Truace
Subscribe
The Index recomputed live from the record

What the evidence says.What the public feels.

The record holds 935 sourced gains and 771 sourced problems, averaging 68 and 66 on the index score. Readers have logged 22 public signals on the Pulse, which is kept apart and never counted as evidence.

935 gains
771 problems
Every sourced claim in the Index, one square each, shaded by the strength of its evidence. High Moderate Emerging

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

1,706 results
Show filters and sorting

Download every matching row, not just this page:Export CSVExport JSON

AI gains · 935

337
HealthStableModerate evidence · 1 source

Deep learning-augmented plasmonic system PANDA profiled alpha-synuclein aggregate size distributions from human serum and classified Parkinson's disease stages with correlation to clinical scores and dopaminergic PET imaging.

Researchers developed PANDA, a plasmonic nanotechnology system that measures the size distribution of alpha-synuclein aggregates in human serum through immunogold clustering scattering signatures, and combined it with a deep neural network to classify Parkinson's disease stages. By publication date 2026-09-16, the system had been tested on human serum samples and showed correlations with clinical scores and dopaminergic PET imaging.

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

Updated Sep 18, 2026 · TRV-2026-1130

68Index score
338
ScienceStableModerate evidence · 1 source

An interpretable SISSO symbolic-regression AI approach identified materials genes, including d-band center and particle diameter, that statistically describe selectivity of supported palladium-based alloy nanoparticles during hydrogenation of concentrated acetylene streams under elevated pressure and temperature.

Researchers applied an interpretable AI method combining SISSO symbolic regression with partial-derivative sensitivity analysis to link physicochemical parameters to measured performance of supported palladium-based alloy nanoparticles during hydrogenation of concentrated acetylene streams under real-world pressure and temperature.

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

Updated Sep 18, 2026 · TRV-2026-1129

68Index score
339
HealthStableModerate evidence · 1 source

An explainable XGBoost model using six routine preoperative variables enabled individualized reflux esophagitis risk prediction for metabolic bariatric surgery candidates with AUC 0.881 training and 0.825 internal test.

In a retrospective study published September 17, 2026, researchers developed and internally validated machine learning models to predict preoperative reflux esophagitis among 1,068 metabolic bariatric surgery candidates. After feature selection, six routine variables were used, and an XGBoost model achieved the best performance with AUCs of 0.881 in training and 0.825 in internal testing, supported by SHAP interpretation and an online risk tool.

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

Updated Sep 18, 2026 · TRV-2026-1128

68Index score
340
HealthStableModerate evidence · 1 source

A relation-aware graph convolutional network method for medical AI legal compliance improved cross-modal legal semantic alignment and bias localization, achieving high entity alignment accuracy and root-node recall.

Researchers built a legal knowledge graph for medical AI compliance using four node types and four directed legal relations, then applied a relation-aware graph convolutional network with edge-type weights and attention to align cross-modal legal semantics. They used gradient-based backpropagation to extract bias-propagation subgraphs and community discovery to locate root-node clusters, mapping real-time decision flows to temporary nodes for compliance checking.

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

Updated Sep 17, 2026 · TRV-2026-1123

68Index score

AI problems · 771

337
HealthStableModerate evidence · 1 source

Clinical AI deployment without human-factors tools risks automation bias, overreliance, fragmentation of care, and other unintended consequences that threaten diagnostic safety.

On August 13, 2026, a peer-reviewed article in Diagnosis reported results from a two-day workshop of 18 interdisciplinary experts from three countries who sought to identify knowledge gaps in applying AI in clinical settings, using medical imaging as the primary use case.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
89

Updated Aug 14, 2026 · TRV-2026-0750

67Index score
338
HealthStableModerate evidence · 1 source

Up to half of patients receiving internet-delivered CBT for depression and anxiety disorders fail to achieve clinically significant symptom reduction, complicating treatment planning at intake.

In a routine-care sample of 1790 patients with major depressive disorder, panic disorder, and social anxiety disorder, researchers built multimodal machine learning models from pretreatment clinical, sociodemographic, and genetic data to predict clinically meaningful improvement after internet-delivered cognitive behavioral therapy, testing them with nested cross-validation and temporal validation in 356 held-out patients as of the August 2026 publication.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
88

Updated Aug 12, 2026 · TRV-2026-0738

67Index score
339
HealthStableModerate evidence · 1 source

AI methods in drug discovery have so far shown disappointingly limited clinically relevant impact on delivering safer and more efficacious medicines to patients faster.

A 2026 Perspective in Nature Reviews Drug Discovery reviews a decade of artificial intelligence in drug discovery, asking where advances have translated into safer and more efficacious medicines delivered faster. The authors note many AI methods have been developed, applied and benchmarked, but conclude evidence of clinically relevant impact is so far disappointingly limited.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
88

Updated Aug 11, 2026 · TRV-2026-0733

67Index score
340
HealthStableModerate evidence · 1 source

Performance varied significantly across models and domains, with Copilot and Perplexity scoring lower and greater inconsistency in clarity, structure, and clinical utility.

On 2026-08-07, a peer-reviewed comparative study tested five AI systems including the EAU Guidelines Bot, ChatGPT-5, Gemini 2.5 Pro, Copilot - Smart GPT-5, and Perplexity Pro on 13 questions drawn from strongly recommended EAU erectile dysfunction statements. Three senior reviewers rated each answer for relevance, clarity, structure, clinical utility, and factual accuracy.

Impact 30%
49
Evidence 25%
95
Scale 20%
35
Confidence 15%
87
Recency 10%
88

Updated Aug 10, 2026 · TRV-2026-0727

67Index score

Recomputed live from the record · Oct 12, 2026, 8:18 PM