The Internet of Medical Things systems involve medical data transmissions between patients, medical experts, and medical centers over public networks. The sensitivity of the medical images' contents and the personal information in the medical images required high levels of security. Chaotic maps are successfully used in image encryption due to their high security and computational efficiency. Initial random sequences generate the keys for chaotic map confusion and diffusion processes. Selection of the initial parameters is the cornerstone of the success of chaotic maps in securing digital images. In this paper, the authors proposed utilizing the novel binary Coronavirus disease optimization algorithm to determine the optimal initial sequences for the chaotic maps that lead to the generation of the optimal secret keys. The proposed algorithm selects the optimal initial keys using a hybrid fitness function. The generated optimal secret keys are then used for medical image encryption/decryption. Several medical images from different modalities are utilized for testing, and the results are compared to the latest encryption techniques according to various criteria. The experimental results ensure the robustness of the proposed algorithm to various attacks and its superior performance to similar algorithms.