Incremental Health Improvement through Advanced Response-Based Treatment Recommendation (HMIPRTR): Integrating LSTM, GRU, and BFO for Enhanced Patient Care

• Sivanagaraju Vallabhuni,
• Kumar Debasis,

Vol. 19, No. 1, January 31, 2025
10.3837/tiis.2025.01.003, Download Paper (Free):

• BFO
• CNN
• GRU
• Healthcare
• IoT
• LSTM

Abstract

This paper proposes a new model, HMIPRTR, in the spirit of using treatment recommendations based on response for enhancing patient health. The proposed HMIPRTR model implements feature extraction for the disease-specific low-density datasets by incorporating LSTM and GRU. Additionally, for feature selection, a BFO method has been applied to this model with the aim of enhancing the classification accuracy in regard to various diseases. The selected features have been evaluated through the technique of multidomain analysis, which includes Frequency, Entropy, Gabor, and Wavelet. A 1D Convolutional Neural Network 1D CNN is combined with a clustered Apriori model to monitor patient responses and make optimized treatment recommendations. The model improves the accuracy of disease classification by 10.5% compared to CARE Net, 2.9% compared to PFNet, and 6.5% compared to GCN. Accuracy in recommendation increased to 6.5%, precision by 8.5%, recall to 19.5%, and response time to 13.7% compared to those models. This framework holds much promise for the generation of more accurate and timely patient care recommendations.

Statistics

Cite this article