Digital Health

Expertise: Digital Health

Machine Learning to Improve the Accuracy of a Period Prediction App

Key Details Increased period prediction accuracy up to 15%. Challenge Improve period prediction accuracy for application users Solution Linear and tree-based models to make better predictions of the menstrual cycle Technologies and tools Machine Learning: regression and gradient boosting models Scientific (Predictive Analytics Python stack): Python, NumPy, scikit-learn, LightGBM, XGBoost; Web-application: hug (web-framework), Gunicorn (web-server) … Read more

Neural Network Implementation for Better Cycle Predictions. Success Story of Flo

Key Details Improved accuracy of cycle predictions by up to 54.2%. Challenge Prediction of irregular women cycles Solution Integration of neural networks for menstrual cycle prediction Technologies and tools Python Client Flo is a smart period tracker that accurately predicts women’s menstrual cycles, ovulation and fertile days. Challenge: improve predictions of irregular cycles for application … Read more

Pose Estimation for Fitness and Physical Therapy Application

Key Details Improved pose estimation and error detection by 64%. Challenge Develop a state-of-the-art pose estimation model to detect a human posture in a real-time scenario and perform error analysis and repetitions counting Solution Deep learning for accurate human pose estimation and data science algorithms for error detection Technologies and tools PyTorch, CoreML, TFLite, OpenCV, … Read more