AI in diagnostics and imaging: what 455 records show

Machine learning's edge over standard statistics in predicting trauma deaths is small and may be inflated · TRV-2026-1068

• Difference in AUC between best machine learning model and logistic regression for trauma mortality prediction

  AUC difference 0.026 (95% CI 0.009 to 0.043)

  17 studies · 243,324 people · search to January 15, 2025

  The review’s own words

  The pooled AUC difference favoring ML was 0.026 (95% CI 0.009-0.043)

• AUC difference between machine learning and logistic regression in studies reporting confidence intervals for both models

  AUC difference 0.017 (95% CI 0.005 to 0.029)

  search to January 15, 2025

  The review’s own words

  the co-primary estimate was 0.017 (95% CI 0.005-0.029)

AI-made MRI scans match real ones overall but lag on diseased areas, a 41-study review found · TRV-2026-0904

• Image similarity (SSIM) of synthesized postcontrast T1-weighted MRI in brain tumor studies

  SSIM 0.92 (95% CI 0.90 to 0.93)

  15 studies · search to January 16, 2025

  The review’s own words

  Meta-analysis of 15 brain tumor studies (30 models) yielded pooled SSIM of 0.92 (95% CI, 0.90-0.93) and pooled PSNR of 30.6 dB (95% CI, 28.6-32.6).

• Peak signal-to-noise ratio (PSNR) of synthesized postcontrast T1-weighted MRI in brain tumor studies

  PSNR 30.6 (95% CI 28.6 to 32.6)

  15 studies · search to January 16, 2025

  The review’s own words

  Meta-analysis of 15 brain tumor studies (30 models) yielded pooled SSIM of 0.92 (95% CI, 0.90-0.93) and pooled PSNR of 30.6 dB (95% CI, 28.6-32.6).

Only 1 of 20 studies of AI fertility predictions had low risk of bias, a review shows · TRV-2026-0674

• Pooled sensitivity of machine learning models for predicting pregnancy outcomes after assisted reproduction

  Sensitivity 0.737 (95% CI 0.662 to 0.799)

  14 studies · search to July 2026

  The review’s own words

  The pooled sensitivity was 0.737 (95% CI, 0.662-0.799) and the pooled specificity was 0.789 (95% CI, 0.709-0.851), with substantial heterogeneity (I2 = 97.7% and 99.0%, respectively).

• Pooled specificity of machine learning models for predicting pregnancy outcomes after assisted reproduction

  Specificity 0.789 (95% CI 0.709 to 0.851)

  14 studies · search to July 2026

  The review’s own words

  The pooled sensitivity was 0.737 (95% CI, 0.662-0.799) and the pooled specificity was 0.789 (95% CI, 0.709-0.851), with substantial heterogeneity (I2 = 97.7% and 99.0%, respectively).

• Diagnostic odds ratio of machine learning models for predicting pregnancy outcomes after assisted reproduction

  Diagnostic odds ratio 10.49 (95% CI 6.28 to 17.53)

  14 studies · search to July 2026

  The review’s own words

  The pooled DOR was 10.49 (95% CI, 6.28-17.53), and the SROC curve indicated acceptable overall discrimination.

For narrowed bile ducts, a review of five studies found AI hit 95% sensitivity · TRV-2026-0679

• Diagnostic accuracy (SROC) of AI-assisted cholangioscopy for biliary strictures

  Accuracy (SROC) 97% (95% CI 95% to 98%)

  The review’s own words

  The pooled sensitivity and specificity were 95% (95% CI: 85-98) and 88% (95% CI: 76-94), with a diagnostic accuracy (SROC) of 97% (95% CI: 95-98).

• Diagnostic accuracy (SROC) of neural network cholangioscopy AI, neural network studies only

  Accuracy (SROC) 97% (95% CI 95% to 98%)

  4 studies · 538 people

  The review’s own words

  Sensitivity analysis of CNN studies (4 studies, 538 patients) demonstrated a pooled sensitivity, specificity, and accuracy (SROC) of 95% (95% CI: 82-99), 88% (95% CI: 72-95), and 97% (95% CI: 95-98), respectively.

