The Role of Cryptography and AI in Defending against Malware Attacks in health care infrastructure

This article presents an AI- driven adaptive cryptographic framework, where the encryption-decryption algorithms are dynamically selected based on the severity index of the malware attack. The severity index is computed based on the weighted average of malware and behavioral anomaly. The severity index is used by a light weight decision model to choose a secured cryptographic algorithm automatically. In this work, based on severity index, either AES-256 or hybrid cryptosystem comprising of ECC & AES or digital signature-supported Edwards Curve based Digital Signature algorithm integrated with ECC is chosen. The parameters considered are the key generation time, time for encryption and decryption, along with the communication overhead for various payload sizes and different severity levels, Depending on severity of threat, performance of the cryptographic system is tested and results are indicated using bar charts and graphs. The experimental analysis exhibits a significant reduction in computational and communication overhead when a low-risk is encountered while stronger security mechanisms are efficient for high-risk transmissions. This achieves a practical balance between privacy, confidentiality, integrity and efficiency in real-time and resource constrained environments.