November 18, 2026
Health news, explained with evidence
LIVE
Preventive Services Task Force Resumes Meetings Under New Leadership Blood-Based Genetic Signature May Predict CAR T-Cell Therapy Success in Lymphoma Ultraprocessed Diets May Impact Brain Reward Systems in Younger Adults Addressing Body Image and the Reality of Gluteal Anatomy Migraine and Smoking Linked to Increased Early Stroke Risk in Women CBD Oil Found Ineffective for Managing Behavioral Issues in Youths with Intellectual Disabilities Expert Advice on Hydration Gear Sparks Debate Among Recreational Runners A Structured Approach to Beginning a Strength and Cardio Routine The Role of Daily Mobility Training in Managing Sedentary Tension New Theoretical Framework Explores Memory Formation and Dementia Preventive Services Task Force Resumes Meetings Under New Leadership Blood-Based Genetic Signature May Predict CAR T-Cell Therapy Success in Lymphoma Ultraprocessed Diets May Impact Brain Reward Systems in Younger Adults Addressing Body Image and the Reality of Gluteal Anatomy Migraine and Smoking Linked to Increased Early Stroke Risk in Women CBD Oil Found Ineffective for Managing Behavioral Issues in Youths with Intellectual Disabilities Expert Advice on Hydration Gear Sparks Debate Among Recreational Runners A Structured Approach to Beginning a Strength and Cardio Routine The Role of Daily Mobility Training in Managing Sedentary Tension New Theoretical Framework Explores Memory Formation and Dementia

New Machine Learning Approach Evaluates Digitized Spiral Drawings for Parkinson’s Disease Detection

🏃 Fitness 2 min read 👁 2

Researchers in Nature Health Sciences evaluated a machine learning framework that uses digitized spiral drawings to classify Parkinson's disease and healthy controls.

🏃Fitness

Key points

  • Machine learning framework tested on digitized spiral drawings for Parkinson's disease
  • Model achieved an 86.46% five-fold cross-validation accuracy on a small development set
  • Authors stress the findings show technical viability, not definitive diagnostic ability

A recent study published in Nature Health Sciences explores an automated machine learning framework designed to classify Parkinson's disease using digitized spiral drawings. Developed by an international team of researchers, the system aims to offer a non-invasive method for assessing motor abnormalities associated with the neurodegenerative disorder, which progressively affects dopaminergic neurons and leads to motor and non-motor impairments.

The investigative model incorporated six spatial and image-derived features processed through Min–Max normalization and a steady-state Markov chain feature selector. Classification utilized a stacking ensemble containing a Decision Tree, a radial basis function support vector machine, and a Random Forest, alongside a Logistic Regression meta-learner. On a development set of 62 files, the stacking model demonstrated a five-fold cross-validation accuracy of 86.46% and a balanced accuracy of 84.23%. Out-of-fold pooling yielded a sensitivity of 95.92% and a specificity of 38.46%. However, ablation experiments noted that stacking did not surpass a single radial basis function support vector machine, and feature selection did not outperform using all six features.

What to know: While the findings highlight the technical viability of analyzing digitized spirals for movement disorders, the authors emphasize that the results do not prove clinical diagnostic ability. The study's small sample size and limited generalizability mean further independent datasets are required before such tools can be broadly applied. Always consult a qualified healthcare provider for personal medical advice, diagnoses, or treatment decisions.

Source Nature Health Sciences — nature.com/articles/s41598-026-80953-9 ↗

Contact the newsroom

Got a tip, a proposal or a partnership question? Write to us and we will reply shortly.