• Specificity of AI-assisted cholangioscopy for indeterminate or malignant biliary strictures

  Specificity 88% (95% CI 76% to 94%)

  The review’s own words

  The pooled sensitivity and specificity were 95% (95% CI: 85-98) and 88% (95% CI: 76-94), with a diagnostic accuracy (SROC) of 97% (95% CI: 95-98).

AI on heart ultrasounds spots a hard-to-diagnose heart failure, but evidence is thin · TRV-2026-2311

• Area under summary ROC curve for AI echocardiography detecting HFpEF or elevated filling pressure

  Area under the summary ROC curve 0.875 (95% CI 0.791 to 0.930)

  2,770 people · search to June 2026

  The review’s own words

  Pooled sensitivity was 84.0% (95% CI: 73.0-91.0%) and specificity 83.0% (95% CI: 73.0-89.0%), with an area under the summary ROC curve of 0.875 (95% CI: 0.791-0.930).

• Diagnostic odds ratio of AI echocardiography for detecting HFpEF or elevated filling pressure

  Diagnostic odds ratio 24.73 (95% CI 18.46 to 33.12)

  2,770 people · search to June 2026

  The review’s own words

  The diagnostic odds ratio was 24.73 (95% CI: 18.46-33.12)

• Sensitivity of AI-enabled echocardiography for detecting HFpEF, elevated filling pressure or diastolic dysfunction

  Sensitivity 84.0% (95% CI 73.0% to 91.0%)

  2,770 people · search to June 2026

  The review’s own words

  Pooled sensitivity was 84.0% (95% CI: 73.0-91.0%) and specificity 83.0% (95% CI: 73.0-89.0%)

AI hit 91% sensitivity on children's ear infections; a nine-study review shows it matched GPs · TRV-2026-2182

• Sensitivity of AI systems in detecting acute otitis media

  Sensitivity 91% (95% CI 77% to 97%)

  search to July 30, 2025

  The review’s own words

  The pooled sensitivity and specificity were 91% (95% confidence interval [CI]: 77%–97%) and 99% (95% CI: 94%–100%), respectively.

• Specificity of AI systems in detecting acute otitis media

  Specificity 99% (95% CI 94% to 100%)

  search to July 30, 2025

  The review’s own words

  The pooled sensitivity and specificity were 91% (95% confidence interval [CI]: 77%–97%) and 99% (95% CI: 94%–100%), respectively.

AI scan tools for glioblastoma show useful but imperfect accuracy in outside tests, with AUC 0.81 · TRV-2026-2075

• Sensitivity of MRI-based machine learning for distinguishing true progression from pseudoprogression in glioblastoma (external test cohorts)

  Sensitivity 0.78 (95% CI 0.63 to 0.88)

  219 people · search to October 1, 2025

  The review’s own words

  Pooled sensitivity was 0.78 (95% confidence interval [CI], 0.63-0.88), specificity was 0.75 (95% CI, 0.63-0.84), and summary area under the receiver operating characteristic curve (AUC) was 0.81 (bootstrap 95% CI, 0.69-0.91).

• Summary AUC of MRI-based machine learning for distinguishing true progression from pseudoprogression in glioblastoma (external test cohorts)

  AUC 0.81 (95% CI 0.69 to 0.91)

  219 people · search to October 1, 2025

  The review’s own words

  Pooled sensitivity was 0.78 (95% confidence interval [CI], 0.63-0.88), specificity was 0.75 (95% CI, 0.63-0.84), and summary area under the receiver operating characteristic curve (AUC) was 0.81 (bootstrap 95% CI, 0.69-0.91).

• Specificity of MRI-based machine learning for distinguishing true progression from pseudoprogression in glioblastoma (external test cohorts)

  Specificity 0.75 (95% CI 0.63 to 0.84)

  219 people · search to October 1, 2025

  The review’s own words

  Pooled sensitivity was 0.78 (95% confidence interval [CI], 0.63-0.88), specificity was 0.75 (95% CI, 0.63-0.84), and summary area under the receiver operating characteristic curve (AUC) was 0.81 (bootstrap 95% CI, 0.69-0.91).

