Ranks distinct AI gain and problem claims from the published record. Scores reward impact, independent source strength, scale, confidence, and recency.
A systematic review and meta-research appraisal examined 20 studies comparing machine learning and logistic regression for trauma mortality prediction, with 17 studies (243,324 patients) in primary synthesis. The pooled within-study AUC difference favoring the best ML model was 0.026 (95% CI 0.009-0.043), 0.017 in co-primary analysis of studies reporting CIs, with extreme heterogeneity and a prediction interval crossing zero.
This PRISMA-guided systematic review examined 60 studies published between 2019 and February 2026 that used deep learning to classify Alzheimer's stages and predict conversion from mild cognitive impairment to Alzheimer's disease. It found cross-sectional designs predominant, CNNs dominant for neuroimaging, and growing use of RNNs and transformers for longitudinal data, with multimodal approaches in 24 studies.
Researchers developed a Categorical Chain Differential framework that couples prediction of four ash fusion temperatures across heterogeneous solid fuels. Using a regressor chain with fuel category information and non-negative constraints on inter-stage differences, the model enforces the physical order DT ≤ ST ≤ HT ≤ FT.
A Frontiers in Cognition review published in November 2023 surveyed how digital devices, social media platforms, and AI tools affect human cognition. It described these technologies as integral to daily life, bringing convenience and efficiency, while also examining their positive and negative influences on attention, memory, addiction, perception, decision-making, critical thinking and learning across different age groups.
A peer-reviewed review published August 13, 2026 examined how AI-enabled medical devices challenge traditional safety-risk management. Drawing on 19 academic and regulatory sources, it found ISO 14971, AAMI CR34971 and the EU AI Act each cover parts of device safety and algorithmic governance but remain fragmented in practice.
A peer-reviewed survey study from January to April 2025 asked 52 US dermatology and dermatopathology professionals to rate AI-simplified versions of six fictitious dermatopathology reports. One version used Basic ChatGPT-4.0 with a simple prompt and the other used a custom DermDecoder GPT with a structured 489-word prompt, evaluated for factualness, completeness, and potential harm.
Researchers tested Gemini 3.1 Pro in a zero-shot setting to detect lumbar disc herniations on sagittal MRI from 119 SPIDER cases (26% prevalence). Using only the mid-sagittal slice and a forced binary prompt, T1-only achieved 70% accuracy with 58% sensitivity and 74% specificity, while paired T1+T2 achieved 58% accuracy with 77% sensitivity and 51% specificity.
On August 12, 2026, a review in the International Journal of Toxicology summarized AI and machine learning use in toxicological risk assessment to predict chemical toxicity and support regulatory decisions, noting that explainable AI methods like SHAP and LIME are being explored to make model decisions interpretable.
ace