AI Core
896 items
Two-center PCD-CT spectral radiomics in 378 patients: VNC-trained model separates benign from malignant liver lesions at AUC 0.899, with high-keV and VNC test reconstructions most stable
In a two-center study of 378 patients with focal hepatic lesions (190 benign, 227 lesions; 188 malignant, 2681 lesions), radiomic features were extracted from nine PCD-CT spectral datasets (40, 50, 70, 90, 110, 140, 180 keV, virtual non-contrast [VNC], and iodine density maps [IDM]) after nnU-Net segmentation confirmed by an abdominal radiologist, multiple machine-learning models were trained with lesion-level predictions aggregated per patient, and the Random Forest model trained on VNC achieved the highest performance (AUC 0.899; 95% CI 0.840-0.951; accuracy 83.5%), while cross-domain analysis showed similar mean AUCs across training reconstructions (0.75 to 0.83) and test reconstruction choice mattered: 110, 140, 180 keV and VNC were most stable (median AUC >= 0.
A scientometric analysis of 269 articles shows XAI research in decision dashboards shifting from algorithm-centered to human- and decision-centered
Using scientometric methods on 269 original research articles retained after PRISMA screening from Scopus and Web of Science, this study finds that the intellectual core of XAI research in decision dashboards is organized around explainable AI, decision making, and deep learning, that themes have shifted over time from classical machine learning such as neural networks and support vector machines toward visual deep learning, visualization techniques, trust, and intelligent decision support systems, and that the field is moving from a technology-centered toward a human- and decision-centered paradigm.
Bibliometric Analysis and Co-word Mapping of the Knowledge Graph Field: A Review-Style Study Charts the Domain's Knowledge Base and Thematic Structure
The article titled "Bibliometric Analysis and Co-word Mapping: The Field of Knowledge Graphs" organizes and reviews literature on the knowledge graph research field using bibliometric and co-word analysis, with references spanning knowledge graph definitions and surveys, embedding methods, completion and refinement, domain-specific graphs, educational applications, and scientometric and co-word methods themselves; however, the text available here is only the reference list, and the body, figures, and specific bibliometric results are not included.
Oxford team review: pericoronary fat imaging turns coronary inflammation into a quantifiable metric, with FAI Score falling after lipid-lowering, biologics and radiotherapy
This review by a University of Oxford group synthesises the anatomical and physiological basis of pericoronary adipose tissue (PCAT) as a biosensor of coronary inflammation, explains how the standardised Fat Attenuation Index (FAI) Score corrects for technical, anatomical and biological variability and predicts MACE, reviews the role of artificial intelligence in automated segmentation and multi-parametric risk modelling, and compiles evidence that FAI falls after statins, anti-oxLDL antibodies, anti-TNF biologics, radiotherapy and cardiometabolic agents, concluding that PCAT imaging may complement traditional risk factors and plaque metrics while the evidence remains evolving.
This review maps AI's role in personalized medicine, from biomarker discovery to clinical decision support, while noting unresolved ethical, legal, and translation barriers
This review examines artificial intelligence in personalized medicine across biomarker discovery, disease subtyping, risk prediction, early diagnosis, treatment-response modelling, pharmacogenomics, digital pathology, longitudinal monitoring, and clinical decision support, and discusses the data types enabling personalization, the conceptual foundations, unresolved ethical and legal issues, and the obstacles between research potential and clinical practice, while emphasizing that AI does not take over the doctor's decision-making but helps doctors access more information than they could retain within their minds all at once.
AI-Driven Clinical Deterioration Index Flags High-Risk Geriatric Hip Fracture Patients After Surgery: 93.3% Complication Rate and 20.0% One-Year Mortality at CDI ≥65
In this single-center retrospective cohort of 197 geriatric patients who underwent fixation of OTA/AO 31-A/B/C hip fractures, an AI-driven Clinical Deterioration Index (CDI, 0–100 from 31 clinical measures) measured in the first 48 postoperative hours identified 15 patients (7.6%) with CDI ≥65 who had more complications (93.3% vs 26.4%), longer stays (10.1 vs 5.2 days), shorter discharge ambulation (8.8 vs 42.3 feet), and higher one-year mortality (20.0% vs 3.8%), while the institutional cutoff of 65 showed high specificity (99.3%) but low sensitivity (22.6%) and an optimized threshold of 47.7 raised sensitivity to 77.4% with comparable accuracy (75.0%).
Author argues psychiatry big data needs theory-driven, within-person intensive sampling and generative AI integration to address atheoretical predictive research
This commentary, responding to Stein et al.'s critical review of big data in psychiatry, argues that psychiatry has swung from 'all theory and no data' to 'all data and no theory,' and proposes three directions: greater consideration of theory in big data use, development of idiographic psychiatry, and integration of generative AI into psychiatric research and practice.
XGBoost beat NeuralProphet for Budapest metro M4 demand forecasting, and its SHAP explanations matched expert expectations
Using one year of hourly boarding data from Budapest metro line M4 (216 usable days, 05:00–24:00) restricted to a single station and direction (Kálvin tér toward Kelenföld vasútállomás) and a 6-hour forecast window, the study fairly compared inherently interpretable NeuralProphet with a black-box XGBoost interpreted post hoc via SHAP, finding XGBoost more accurate (MAE 91.0753 vs 230.2803, RMSE 144.7094 vs 363.2115, R-squared 0.9600 vs 0.7896) and SHAP temporal-feature insights consistent with NeuralProphet seasonality components and transport expert knowledge.
Pretraining on diverse data and fine-tuning with 500 target recordings lifted a sperm whale click-train detector's AUC from as low as 0.60 to 0.78–0.97 on unseen datasets
Starting from a previously trained temporal convolutional network sperm whale click-train detector, the study compared four transfer approaches across four sources (BAL, CS, ICE, MED)—cross-dataset baseline evaluation, training from scratch on only 500 target recordings, pretraining with random fine-tuning, and pretraining with active (uncertainty-based) fine-tuning—and found that pretrained models dropped in performance on unseen data but that fine-tuning with 500 target recordings effectively mitigated the drop, with active fine-tuning consistently outperforming the other approaches although its gain over random fine-tuning was marginal.
Page 1 · showing 10