In 12 studies, AI-read ECGs spot weak heart pumping in people without symptoms, a review finds · TRV-2026-1837

• AI-ECG discrimination of left ventricular ejection fraction 35% or lower in symptomatic people

  AUC 0.89 (95% CI 0.86 to 0.92)

  search to 8th September 2025

  The review’s own words

  Symptomatic cohorts showed AUROCs of 0.89 (95% CI 0.86-0.92), 0.90 (95% CI 0.88-0.92), and 0.85 (95% CI 0.83-0.86) respectively.

• AI-ECG discrimination of left ventricular ejection fraction 50% or lower in asymptomatic people

  AUC 0.89 (95% CI 0.85 to 0.97)

  search to 8th September 2025

  The review’s own words

  In asymptomatic cohorts evaluating left ventricular systolic dysfunction (LVSD), AI-ECG demonstrated promising but heterogeneous discriminatory performance (I²>85%) with pooled AUROCs of 0.87 (95% CI 0.75-0.99) for LVEF ≤35%, 0.93 (95% CI 0.88-0.97) for LVEF ≤40%, and 0.89 (95% CI 0.85-0.97) for LVEF ≤50%.

• AI-ECG discrimination of left ventricular ejection fraction 40% or lower in asymptomatic people

  AUC 0.93 (95% CI 0.88 to 0.97)

  search to 8th September 2025

  The review’s own words

  In asymptomatic cohorts evaluating left ventricular systolic dysfunction (LVSD), AI-ECG demonstrated promising but heterogeneous discriminatory performance (I²>85%) with pooled AUROCs of 0.87 (95% CI 0.75-0.99) for LVEF ≤35%, 0.93 (95% CI 0.88-0.97) for LVEF ≤40%, and 0.89 (95% CI 0.85-0.97) for LVEF ≤50%.

A review pooling nine studies shows AI detects eye toxoplasmosis in photos with 0.965 sensitivity · TRV-2026-1770

• Specificity of deep learning in detecting ocular toxoplasmosis from fundus photographs

  Specificity 0.955 (95% CI 0.929 to 0.972)

  search to January 2026

  The review’s own words

  Pooled sensitivity was 0.965 (95% CI: 0.944-0.978) and pooled specificity was 0.955 (95% CI: 0.929-0.972).

• Sensitivity of deep learning in detecting ocular toxoplasmosis from fundus photographs

  Sensitivity 0.965 (95% CI 0.944 to 0.978)

  search to January 2026

  The review’s own words

  Pooled sensitivity was 0.965 (95% CI: 0.944-0.978) and pooled specificity was 0.955 (95% CI: 0.929-0.972).

• Diagnostic odds ratio of deep learning for ocular toxoplasmosis on fundus photographs

  Diagnostic odds ratio 594.1 (95% CI 263.3 to 1340.4)

  search to January 2026

  The review’s own words

  The diagnostic odds ratio reached 594.1 (95% CI: 263.3-1340.4), with positive and negative likelihood ratios of 21.6 and 0.036.

Surgical AI built on a US registry scored 0.71 vs 0.80 for other data in a 295-model review · TRV-2026-1489

• Adjusted difference in AUROC of surgical prediction models, NSQIP vs non-NSQIP data sources

  Adjusted regression coefficient (β) for AUROC -0.056 (95% CI -0.105 to -0.007)

  The review’s own words

  NSQIP remained associated with lower AUROC (β=-0.056, 95% CI -0.105 to -0.007; p=0.026)

Generative AI was 13.0 points less accurate than physicians at radiology diagnosis, a review of 48 studies found · TRV-2026-1284

• Pooled diagnostic accuracy of generative AI on free-text radiology diagnostic tasks

  Accuracy 42.9% (95% CI 35.8% to 50.1%)

  search to March 2025

  The review’s own words

  Pooled diagnostic accuracy of generative AI was 42.9% (95% CI, 35.8-50.1%) for free-text tasks and 58.1% (95% CI, 50.0-66.2%) for choice tasks.

• Difference in accuracy of generative AI between image-only and text-only input

  Difference in accuracy (percentage points) -25.9 (95% CI -39.1 to -12.8)

  search to March 2025

  The review’s own words

  Text-only input was associated with higher accuracy than image-only input (difference in accuracy [image-only minus text-only], -25.9 percentage points [95% CI, -39.1 to -12.8]; P = .001)

• Pooled diagnostic accuracy of generative AI on choice-based radiology diagnostic tasks

  Accuracy 58.1% (95% CI 50.0% to 66.2%)

  search to March 2025

  The review’s own words

  Pooled diagnostic accuracy of generative AI was 42.9% (95% CI, 35.8-50.1%) for free-text tasks and 58.1% (95% CI, 50.0-66.2%) for choice tasks.

