Diagnostics & Imaging
281 gains94 Traces87 problems
Every measured gain and problem on the record for this topic, newest first. Updated .
AI in diagnostics and imaging
One square per record, darker for stronger evidence:Evidence: 7 high408 moderate47 emerging
Most of the evidence here is retrospective accuracy testing on past scans and records, with few randomized trials and little tracking of what happens to patients afterwards. TRV-2026-1434TRV-2026-1490TRV-2026-0486 Reviewers repeatedly say strong test scores have not yet been matched by proof of better care. TRV-2026-1017TRV-2026-0486TRV-2026-1924 Results look best on narrow detection tasks and weakest when models move to new hospitals or populations. TRV-2026-0638TRV-2026-1244TRV-2026-1220
A randomized trial of 14,665 adults showed AI-ECG-guided monitoring found new atrial fibrillation, flutter or pre-AF findings in 0.74% of intervention participants versus 0.48% of controls, though the authors call the result statistically fragile. TRV-2026-2462 In a randomized trial of 300 people with suspected inherited retinal disease, eye specialists using an AI tool had the right gene category among their top four answers 81.8% of the time versus 53.1% alone. TRV-2026-0575 A systematic review found AI-read ECGs detected weak heart pumping in people without symptoms with pooled AUROC of 0.87 to 0.93, and another review found AI detected STEMI heart attacks with sensitivity 0.91 versus 0.78 for clinicians. TRV-2026-1837TRV-2026-2312 A systematic review of pooled studies found AI assistance raised doctors' fracture detection sensitivity to 87% from 73%, with junior clinicians gaining most. TRV-2026-1145
A randomized trial of 171 veterans found AI-guided monitoring after a heart failure hospitalization did not significantly reduce rehospitalization (16.3% versus 18.7% for controls). TRV-2026-1587 In a crossover study of four commercial chest X-ray AI tools, AI assistance did not improve diagnostic accuracy for five readers reviewing 1200 patients at a single center. TRV-2026-0854 A systematic review found generative AI was 13.0 percentage points less accurate than physicians at radiology diagnosis. TRV-2026-1284 In Nepal, a study of 826 chest X-rays found only 17.89% of the AI's abnormal flags were correct at a threshold set for at least 95% sensitivity. TRV-2026-1265
Reviews of eye disease, echocardiography and heart ischemia AI say effects on safety, workflows and patient outcomes remain largely untested. TRV-2026-1490TRV-2026-1924TRV-2026-1434 A few encouraging results rest on very small samples: an AI app flagged all 13 occlusion heart attacks, but in only 33 selected, hard-to-read ECGs. TRV-2026-1923 In a study of 30 bowel disease patients, AI classifiers working on biopsy light scans did no better than chance (p = 0.51). TRV-2026-2047 A systematic review of rotator cuff tear AI found most diagnostic studies were rated high risk of bias, so reliability in real clinics is unproven. TRV-2026-1698
Whether AI help improves human readers is split: in a randomized crossover test, 21 physician assistants reached 92.9% accuracy on heart attack ECGs with AI versus 79.0% without it. TRV-2026-0651 A crossover study of four commercial chest X-ray tools found no accuracy gain for five readers. TRV-2026-0854 In a meta-analysis of urological cancer imaging, AI scored a pooled AUC of 0.91 against 0.83 for clinicians, yet was less sensitive than clinicians in the prostate and MRI subgroups. TRV-2026-0640
AI reading of scans, slides and ECGs often reports high accuracy for specific findings, mostly on retrospective data from one centre. TRV-2026-0676TRV-2026-0764TRV-2026-0640TRV-2026-1145TRV-2026-2312
See these 130 recordsModels forecasting survival, complications and recurrence usually beat standard scores on past data, often by modest margins. TRV-2026-1068TRV-2026-1828TRV-2026-1009TRV-2026-0919TRV-2026-1081
See these 130 recordsAccuracy often drops when models meet new hospitals, countries or scanners; reviews say external validation is sparse. TRV-2026-0638
A Trace is one record with both a gain and a problem, so it counts on both sides: 375 records show a gain and 181 show a problem. 172 context records are in the record list but not counted, because they report no measured gain or problem.
