Truace
Subscribe
Good Space a ranking of the record

TheGood surrounding AI

574 documented gains are on the board. 82% rest on at least one peer-reviewed study, and 8% clear the high-evidence bar. Health carries 49% of it. The latest was filed Oct 16, 2026.

Every gain on the board, one square each, shaded by the strength of its evidence. High 45 Moderate 529 Emerging 0
574 results
Show filters and sorting

The board · page 15 of 20

421
Evidence-backed gainPeer-reviewedScience

GPT-3.5 and especially GPT-4 produced narratives that were more progressive on gender roles and sexuality than human crowdworker narratives when responding to identical prompts about creating and falling in love with an artificial human.

Source article: Experimental narratives: A comparison of human crowdsourced storytelling and AI storytelling

Humanities and Social Sciences Communications
Reader signal

How should this claim be treated?

423
Evidence-backed gainPeer-reviewedHealth

AI techniques including machine learning and deep learning accelerate drug discovery and delivery and improve patient outcomes and treatment regimens in personalized medicine.

Source article: Artificial Intelligence (AI) Applications in Drug Discovery and Drug Delivery: Revolutionizing Personalized Medicine

Pharmaceutics
Reader signal

How should this claim be treated?

424
Model-prefilled gainPeer-reviewedHealth

A gradient boosting classifier trained on eight routine treatment-time features improved prediction of local control after Gamma Knife radiosurgery to post-resection cavities from brain metastases, achieving ROC-AUC 0.735 compared to chance-level baseline.

Source article: Machine learning prediction of local control after Gamma Knife radiosurgery to post-resection cavities from brain metastases: a proof-of-concept study

Journal of Neuro-Oncology
Reader signal

How should this claim be treated?

425
Model-prefilled gainPeer-reviewedHealth

AI-enhanced insertable cardiac monitors reduced nonactionable alerts from 5,078 to 2,110 per year for a 600-patient clinic, saving 559 staffing hours and $29,470 annually compared to non-AI monitors.

Source article: Impact of Artificial Intelligence-Enhanced Insertable Cardiac Monitors on Device Clinic Workflow and Resource Utilization

JACC: Advances
Reader signal

How should this claim be treated?

426
Model-prefilled gainPeer-reviewedHealth

AI models applied to neurosurgical data achieved Dice scores of 0.82-0.84 for tumor segmentation and AUC values of 0.80-0.90 for molecular prediction and outcome forecasting, supporting lesion detection, surgical navigation, and prognostication.

Source article: From severity scoring to predictive analytics: the emerging role of AI in neurosurgery

Frontiers in Neurology
Reader signal

How should this claim be treated?

427
Model-prefilled gainPeer-reviewedHealth

In real-world use among 254 young children with developmental concern, the AI-based Canvas Dx provided accurate autism predictions with high negative and positive predictive values and enabled diagnosis at a median age of 37.2 months.

Source article: An analysis of the real world performance of an artificial intelligence based autism diagnostic

Scientific Reports
Reader signal

How should this claim be treated?

428
Evidence-backed gainPeer-reviewedHealth

A transformer-based deep learning framework using clinically indicated non-contrast chest CT stratified risk of refractory Mycoplasma pneumoniae pneumonia in children with AUCs around 0.89-0.90 on internal and external test cohorts.

Source article: A CT-based deep learning model for the automated risk stratification of refractory Mycoplasma pneumoniae pneumonia in children

BMC Medical Imaging
Reader signal

How should this claim be treated?

429
Evidence-backed gainPeer-reviewedMedia & Arts

Cloud-edge synergy architecture for AI music generation empowers educational use by enabling resource democratization, personalized learning, and pedagogical innovation in latency-sensitive settings.

Source article: Integrating AI and cloud-edge technologies for music creation in educational and performance domains

Journal of Cloud Computing
Reader signal

How should this claim be treated?

430
Evidence-backed gainPeer-reviewedMedia & Arts

Among 496 Arabian users, intention to use GenAI for media content creation was most strongly driven by hedonic motivation alongside performance expectancy and other UTAUT2 factors, and actual use was positively driven by facilitating conditions, habit, trust and intention.

Source article: Generative AI and Media Content Creation: Investigating the Factors Shaping User Acceptance in the Arab Gulf States

Journalism and Media
Reader signal

How should this claim be treated?

434
Evidence-backed gainPeer-reviewedHealth

Machine learning models that incorporate compressed respiratory and oxygenation timeseries improved early prediction of BPD at 7 days after birth compared to clinical-data-only models.

Source article: Prediction of bronchopulmonary dysplasia seven days after birth using respiratory and oxygenation timeseries with machine learning

Pediatric Research
Reader signal

How should this claim be treated?

435
Evidence-backed gainPeer-reviewedHealth

Explainable ML models used after Wells/Geneva triage improved early PTE prediction in ED patients, with Extra Trees reaching 0.82 accuracy and 0.83 AUC.

