urgency: Moderate

Performance metrics for this classifier are under review due to a labeling schema change and will be republished once validated.

Spine/Musculoskeletal (8 conditions)

Intervertebral Disc Disease Thoracic

Canine & Feline

Clinical Urgency: Moderate

Sensitivity

81.94%

95% Confidence Interval: 80.41% - 83.47%

Specificity

73.75%

95% Confidence Interval: 72.8% - 74.69%

Radiologist Agreement Rate: N/A

Total Cases Evaluated: 10,717

▼ View Full Metrics

Area Under Curve (AUC) 0.8512

Balanced Accuracy 77.84%

Overall Accuracy 75.6%

Matthews Correlation Coefficient (MCC) 0.4780

Positive Predictive Value (PPV) 3.59%

Negative Predictive Value (NPV) 99.71%

Prevalence 1.18%

Confusion Matrix

True Positives (TP) 1,987

False Negatives (FN) 438

True Negatives (TN) 6,115

False Positives (FP) 2,177

Ground Truth Cases

Positive Cases 2,425

Negative Cases 8,292

Release Date December 2025

Intervertebral Disc Disease Lumbar

Canine & Feline

Clinical Urgency: Moderate

Sensitivity

77.32%

95% Confidence Interval: 74.71% - 79.92%

Specificity

68.01%

95% Confidence Interval: 66.76% - 69.27%

Radiologist Agreement Rate: 87.5%

Total Cases Evaluated: 6,294

▼ View Full Metrics

Area Under Curve (AUC) 0.8360

Balanced Accuracy 72.67%

Overall Accuracy 69.48%

Matthews Correlation Coefficient (MCC) 0.3384

Positive Predictive Value (PPV) 7.93%

Negative Predictive Value (NPV) 98.83%

Prevalence 3.44%

Confusion Matrix

True Positives (TP) 767

False Negatives (FN) 225

True Negatives (TN) 3,606

False Positives (FP) 1,696

Ground Truth Cases

Positive Cases 992

Negative Cases 5,302

Release Date December 2025

Spondylosis Lumbar Spine

Canine & Feline

Clinical Urgency: Low

Sensitivity

80.55%

95% Confidence Interval: 78.38% - 82.72%

Specificity

71.62%

95% Confidence Interval: 70.37% - 72.87%

Radiologist Agreement Rate: 79%

Total Cases Evaluated: 6,294

▼ View Full Metrics

Area Under Curve (AUC) 0.8071

Balanced Accuracy 76.08%

Overall Accuracy 73.44%

Matthews Correlation Coefficient (MCC) 0.4305

Positive Predictive Value (PPV) 8.7%

Negative Predictive Value (NPV) 99.1%

Prevalence 3.25%

Confusion Matrix

True Positives (TP) 1,031

False Negatives (FN) 249

True Negatives (TN) 3,591

False Positives (FP) 1,423

Ground Truth Cases

Positive Cases 1,280

Negative Cases 5,014

Release Date December 2025

Congenital Thoracic Vertebrae Anomaly

Canine & Feline

Clinical Urgency: Moderate

Performance metrics for this classifier are under review and will be republished once validated.