| Evidence | Gain side | Problem side |
|---|---|---|
| HighStrong study design, backed by at least two independent sources | 5 | 4 |
| ModerateA measured study or an official source | 346 | 141 |
| EmergingEarly, small, unreviewed or reported second-hand | 24 | 36 |
By type of evidence: 160 observational studies; 130 controlled comparisons; 74 benchmark or lab tests; 47 systematic reviews or meta-analyses; 35 commentaries or narrative reviews; 4 randomized trials; 4 news reports; 3 regulatory or legal actions; 2 sets of official statistics; 1 interview study; 1 survey; 1 case report. How evidence is scored
In a review of 150 clinically validated studies, AI reached expert-level accuracy in medical imaging, with AUC up to 0.94 for cancer detection.
In a model development and validation study of cancer patients, the highest-risk quarter flagged by a face-photo AI had more than triple the mortality of the lowest (adjusted hazard ratio 3.23); its age tool erred by 5.1 years on average.
In 12 externally validated studies of people without symptoms, AI-read ECGs detected weak heart pumping with pooled AUROC of 0.87 to 0.93, similar to results in people with symptoms.
In a study of kidney patients in the US and China, models built to classify heart disease performed worse when tested on the other country's patients, and key risk factors differed: age in the US, hypertension in China.
A single-center study found the software's sensitivity fell to 73% for distal M1 blockages, with only moderate agreement with experts; non-target occlusions were excluded from the main analysis, so it needs expert review rather than standalone use.
22 reviews in this topic printed a pooled result: one figure calculated across many studies. They are the reviews’ own numbers, checked word for word against their abstracts. They often measure different things, so they are listed side by side, not added up.
In a 110-study review, AI beat clinicians on urological cancer scans but lagged on prostate · TRV-2026-0640
Specificity 0.83 (95% CI 0.80 to 0.86)
110 studies · search to June 2026
AI models achieved pooled sensitivity and specificity of 0.85 (95% CI: 0.83-0.87) and 0.83 (95% CI: 0.80-0.86), with an AUC of 0.91 (95% CI: 0.88-0.93)
Sensitivity 0.82 (95% CI 0.79 to 0.85)
110 studies · search to June 2026
Clinicians demonstrated a pooled sensitivity of 0.82 (95% CI: 0.79-0.85) and specificity of 0.68 (95% CI: 0.62-0.73), with an AUC of 0.83 (95% CI: 0.80-0.86).
AUC 0.91 (95% CI 0.88 to 0.93)
110 studies · search to June 2026
AI models achieved pooled sensitivity and specificity of 0.85 (95% CI: 0.83-0.87) and 0.83 (95% CI: 0.80-0.86), with an AUC of 0.91 (95% CI: 0.88-0.93)
General chatbots do well on some narrow tasks but trail specialists and clinicians on others. TRV-2026-2253TRV-2026-1679TRV-2026-1352TRV-2026-2004TRV-2026-0758TRV-2026-1284
See these 34 recordsFew studies measure whether AI changes decisions or patient outcomes, and the randomized evidence that exists is mixed. TRV-2026-2462TRV-2026-1587TRV-2026-0575TRV-2026-0651TRV-2026-0854TRV-2026-1490
See these 31 recordsMachine learning on biomarkers and spectra often separates patients from controls in small studies that have not been tested in clinics. TRV-2026-0728TRV-2026-0896TRV-2026-0616TRV-2026-1966TRV-2026-1256
See these 38 recordsEach record sits in one thread at most. A record with both a gain and a problem counts on both sides. 45 records fit none of these threads.
Blue Ortho recalled Predict+, a machine-learning tool that predicts outcomes after shoulder replacement surgery, on 2026-02-05 because it lacked pre-market clearance/approval.
AI picked out subdural brain bleeds on CT with high accuracy, a review of 67,266 scans shows · TRV-2026-0676
Precision 0.983 (95% CI 0.933 to 0.996)
3 studies · 67,266 people · search to December 2025
Among DL models, the pooled precision of CNN and U-Net models was 0.763 (95% CI [0.560, 0.891]; I2 = 98.9%; n = 12) and 0.983 (95% CI [0.933, 0.996]; I2 = 4.6%; n = 3), respectively (Fig. 5).
