MS. MONIKA SHARMA, DR. AMIT BINDAL, SEEMA SHUKLA, SWAPNIL VICHARE, ZATIN GUPTA, SWAMY TN, DR AAYUSH SHRIVASTAVA
DOI: https://doi.org/The introduction of Internet of Things (IoT) into the healthcare system has transformed the traditional healthcare system by ensuring the real-time monitoring, early detection of diseases and health information at any time. However, despite these developments, IoT-based healthcare infrastructures encounter substantial challenges regarding cyber-attacks, data incompleteness, and low throughput in wide-scale solutions. For tackling these issues, in this paper, we suggest a secure and scalable framework based on CNNs for intelligent health status classification and cyber threat detection. The model includes sensor data in a patient’s life, from three collected, representative patientMonitoring. csv, Attack. csv, and environmentMonitoring. csv. A deep CNN model is developed and optimized with methods like batch normalization, dropout regularization, and early stopping to improve model generalization and avoid overfitting. Experiments show that the model achieved high classification accuracy, low loss, and good robustness against different types of input data. The outcomes verify the feasibility of enhancing the security, adaptability, and efficiency of IoT-driven healthcare systems by using deep learning-based models.