Source article: Explainable machine learning for early prediction and anatomical classification of pulmonary embolism in the emergency department

BMC Emergency Medicine
Reader signal

How should this claim be treated?

436
Evidence-backed gainPeer-reviewedHealth

A six-protein plasma signature tested with SVM distinguished fibrotic hypersensitivity pneumonitis from idiopathic pulmonary fibrosis on an independent test set with 71.4% accuracy, offering a non-invasive diagnostic aid.

Source article: Plasma proteomics and machine learning deliver non-invasive distinction between fibrotic hypersensitivity pneumonitis and idiopathic pulmonary fibrosis

Journal of Translational Medicine
Reader signal

How should this claim be treated?

437
Evidence-backed gainPeer-reviewedHealth

LightGBM model trained on 282 Lenke 1/2 AIS patients predicted postoperative coronal imbalance with AUC 0.885 training and 0.824 internal validation, identifying LIV-LSTV, Lumbar Modifier, and Risser grade as key predictors.

Source article: Predicting postoperative coronal imbalance in Lenke 1/2 adolescent idiopathic scoliosis: A machine learning model with clinical interpretability

International Orthopaedics
Reader signal

How should this claim be treated?

438
Evidence-backed gainPeer-reviewedHealth

Multi-Scale Feature Fusion model combining U-Net segmentation, EfficientNet and attention autoencoder features fused via CCA and YOLO classification achieved 99.95% accuracy on apple leaf disease datasets, enabling early detection for sustainable agriculture.

Source article: Comprehensive plant disease classification and severity estimation for sustainable farming via automatic segmentation and multi-scale feature fusion

BMC Plant Biology
Reader signal

How should this claim be treated?

440
Evidence-backed gainPeer-reviewedHealth

AI-assisted caries annotation increased dental students' diagnostic accuracy, sensitivity, and confidence while reducing interpretation time on panoramic radiographs.

Source article: Effect of AI-assisted caries annotation on dental students' performance in caries detection on panoramic radiographs

BMC Medical Education
Reader signal

How should this claim be treated?

441
Evidence-backed gainPeer-reviewedHealth

Convolutional neural networks trained on four Scheimpflug-based corneal maps differentiated keratoconus from normal astigmatic eyes with up to 99.2% accuracy and AUC 1.00, with external validation retaining 97-98% accuracy.

Source article: Artificial intelligence for diagnosis of keratoconus using Scheimpflug based corneal tomography

International Journal of Ophthalmology
Reader signal

How should this claim be treated?

443
Evidence-backed gainPeer-reviewedPolicy

Researchers developed the University Policy Development Framework for Generative AI to help universities assess priorities and build sustainable governance capacity.

Source article: A framework for developing university policies on generative AI governance: a cross-national comparative study

Studies in Higher Education
Reader signal

How should this claim be treated?

445
Evidence-backed gainPeer-reviewedHealth

General-purpose LLMs can encode EHRs as plain-text descriptions to produce embeddings for clinical prediction without private medical training data, matching a specialized EHR foundation model across 15 tasks and improving on some tasks in UK Biobank external validation.

Source article: Large language models are powerful electronic health record encoders

npj Digital Medicine
Reader signal

How should this claim be treated?

446
Evidence-backed gainPeer-reviewedHealth

Machine learning classifiers using demographic, clinicopathological and inflammatory markers predicted 5-year overall survival and second primary cancer occurrence in a European nasopharyngeal carcinoma cohort.

Source article: Machine learning prognostication in nasopharyngeal carcinoma: a european multicentre analysis of survival and risk of second malignancy

European Archives of Oto-Rhino-Laryngology
Reader signal

How should this claim be treated?

447
Evidence-backed gainPeer-reviewedHealth

XGBoost model integrating UHR and iPTH predicted protein-energy wasting in incident hemodialysis patients with AUC 0.801, enabling individualized risk assessment.

Source article: Predictive value of the uric acid to high-density cholesterol ratio (UHR) combined with intact parathyroid hormone for protein-energy wasting after incident hemodialysis: a multicenter study

Renal Failure
Reader signal

How should this claim be treated?

448
Evidence-backed gainPeer-reviewedSports

An AI decision support system using ID3 entropy and enhanced Monte Carlo tree search delivered real-time basketball strategy evaluation in about 3.13 seconds with 74% win rate and 84% decision rationality.

Source article: Application of Artificial Intelligence in Real-Time Strategic Decision Support System for Sports Competitions

Ingegneria Sismica
Reader signal

How should this claim be treated?

449
Evidence-backed gainPeer-reviewedHealth

Deep learning CT segmentation using Comp2Comp enabled compartment-specific measurement of visceral adipose tissue, subcutaneous fat, and muscle, revealing sustained visceral fat reduction after metabolic and bariatric surgery that BMI alone does not capture.

Source article: Beyond BMI: Deep Learning Segmentation-Driven CT Reveals Body Composition Changes after Metabolic and Bariatric Surgery

Journal of the American College of Surgeons
Reader signal

How should this claim be treated?