Accuracy 0.951 (95% CI 0.751 to 0.992)
3 studies · 67,266 people · search to December 2025
Among DL models, the pooled accuracy of CNN and U-Net models was 0.903 (95% CI [0.863, 0.932]; I2 = 97.7%; n = 13) and 0.951 (95% CI [0.751, 0.992]; I2 = 93.2%; n = 3), respectively.
Specificity 0.899 (95% CI 0.865 to 0.926)
15 studies · 67,266 people · search to December 2025
Among DL models, the pooled specificity of CNN, U-Net, and hybrid models was 0.899 (95% CI [0.865, 0.926]; I2 = 98.7%; n = 15), 0.981 (95%
AI beat clinicians at spotting serious heart attacks on ECGs in 8 studies, but patient impact is unproven · TRV-2026-2312
Specificity 0.95 (95% CI 0.90 to 0.98)
20 studies · search to March 10, 2025
In 20 studies comparing AI with a clinical reference standard, pooled sensitivity was 0.95 (95% confidence interval, 0.91-0.97), specificity 0.95 (0.90-0.98), AUC 0.98 (0.97-0.99), positive likelihood ratio 19.2 (9.3-39.5), and negative likelihood ratio 0.05 (0.03-0.10).
Sensitivity 0.95 (95% CI 0.91 to 0.97)
20 studies · search to March 10, 2025
In 20 studies comparing AI with a clinical reference standard, pooled sensitivity was 0.95 (95% confidence interval, 0.91-0.97), specificity 0.95 (0.90-0.98), AUC 0.98 (0.97-0.99), positive likelihood ratio 19.2 (9.3-39.5), and negative likelihood ratio 0.05 (0.03-0.10).
AUC 0.98 (95% CI 0.97 to 0.99)
20 studies · search to March 10, 2025
In 20 studies comparing AI with a clinical reference standard, pooled sensitivity was 0.95 (95% confidence interval, 0.91-0.97), specificity 0.95 (0.90-0.98), AUC 0.98 (0.97-0.99), positive likelihood ratio 19.2 (9.3-39.5), and negative likelihood ratio 0.05 (0.03-0.10).
AI help raised fracture detection to 87% from 73%; junior doctors gained most, a review shows · TRV-2026-1145
Sensitivity 73% (95% CI 69% to 78%)
17 studies · search to September 17, 2025
Compared to unassisted diagnosis, AI significantly improved pooled sensitivity (87%, 95% confidence interval [CI]: 84-89%, versus 73%, 95% CI: 69-78%) and maintained high specificity (95%, 95% CI: 92-97%, versus 94%, 95% CI: 89-96%).
Specificity 95% (95% CI 92% to 97%)
17 studies · search to September 17, 2025
Compared to unassisted diagnosis, AI significantly improved pooled sensitivity (87%, 95% confidence interval [CI]: 84-89%, versus 73%, 95% CI: 69-78%) and maintained high specificity (95%, 95% CI: 92-97%, versus 94%, 95% CI: 89-96%).
Specificity 94% (95% CI 89% to 96%)
17 studies · search to September 17, 2025
Compared to unassisted diagnosis, AI significantly improved pooled sensitivity (87%, 95% confidence interval [CI]: 84-89%, versus 73%, 95% CI: 69-78%) and maintained high specificity (95%, 95% CI: 92-97%, versus 94%, 95% CI: 89-96%).
Machine learning's edge over standard statistics in predicting trauma deaths is small and may be inflated · TRV-2026-1068
AUC difference 0.026 (95% CI 0.009 to 0.043)
17 studies · 243,324 people · search to January 15, 2025
The pooled AUC difference favoring ML was 0.026 (95% CI 0.009-0.043)
AUC difference 0.017 (95% CI 0.005 to 0.029)
search to January 15, 2025
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
SSIM 0.92 (95% CI 0.90 to 0.93)
15 studies · search to January 16, 2025
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).
PSNR 30.6 (95% CI 28.6 to 32.6)
15 studies · search to January 16, 2025
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
Sensitivity 0.737 (95% CI 0.662 to 0.799)
14 studies · search to July 2026
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).
