The largest source of information in healthcare during the present epidemic is radiological imaging, which is also one of the most difficult sources to interpret. Clinicians today are forced to rely heavily on therapeutic image analysis that has been filtered and sometimes performed by worn-out radiologists. Transmission of these medical data increases in frequency due to patient overflow, and protecting confidentiality, along with integrity and availability, emerges as one of the most crucial components of security. Medical images generally contain sensitive information about patients and are therefore vulnerable to various security threats during transmission over public networks. These images must be protected before being transmitted over this network to the public. In this paper, an efficient SCAN and chaotic-map-based image encryption model is proposed. This paper describes pixel value and pixel position manipulation based on SCAN and chaotic theory. The SCAN method involves translating an image’s pixel value to a different pixel value and rearranging pixels in a predetermined order. A chaotic map is used to shift the positions of the pixels within the block. Decryption follows the reverse process of encryption. The effectiveness of the suggested strategy is evaluated by computing the histogram chi-square test, MSE, PSNR, NPCR, UACI, SSIM, and UQI. The efficiency of the suggested strategy is demonstrated by comparison analysis. The results of analysis and testing show that the proposed program can achieve the concept of partial encryption. In addition, simulation experiments demonstrate that our approach has both a faster encryption speed and higher security when compared to existing techniques.

Electroencephalography (EEG) is a fundamental technique for epileptic seizure detection due to its ability to capture brain activity. Symmetry and asymmetry in EEG signals serve as indicators of seizure events, as normal EEGs exhibit bilateral patterns that may become asymmetric during seizures. The paper discusses applying symmetry/asymmetry analysis in the diagnostic process for epilepsy. To improve how accurate detection can be, Improved Feature Space Method (ICFS) with DWT extracts features from the datasets in time, frequency and entropy domains. The features are processed by various Support Vector Machine (SVM) models during ensemble training. Tests on standard EEG datasets found that our method achieved 97% accuracy during validation and outperformed both traditional correlation methods and advanced seizure detection approaches. The study reveals the model works effectively and can accurately detect seizures.