DBFEH: Design of a Deep-Learning-Based Bioinspired Model to Improve Feature Extraction Capabilities of Healthcare Device Sets

Abstract

Extraction of features from healthcare devices requires the efficient deployment of high-sample rate components. A wide variety of deep learning models are proposed for this task, and each of these models showcases non-uniform performance & complexity levels when applied to real-time scenarios. To overcome such issues, this text proposes the design of a deep-learning-based bioinspired model that assists in improving the feature extraction capabilities of healthcare device sets. The model uses Elephant Herding Optimization (EHO) for real-time control of data rate & feature-extraction process. The collected samples are processed via an Augmented 1D Convolutional Neural Network (AOD CNN) model, which assists in the identification of different healthcare conditions. The accuracy of the proposed AOD CNN is optimized via the same EHO process via iterative learning operations. Due to adaptive data-rate control, the model is capable of performing temporal learning for the identification of multiple disease progressions. These progressions are also evaluated via the 1D CNN model, which can be tuned for heterogeneous disease types. Due to the integration of these methods, the proposed model is able to improve classification accuracy by 2.5%, while reducing the delay needed for data collection by 8.3%, with an improvement in temporal disease detection accuracy of 5.4% when compared with standard feature classification techniques. This assists in deploying the model for a wide variety of clinical scenarios.