Machine Learning and Entropy Indices Reveal Most Mahanadi Water Near Paradip Unfit to Drink
A three-year geospatial study of the Mahanadi River near Paradip, Odisha, found that most monsoon-season surface water samples were unfit ...
Tag: Machine Learning
Particle shape beats gradation in setting sand shear strength, AI model shows
Triaxial tests on four granular mixtures show particle shape, not gradation, dominates shear strength, with XGBoost outperforming a neural network ...
Turbidity Currents Meet AI: Review Maps Numerical Models and Machine Learning
A new review maps how numerical simulations and artificial intelligence are reshaping the study of turbidity currents, the powerful underwater ...
Smart Farming’s Energy Problem: Review Maps Efficient Sensor Network Routing Protocols
A PRISMA-based systematic review of 43 studies published between 2016 and 2025 classifies energy-efficient routing protocols for agricultural wireless sensor ...
Diet scores boost genetic prediction of BMI and diabetes in Native Hawaiians
A study of 14,346 Native Hawaiians shows that combining trans-ancestry polygenic scores with machine-learning dietary scores substantially improves BMI prediction, ...
Macrophage-linked biomarkers flagged for esophageal cancer chemoradiotherapy response in computational study
A computational study identified KRT14 and MMP10 as macrophage-associated candidate biomarkers of chemoradiotherapy response in esophageal squamous cell carcinoma.
Machine Learning Flags Gambling Risk From Routine Betting Data Without Clinical Labels
A leakage-controlled machine learning pipeline converts routine online gambling telemetry into an auditable three-level risk flag that outperformed incumbent queues ...
AI Meets Solar Simulation to Track Moisture Damage in Perovskite Cells
A PRISMA-guided review of 105 studies argues that moisture-driven degradation, not efficiency, is the key barrier facing perovskite solar cells ...
Machine Learning Model Boosts Microgrid Power Forecasting Accuracy in Simulation
A simulation study reports that an Adaptive Tunicate Swarm Optimized Random Forest achieved higher power generation forecasting accuracy than LGBM ...
Mixed-effects radiomics model hits AUC 0.91 for lung cancer prediction across seven UK hospitals
A retrospective UK study of 1,026 patients found that a mixed-effects radiomics model achieved an AUC of 0.91 for predicting ...
Hybrid Physics and Data Framework Delivers Real-Time Forecasts of Atmospheric Pollutants
A hybrid physics-and-data framework decomposes pollutant forecasting into advection, learned diffusion, and a regressed gap term, achieving predictions in under ...
AI Learns the Physics of Magnetized Nanofluid Flow for Solar Thermal Design
Researchers built a physics-informed neural network surrogate that rapidly predicts heat and mass transfer in magnetized ternary nanofluid flow over ...
AI Model Predicts Strength and Surface Quality of 3D-Printed PLA Parts
A hybrid RVFL-HGSO machine learning model achieved a coefficient of determination (R2) of 98.95 percent in predicting the surface roughness ...
Research report: A proposed hybrid machine learning and physics-based optimization framework for energy-efficient smart buildings
A hybrid framework combining machine learning, physics-based thermal modelling and optimization achieved up to 26.4 percent energy savings on the ...
AI Shows Promise for Multiple Sclerosis Diagnosis, Review Finds Validation Gaps
A systematic review of 26 studies finds AI models diagnosing multiple sclerosis with over 95 percent accuracy in some cases, ...
British Society of Echocardiography Sets Out How AI Should Enter the Heart Scan Workflow
A new British Society of Echocardiography position statement maps where artificial intelligence is ready for clinical use across the echocardiography ...
Simple blood and lung tests may predict airway hyperresponsiveness before provocation testing
An interpretable model and scorecard built from routine lung function and blood markers predicted bronchial hyperresponsiveness with an AUC near ...
Massive Patch-Clamp Dataset Opens a Window on the Mouse Brain’s Sensory Gatekeeper
A new FAIR² dataset from EPFL releases full raw patch-clamp recordings from 19 juvenile mouse thalamic reticular nucleus neurons, spanning ...
Digital Pathology Tool Quantifies Amyloid Vessels and Plaques in CAA
A validated HALO AI digital pathology pipeline automatically distinguishes CAA-positive vessels from β-amyloid plaques and shows that higher CAA scores ...
Lightweight Knowledge Distillation Speeds Up Vulnerability Detection in Python Code
A new knowledge distillation method compresses large pre-trained models into lightweight detectors that identify vulnerabilities in Python code with Macro-F1 ...