Specificity 0.789 (95% CI 0.709 to 0.851)
14 studies · search to July 2026
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 10.49 (95% CI 6.28 to 17.53)
14 studies · search to July 2026
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
Accuracy (SROC) 97% (95% CI 95% to 98%)
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).
Accuracy (SROC) 97% (95% CI 95% to 98%)
4 studies · 538 people
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 88% (95% CI 76% to 94%)
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).
Review of 16 studies finds jaw-joint AI spots disc displacement better on MRI than on X-ray scans · TRV-2026-2861
Specificity 0.741 (95% CI 0.592 to 0.850)
search to 18 June 2026
For disc displacement, magnetic resonance imaging input yielded pooled sensitivity 0.924 (95% CI, 0.859-0.960) and specificity 0.812 (0.709-0.885), compared with 0.765 (0.618-0.867) and 0.741 (0.592-0.850) for panoramic or cone-beam computed tomography input.
Sensitivity 0.765 (95% CI 0.618 to 0.867)
search to 18 June 2026
For disc displacement, magnetic resonance imaging input yielded pooled sensitivity 0.924 (95% CI, 0.859-0.960) and specificity 0.812 (0.709-0.885), compared with 0.765 (0.618-0.867) and 0.741 (0.592-0.850) for panoramic or cone-beam computed tomography input.
Specificity 0.812 (95% CI 0.709 to 0.885)
search to 18 June 2026
For disc displacement, magnetic resonance imaging input yielded pooled sensitivity 0.924 (95% CI, 0.859-0.960) and specificity 0.812 (0.709-0.885), compared with 0.765 (0.618-0.867) and 0.741 (0.592-0.850) for panoramic or cone-beam computed tomography input.
Review of 7 studies finds AI mammogram reading trades missed cancers for false alarms · TRV-2026-2725
Proportion 0.12% (95% CI 0.06% to 0.26%)
The pooled FNP increased incrementally from 0.02% [95% CI 0.01, 0.03] at Transpara 3 to 0.12% [95% CI 0.06, 0.26] (Transpara 9), consistent with a trade-off with FPP.
Proportion 71.83% (95% CI 69.67% to 73.90%)
Pooled FPP decreased incrementally with increasing Transpara threshold (71.83% [95% CI 69.67, 73.90] at Transpara 3 to 10.77% [95% CI 8.34, 13.79] at Transpara 9).
Proportion 10.77% (95% CI 8.34% to 13.79%)
Pooled FPP decreased incrementally with increasing Transpara threshold (71.83% [95% CI 69.67, 73.90] at Transpara 3 to 10.77% [95% CI 8.34, 13.79] at Transpara 9).
AI on heart ultrasounds spots a hard-to-diagnose heart failure, but evidence is thin · TRV-2026-2311
Area under the summary ROC curve 0.875 (95% CI 0.791 to 0.930)
2,770 people · search to June 2026
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 24.73 (95% CI 18.46 to 33.12)
2,770 people · search to June 2026
The diagnostic odds ratio was 24.73 (95% CI: 18.46-33.12)
Sensitivity 84.0% (95% CI 73.0% to 91.0%)
2,770 people · search to June 2026
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 91% (95% CI 77% to 97%)
search to July 30, 2025
The pooled sensitivity and specificity were 91% (95% confidence interval [CI]: 77%–97%) and 99% (95% CI: 94%–100%), respectively.
Specificity 99% (95% CI 94% to 100%)
search to July 30, 2025
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 0.78 (95% CI 0.63 to 0.88)
219 people · search to October 1, 2025
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).
AUC 0.81 (95% CI 0.69 to 0.91)
219 people · search to October 1, 2025
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 0.75 (95% CI 0.63 to 0.84)
219 people · search to October 1, 2025
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
AUC 0.89 (95% CI 0.86 to 0.92)
search to 8th September 2025
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.
AUC 0.89 (95% CI 0.85 to 0.97)
search to 8th September 2025
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%.
AUC 0.93 (95% CI 0.88 to 0.97)
search to 8th September 2025
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 0.955 (95% CI 0.929 to 0.972)
search to January 2026
Pooled sensitivity was 0.965 (95% CI: 0.944-0.978) and pooled specificity was 0.955 (95% CI: 0.929-0.972).