Review shows AI spots hip bone death at 88.8% sensitivity; only five tests used outside data · TRV-2026-1280

• Sensitivity of imaging AI models in detecting avascular necrosis of the femoral head

  Sensitivity 0.888 (95% CI 0.789 to 0.944)

  The review’s own words

  Pooled sensitivity was 0.888 (95% CI, 0.789-0.944), specificity was 0.937 (95% CI, 0.906-0.958), and SROC AUC was 0.967.

• Specificity of imaging AI models in detecting avascular necrosis of the femoral head

  Specificity 0.937 (95% CI 0.906 to 0.958)

  The review’s own words

  Pooled sensitivity was 0.888 (95% CI, 0.789-0.944), specificity was 0.937 (95% CI, 0.906-0.958), and SROC AUC was 0.967.

AI predicts death from heart valve infection well, but half the studies had bias risk · TRV-2026-1049

• Discrimination of machine learning models for in-hospital or early mortality in infective endocarditis

  AUC 0.85 (95% CI 0.81 to 0.89)

  search to April 2026

  The review’s own words

  Pooled AUC was 0.85 (95% confidence interval [CI], 0.81-0.89) for in-hospital/early mortality (I2 = 35.3%) and 0.85 (95% CI, 0.82-0.88) for 6-month mortality (I2 = 0%)

• Discrimination of machine learning models for 6-month mortality in infective endocarditis

  AUC 0.85 (95% CI 0.82 to 0.88)

  search to April 2026

  The review’s own words

  Pooled AUC was 0.85 (95% confidence interval [CI], 0.81-0.89) for in-hospital/early mortality (I2 = 35.3%) and 0.85 (95% CI, 0.82-0.88) for 6-month mortality (I2 = 0%)

AI models flag TB treatment failure, but only 23.5% of 34 studies tested them externally · TRV-2026-0817

• Discrimination of AI/ML models predicting tuberculosis treatment failure

  AUC 0.836 (95% CI 0.799 to 0.868)

  100,790 people · search to October 2025

  The review’s own words

  The pooled AUC was 0.836 (95% CI 0.799-0.868), with substantial heterogeneity (I² = 97.9%).

AI predicts brain bleed outcomes well in 83 studies, but mostly in internal tests · TRV-2026-0611

• Discrimination of clinical-radiomics models for predicting poor functional outcome in intracerebral hemorrhage

  C-index 0.850 (95% CI 0.830 to 0.869)

  search to September 2025

  The review’s own words

  models integrating both clinical and radiomics features achieved the highest discriminative performance across key prognostic prediction tasks: predicting HE (pooled C-index 0.822, 95% confidence interval [CI] 0.789-0.855), poor functional outcome (C-index 0.850, 95% CI 0.830-0.869), and mortality (C-index 0.860, 95% CI 0.809-0.911)

• Discrimination of clinical-radiomics models for predicting mortality in intracerebral hemorrhage

  C-index 0.860 (95% CI 0.809 to 0.911)

  search to September 2025

  The review’s own words

  models integrating both clinical and radiomics features achieved the highest discriminative performance across key prognostic prediction tasks: predicting HE (pooled C-index 0.822, 95% confidence interval [CI] 0.789-0.855), poor functional outcome (C-index 0.850, 95% CI 0.830-0.869), and mortality (C-index 0.860, 95% CI 0.809-0.911)

• Discrimination of clinical-radiomics models for predicting hematoma expansion in intracerebral hemorrhage

  C-index 0.822 (95% CI 0.789 to 0.855)

  search to September 2025

  The review’s own words

  models integrating both clinical and radiomics features achieved the highest discriminative performance across key prognostic prediction tasks: predicting HE (pooled C-index 0.822, 95% confidence interval [CI] 0.789-0.855), poor functional outcome (C-index 0.850, 95% CI 0.830-0.869), and mortality (C-index 0.860, 95% CI 0.809-0.911)