Real-Time Smart Healthcare Framework Using AI, IoT, and Deep Learning for Predictive Genetic Disorder Detection

Abstract

As genetic disorders become more common, proactive and personalized healthcare solutions are needed. This study introduces a Real-Time Smart Healthcare Framework Using AI, IoT, and Deep Learning for Predictive Genetic Disorder Detection, which is based on Hybrid Deep Learning with Multimodal Data Fusion (MDF) using PyTorch. The framework combines genomic sequences, IoT sensor data, and electronic health records to provide real-time predictive analyses of patients' genetic risk factors. To capture complex interactions across modalities, a hybrid architecture that integrates CNN for genomic feature extraction, LSTM for temporal IoT data was used. Experimental results show the framework has an accuracy of 94.6%, a precision of 92.8% and a recall of 93.5%, outperforming the traditional single-modality models like CNN, LSTM, and random forest. Explainable AI techniques offer clarity in prediction, allowing clinicians to better understand what led to a heightened risk. The proposed system provides a scalable, real-time, and interpretable solution, paving the way for smart healthcare systems to become more personalized in delivering preventive care and timely intervention for genetic disorders.