± HEALTH Colorectal Cancer (CRC) is a common cause of cancer death and prognostic factors are used to determine manage…+ HEALTH This multicenter study developed and validated an interpretable machine learning model integrating granular n…+ HEALTH Multiple stakeholders need to locate results of registered clinical trials but frequently struggle to find th…+ HEALTH Accurate preoperative glioma grading and molecular subtyping are important for treatment. The vascular microe…+ HEALTH Gastrointestinal endoscopy generates extensive high-resolution video data, posing significant challenges for…+ HEALTH Non-invasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP) and invasive encapsulat…+ POLICY Artificial intelligence (AI) has the potential to strengthen real-world evidence (RWE) for regulatory decisio…± HEALTH Echocardiography remains the cornerstone of cardiovascular imaging. However, traditional workflows including…
TruaceTracing the truth around AIFriday, August 28, 2026
Subscribe
The Index

What the evidence says.What the public feels.

Ranks distinct AI gain and problem claims from the published record. Scores reward impact, independent source strength, scale, confidence, and recency.

1,182 results
Show filters and sorting

AI gains · 658

68
GainHealth· Stable· Evidence: Moderate (1 source)

AI models for CT, MRI, and ultrasound diagnosis of urological cancers achieved higher pooled specificity and AUC than clinicians in a 110-study meta-analysis.

A meta-analysis of 110 studies up to June 2026 evaluated AI algorithms for diagnosing urological cancers on CT, MRI, and ultrasound, pooling sensitivity, specificity, and AUC and comparing to clinician performance where reported.

Impact 30%49
Evidence 25%95
Scale 20%35
Confidence 15%87
Recency 10%95

Updated Aug 4, 2026 · TRV-2026-0640

68
GainHealth· Stable· Evidence: Moderate (1 source)

An ensemble Stacking model using baseline clinical, neurological, and MRI features predicted one-year AIS grade and motor/independence scores in TCSCI patients with high discrimination and low error on external testing.

By August 2026, researchers had developed and externally validated a two-layer Stacking ensemble that integrates baseline clinical data, neurological assessments, and cervical MRI features to predict one-year outcomes after traumatic cervical spinal cord injury. In 340 patients analyzed, the model predicted AIS grade with AUC 0.85 and predicted continuous motor and independence scores with R8 above 0.986.

Impact 30%49
Evidence 25%95
Scale 20%35
Confidence 15%87
Recency 10%95

Updated Aug 4, 2026 · TRV-2026-0639

68
GainHealth· Stable· Evidence: Moderate (1 source)

Deep learning models demonstrated proof-of-concept ability to predict knee osteoarthritis progression from medical imaging, with internal median AUCs up to 0.87 for surgical endpoints.

This PRISMA systematic review evaluated 33 peer-reviewed studies (2019-2026) comprising 50 deep learning models that predict knee osteoarthritis progression from medical imaging. It extracted AUC as primary outcome, categorized nine different progression definitions, and assessed bias with PROBAST-AI, finding median internal AUCs of 0.87 for surgery, 0.78 for structural, and 0.79 for symptomatic endpoints.

Impact 30%49
Evidence 25%95
Scale 20%35
Confidence 15%87
Recency 10%95

Updated Aug 4, 2026 · TRV-2026-0638

68
GainClimate· Stable· Evidence: Moderate (1 source)

Hybrid STICS plus random forest integration improved accuracy and interpretability of apple fruit maturity date prediction to support harvest timing and climate adaptation across China's apple regions.

Researchers developed a hybrid framework that couples the STICS biophysical crop model with machine learning to predict apple fruit maturity dates across China. Using phenology records from 24 sites and weather data from 250 stations for 1991-2020, they found a random forest integration improved prediction accuracy and interpretability at regional scales.

Impact 30%49
Evidence 25%95
Scale 20%35
Confidence 15%87
Recency 10%95

Updated Aug 3, 2026 · TRV-2026-0632

AI problems · 524

68
ProblemEducation· Stable· Evidence: Moderate (1 source)

Generative AI adoption in higher education created persistent risks to academic integrity, data privacy, equity, and responsible governance.

This March 2026 systematic and thematic review examined generative AI tools such as ChatGPT in higher education, analyzing 46 Web of Science documents and qualitatively synthesizing 27 peer-reviewed articles to map implementation trends.

Impact 30%49
Evidence 25%95
Scale 20%35
Confidence 15%87
Recency 10%93

Updated Jul 20, 2026 · TRV-2026-0342

68
ProblemHealth· Stable· Evidence: Moderate (1 source)

In patients with LAD myocardial bridging, longer bridging length increased risk of abnormal FFRCT, with a larger effect in females, and females with isolated bridging had more pronounced distal hemodynamic compromise.

A retrospective study of 300 patients with left anterior descending artery myocardial bridging and 104 controls used an AI-based platform, Shukun-FFRCT, to obtain whole-vessel and segmental FFRCT values and relate them to bridging morphology and sex.

Impact 30%49
Evidence 25%95
Scale 20%35
Confidence 15%87
Recency 10%93

Updated Jul 20, 2026 · TRV-2026-0331

68
ProblemHealth· Stable· Evidence: Moderate (1 source)

Use of the same ambient AI scribes was associated with increased length of notes and no change in physician productivity measured by billing metrics.

A rapid review published April 29 2025 synthesized 6 real-world studies of digital scribes using ambient listening and generative AI from 1450 screened records spanning academic health systems, community settings, and outpatient practices. Across observational, case report, cohort, and survey designs, authors reported decreased self-reported documentation times with associated increased length of notes.

Impact 30%49
Evidence 25%95
Scale 20%35
Confidence 15%87
Recency 10%93

Updated Jul 20, 2026 · TRV-2026-0330

68
ProblemHealth· Stable· Evidence: Moderate (1 source)

Despite higher scores, the evaluated large language models have identified limitations that prohibit them from replacing human experts for triage in overcrowded emergency departments.

Researchers designed the Skyer benchmark to evaluate fifteen large language models on 55 realistic pediatric emergency department scenarios using a weighting system for over-triage and under-triage plus three repeat runs for consistency. By the publication date of July 11 2026, ChatGPT-4.5-preview and Gemini-2.5_05-06 had shown 77% and 74% accuracy with mean weights of 377.5 and 365 out of 550, compared to 64% and 253.5 for human experts.

Impact 30%49
Evidence 25%95
Scale 20%35
Confidence 15%87
Recency 10%93

Updated Jul 20, 2026 · TRV-2026-0327

Recomputed live from the record · Aug 28, 2026, 10:50 AM