With the deep integration of “AI + medicine”, AI-assisted technology has been of great help to human beings in the medical field, especially in the area of predicting and diagnosing diseases based on big data, because it is faster and more accurate. However, concerns about data security seriously hinder data sharing among medical institutions. This research work introduced a novel Hybrid Fuzzy Neural Network (FNN) with Modified Deer Hunt Optimization (HMDH) based Pallier Homomorphic Encryption (PHE) scheme for enhancing the data security of the cloud from malware and attacks. Initially, collected datasets are stored in the cloud using the cloud sim tool, and collected datasets are transferred into the developed FNNHMDH-PHE framework. At first, generate the key for each dataset and separate the private key for all datasets. Moreover, convert the plain text into ciphertext using the FNN and deer fitness function in PHE. Finally, cloud-stored data are encrypted successfully and the attained performance outcomes of the developed framework are associated with other existing techniques in terms of decryption time, encryption time, efficiency, and throughput.