Researchers developed a machine learning model that uses ultrasound and MRI data to predict cerebral blood flow in simulated microgravity for astronaut health.
Key Details
- 1Study used 36 healthy male participants in a 90-day head-down tilt bed rest (HDTBR) to simulate microgravity.
- 2Multimodal data collected included carotid Doppler ultrasound, 3D-pCASL brain MRI, and clinical metrics.
- 3Eight machine learning models tested; CatBoost delivered the best predictive performance (AUCs up to 0.92, accuracy up to 0.84).
- 4Model features were interpreted using SHAP, identifying BMI, ICA pulsatility index, and blood flow volume as top predictors.
- 5A web application was developed for real-time CBF prediction using clinical and ultrasound data, intended for spaceflight use.
- 6Limitations include exclusion of vertebral artery flow data and a male-only cohort.
Why It Matters
This ML-driven, imaging-integrated approach advances lightweight, non-invasive cerebral blood flow monitoring in space, potentially improving astronaut neurological care and illustrating cross-modal AI application in radiology. It also underscores the growing relevance of explainable AI and real-world deployment via web tools.
Source
EurekAlert
Related News
•EurekAlert
UVA's $4M NIH Grant Backs AI to Optimize Cardiac MRI Therapies
UVA secures $4 million NIH funding to develop GenStrain, an AI model using MRI data to personalize cardiac pacing therapies.
•EurekAlert
Machine Learning Models Predict Cancer Risk with Routine Health Data
Researchers developed machine learning models using routinely collected health data to predict individual risk for 20 common cancers, outperforming traditional methods for several cancer types.
•EurekAlert
AI Deep Learning Model Enhances Echocardiogram RV Assessment and Prognosis
Researchers developed DROID-RV, a deep learning tool that automates right ventricular assessment from echocardiograms, showing strong prognostic links with cardiovascular outcomes.