Sensitivity 0.965 (95% CI 0.944 to 0.978)
search to January 2026
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 594.1 (95% CI 263.3 to 1340.4)
search to January 2026
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 regression coefficient (β) for AUROC -0.056 (95% CI -0.105 to -0.007)
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
Accuracy 42.9% (95% CI 35.8% to 50.1%)
search to March 2025
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 (percentage points) -25.9 (95% CI -39.1 to -12.8)
search to March 2025
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)
Accuracy 58.1% (95% CI 50.0% to 66.2%)
search to March 2025
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 0.888 (95% CI 0.789 to 0.944)
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 0.937 (95% CI 0.906 to 0.958)
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.
Review of AI phone-photo mouth screening finds 90% detection, but no real-world validation · TRV-2026-1261
Specificity 89% (95% CI 86% to 92%)
The pooled sensitivity was 90% (95% CI: 88.8%-92.6%) and specificity was 89% (95% CI: 86%-92%).
Sensitivity 90% (95% CI 88.8% to 92.6%)
The pooled sensitivity was 90% (95% CI: 88.8%-92.6%) and specificity was 89% (95% CI: 86%-92%).
AI predicts death from heart valve infection well, but half the studies had bias risk · TRV-2026-1049
AUC 0.85 (95% CI 0.81 to 0.89)
search to April 2026
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%)
AUC 0.85 (95% CI 0.82 to 0.88)
search to April 2026
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
AUC 0.836 (95% CI 0.799 to 0.868)
100,790 people · search to October 2025
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
C-index 0.850 (95% CI 0.830 to 0.869)
search to September 2025
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)
C-index 0.860 (95% CI 0.809 to 0.911)
search to September 2025
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)
C-index 0.822 (95% CI 0.789 to 0.855)
search to September 2025
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)
Page 17 of 43
Researchers built an interpretable Random Forest model using retinal measurements pulled automatically from the first OCT scan of 329 untreated eyes with central serous chorioretinopathy. Eyes were later labeled acute or chronic depending on whether subretinal fluid resolved during follow-up averaging 4.2 years.
American Journal of Ophthalmology
Researchers analyzed data from 260 cancer patients with machine learning to group them by pain profile rather than by pain intensity alone.
Researchers built an inpatient prototype that connects scalp or intracranial EEG to a secure large language model, which exchanges messages with patients through a smartphone app.
The paper examines why AI models for reading radiological images are rarely adopted in clinics even when academic versions perform accurately.
Researchers tested ChatGPT-5.6 Sol on 100 synthetic thyroid ultrasound reports, each describing one nodule and the five ACR TI-RADS feature categories.
Researchers built TrajOnco, a chain-of-agents large language model framework that reads sequences of clinical events in electronic health records and produces patient-level summaries, rationales…
Researchers ran a pilot experiment in which 44 medical geneticists, split into an AI-only group and an XAI-supported group, judged 18 facial images of people with or without genetic conditions.
Researchers tested a convolutional neural network system that sorts blood cells on peripheral smears imaged with the Sysmex DI-60.
Researchers used records from two intensive care databases, MIMIC-IV and MIMIC-III, to identify 1249 patients who had both sepsis-associated encephalopathy and heart failure.
A systematic review searched Embase, MEDLINE and CENTRAL and screened 10,645 records. Six studies met the criteria for models that use routine bedside vital signs to predict cardiac arrest in general…
Researchers built RARE-PHENIX, a framework that chains together three steps of rare disease phenotyping: large language model extraction of features from clinical text, mapping of those features to…
Researchers built AutoRSR, a system that uses speech recognition, natural language processing, and machine learning to transcribe children's recordings of a sentence recall task, count errors, score…
Researchers trained a deep residual neural network with attention on single-lead ECGs recorded during overnight polysomnography, combined with sleep stage data, to predict 10-year risk of atrial…
Researchers built the Deep Learning-Radiomic-Text Transformer, which combines CT-derived deep learning features, radiomics, clinical variables, and pathology report embeddings from a large language…
The LINK-HF2 randomized study, run at five Veterans Affairs medical centers, enrolled adults discharged after acute heart failure hospitalization.
ace