Congenital Lumbar Vertebrae Anomaly

Canine & Feline

Clinical Urgency: Moderate

Sensitivity

87.91%

95% Confidence Interval: 83.55% - 92.27%

Specificity

69.99%

95% Confidence Interval: 69.07% - 70.9%

Radiologist Agreement Rate: 72%

Total Cases Evaluated: 9,767

▼ View Full Metrics

Area Under Curve (AUC) 0.8201

Balanced Accuracy 78.95%

Overall Accuracy 70.38%

Matthews Correlation Coefficient (MCC) 0.1832

Positive Predictive Value (PPV) 1.85%

Negative Predictive Value (NPV) 99.89%

Prevalence 0.64%

Confusion Matrix

True Positives (TP) 189

False Negatives (FN) 26

True Negatives (TN) 6,685

False Positives (FP) 2,867

Ground Truth Cases

Positive Cases 215

Negative Cases 9,552

Release Date December 2025

Hip Dysplasia

Canine & Feline

Clinical Urgency: Moderate

Sensitivity

78.68%

95% Confidence Interval: 73.68% - 83.68%

Specificity

73.42%

95% Confidence Interval: 72.32% - 74.52%

Radiologist Agreement Rate: 71.36%

Total Cases Evaluated: 6,435

▼ View Full Metrics

Area Under Curve (AUC) 0.8259

Balanced Accuracy 76.05%

Overall Accuracy 73.63%

Matthews Correlation Coefficient (MCC) 0.2260

Positive Predictive Value (PPV) 2.5%

Negative Predictive Value (NPV) 99.75%

Prevalence 0.86%

Confusion Matrix

True Positives (TP) 203

False Negatives (FN) 55

True Negatives (TN) 4,535

False Positives (FP) 1,642

Ground Truth Cases

Positive Cases 258

Negative Cases 6,177

Release Date July 2025

DJD Pelvis

Canine & Feline

Clinical Urgency: Low

Sensitivity

73.6%

95% Confidence Interval: 69.25% - 77.96%

Specificity

73.6%

95% Confidence Interval: 72.49% - 74.71%

Radiologist Agreement Rate: 87.61%

Total Cases Evaluated: 6,435

▼ View Full Metrics

Area Under Curve (AUC) 0.7689

Balanced Accuracy 73.6%

Overall Accuracy 73.6%

Matthews Correlation Coefficient (MCC) 0.2487

Positive Predictive Value (PPV) 6.18%

Negative Predictive Value (NPV) 99.16%

Prevalence 2.31%

Confusion Matrix

True Positives (TP) 290

False Negatives (FN) 104

True Negatives (TN) 4,446

False Positives (FP) 1,595

Ground Truth Cases

Positive Cases 394

Negative Cases 6,041

Release Date August 2025

Luxation Pelvis

Canine & Feline

Clinical Urgency: High

Sensitivity

75.77%

95% Confidence Interval: 71.32% - 80.23%

Specificity

77.88%

95% Confidence Interval: 76.83% - 78.92%

Radiologist Agreement Rate: 89%

Total Cases Evaluated: 6,435

▼ View Full Metrics

Area Under Curve (AUC) 0.8444

Balanced Accuracy 76.83%

Overall Accuracy 77.76%

Matthews Correlation Coefficient (MCC) 0.2826

Positive Predictive Value (PPV) 1.42%

Negative Predictive Value (NPV) 99.87%

Prevalence 0.42%

Confusion Matrix

True Positives (TP) 269

False Negatives (FN) 86

True Negatives (TN) 4,735

False Positives (FP) 1,345

Ground Truth Cases

Positive Cases 355

Negative Cases 6,080

Release Date August 2025

[1] Hosmer DW, Lemeshow S. Applied Logistic Regression. 2nd ed. New York: Wiley; 2000. AUC interpretation: 0.70–0.80 = acceptable, 0.80–0.90 = excellent, ≥0.90 = outstanding. See also: Çorbacιoğlu ŞK, Aksel G. “Receiver operating characteristic curve analysis in diagnostic accuracy studies: A guide to interpreting the area under the curve value.” Turk J Emerg Med. 2023;23(4):195–198. PMC10664195

Frequently Asked Questions

How accurate are Vetology's AI classifiers?

Our classifiers average 70% sensitivity and 73% specificity across all 83 conditions, validated on over 300,000 test cases from real-world veterinary practices. Performance varies by condition – scroll up to see specific metrics for each classifier by species.

What conditions can Vetology's AI detect?

We detect 83 conditions including cardiomegaly, pleural fluid, hepatomegaly and microhepatia, splenomegaly, kidney size, thoracic masses, abdominal masses, bladder and kidney stones, and more across canine and feline patients.

Our classifiers cover thorax (20 canine, 15 feline), abdomen (25 canine, 14 feline), and spine/musculoskeletal (9 conditions) imaging.

How does Vetology compare to other veterinary imaging AI platforms?

Unlike competitors, we publish complete performance metrics for every classifier. Our transparent data includes sensitivity, specificity, case counts, and Radiologist Agreement Rate, allowing you to make informed decisions about diagnostic accuracy rather than relying on marketing claims alone.

What is Radiologist Agreement Rate (RAR)?

Radiologist Agreement Rate measures how often our AI classifiers agree with board-certified veterinary radiologists on the same cases. This benchmark helps practices understand how AI performance compares to specialist interpretation and provides context for clinical decision-making.

How was this data validated?

All performance metrics are based on a rigorous testing process using over 300,000 real-world veterinary cases.

Each classifier was evaluated independently, with sensitivity, specificity, and case counts calculated from actual clinical imaging studies.

Radiologist Agreement Rates compare AI predictions against board-certified veterinary radiologist interpretations.

Ready to Experience AI-Assisted Radiology?

See how Vetology’s classifiers can improve diagnostic confidence and workflow efficiency in your practice.

Request a Demo

View Pricing

Clinical Validation across 89 Radiographic Condition Classifiers

Veterinary AI Classifier Performance Metrics

89+ Condition Classifiers Built on 300,000 Expert-Reviewed Veterinary Imaging Cases

What Makes Our Data Different

Full Transparency

Species-Specific Classifier Development

Real-World Testing

Understanding Our AI Virtual Radiologist Report Screening Metrics

Why Radiologist Benchmarking Matters

Making Sense of the Metrics

Radiologist Agreement Rate

Total Test Cases

Sensitivity

How reliably the AI detects a condition when it is present. (True Positive Rate)

Specificity

How accurately the AI rules out a condition when it is absent.(True Negative Rate)

Confusion Matrix (TP, FN, TN, FP)

95% Confidence Intervals

(for Sensitivity and Specificity)

Area Under Curve (AUC)

Positive Predictive Value (PPV)

Negative Predictive Value (NPV)

Overall Accuracy

Balanced Accuracy

Matthews Correlation Coefficient (MCC)

Prevalence in Test Set

Clinical Urgency Rating

Ground Truth Case Counts

(Positive + Negative)

Model Release Date

Vetology’s Published Condition Classifier Metrics

Sensitivity

Specificity

95% Confidence Interval

Radiologist Agreement Rate

Area Under Curve (AUC)

Positive Predictive Value (PPV)

Negative Predictive Value (NPV)

Prevalence

Overall Accuracy

Balanced Accuracy

Matthews Correlation Coefficient (MCC)

Confusion Matrix

Clinical Urgency

Ground Truth Cases

Frequently Asked Questions

How accurate are Vetology's AI classifiers?

What conditions can Vetology's AI detect?

How does Vetology compare to other veterinary imaging AI platforms?

What is Radiologist Agreement Rate (RAR)?

How was this data validated?

Ready to Experience AI-Assisted Radiology?

Pin It on Pinterest

How reliably the AI detects a condition when it is present.
(True Positive Rate)

How accurately the AI rules out a condition when it is absent.
(True Negative Rate)