Optimization enabled ensemble based deep learning model for elderly falling risk prediction

Predicting fall risk in the elderly is crucial for enhancing safety and well-being. Aging and chronic diseases often impair balance, increasing fall risk. This study aims to develop an advanced fall risk prediction model using an optimized deep learning approach. Data undergoes pre-processing and augmentation to increase size, then is fed into an ensemble learning model,like Extreme Gradient Boosting (XGBoost), One Dimensional Convolutional Neural Network, and Deep Belief Network. The model is trained with a novel Double Exponential Lyrebird Optimization algorithm, combining double exponential smoothing and Lyrebird Optimization . Experimental results show that DELOA-based ensemble learning model achieved better results.