OPCNet Learns Oscillatory Patterns by Reading Data as an ODE Solution
The new OPCNet neural network fits highly oscillatory data by learning the dynamics of a linear ODE and fitting the ...
Tag: Machine Learning
Machine Learning Model Predicts Residual Strength of Aged Silk Heritage Without Damage
Researchers trained an XGBoost machine learning model to estimate the residual tensile strength of artificially aged silk textiles from nondestructive ...
New open-source tool reads building types and heating loads straight from smart meters
Hourly smart meter readings can now reveal far more than a building's total electricity consumption. Researchers have developed COF-Tool, an ...
AI Model Designed to Predict Antiretroviral Resistance in HIV Patients in Cameroon
Researchers in Cameroon built and tested a machine learning model that predicts antiretroviral resistance from routine clinical data, with XGBoost ...
AI in kidney care shows promise but clinical proof still lags behind
A narrative review finds AI models across nephrology achieve strong predictive performance, yet prospective validation and evidence of patient benefit ...
Uncertainty-aware machine learning maps the real range of electric cargo vans
A comparative study of seven uncertainty-aware machine learning frameworks on real-world cargo EV data shows trade-offs between prediction accuracy, interval ...
Explainable AI forecasts hourly traffic on Interstate I-94
A condition-stratified study of westbound Interstate I-94 found that XGBoost without weather variables delivered the most accurate hour-ahead traffic forecasts.
Wristwatch-style sensor reads touch through the arm with deep learning
A wrist-worn motion sensor and a compact neural network detected prosthetic hand contact with 99.1 percent accuracy in lab tests.
Machine learning and robotic synthesis race to break methane conversion barrier
Machine learning and robotic synthesis are accelerating the search for catalysts that convert methane directly into ethylene, though researchers say ...
Climate change threatens food safety: three computational tools could help manage the risk
A new Perspective article outlines how Multi-Criteria Decision Analysis, predictive microbiology and machine learning can support microbiological food safety management ...
Five bacterial species flag bacterial vaginosis in sequencing-based diagnostic test
A two-center study reports that a random forest model using five bacterial species detected by probe capture-based metagenomic sequencing classified ...
One-motion and two-motion free throws rely on distinct biomechanical and neuromuscular strategies in male basketball enthusiasts
A motion capture and EMG study of twenty male basketball enthusiasts found that one-motion and two-motion free throws differ in ...
Stroke sarcopenia risk models show promise but lack validation
A scoping review of 12 studies finds that risk prediction models for stroke-associated sarcopenia show promising discrimination but almost universally ...
Machine learning reveals when changing shot direction pays off in elite tennis
A machine learning framework analysing 33,975 strokes from Grand Slam men's singles matches shows that the tactical value of redirecting ...
Lung Cancer Pneumonia Risk Models Look Accurate but Almost All Are Flawed
A systematic review of 26 lung cancer pneumonia risk models finds a pooled AUC of 0.89 but high risk of ...
Machine learning predicts wastewater aeration efficiency, but only when data stay close to home
Machine learning can predict the dynamic alpha-zero factor governing wastewater aeration efficiency with high accuracy on random data splits, but ...
Machine Learning Flags Airway Cast Risk in Children With Mycoplasma Pneumonia
A single-center retrospective study of 702 hospitalized children with Mycoplasma pneumoniae pneumonia developed eight machine learning models that identified nine ...
Left Atrial Strain Model Flags Heart Risk in Uremia Patients
A retrospective Chinese study built a LASSO-logistic model using four-dimensional left atrial quantification parameters that predicted major adverse cardiovascular events ...
Machine Learning Flags Postoperative Clot Risk After Hip Fracture Surgery, but TCM Patterns Add Little
A retrospective study of 380 patients found a random-forest model discriminated well in a held-out validation set, while two Traditional ...
No single number decides which teeth come out, systematic review finds
A systematic review of 19 studies finds that orthodontic extraction decisions depend on crowding, facial profile, incisor position, skeletal and ...
POPULAR NEWS
-
Vaping adults without smoking history report slightly more respiratory symptoms, large survey finds
-
AI Advances in Upper Limb Prosthetics Reviewed
-
Papain Pre-Digestion Unlocks Potent Bioactive Peptides From Snapper Salmon Skin Collagen
-
Deep Learning Model Accelerates Drug Target Screening and Repurposing Opportunities