<p>Combining traditional healthcare systems with the Internet of Things (IoT) has significantly improved healthcare service quality. However, continuous monitoring and data transmission by wearable devices and sensors in such environments often occur over unsecured open channels, making the system vulnerable to cyberattacks, especially in critical care situations. To address these concerns, a novel solution leveraging Deep Learning (DL) and Blockchain technologies is proposed for secure data sharing in IoT-enabled medical facilities. The model combines an Encoder-Elliptic Curve Deep Neural Network (EECDNN) with blockchain technology and a Quaternion Evolutionary Gravitational Neocognitron Neural Network (QEGNNnet) for intrusion detection and secure cloud-based healthcare data. The system encrypts medical information using EECDNN, with the encryption key managed securely within a blockchain system where it is segmented into blocks for increased security. Authorized users, such as patients and healthcare providers, can retrieve and decrypt data using the secret key. The model’s efficiency was tested on public datasets like ToN-IoT, CIC-IDS 2017, and CIC-IDS 2018, and was compared to other DL models. Results demonstrated substantial improvements, achieving an impressive 99.9% accuracy and a reduced error rate of 2%. The model also exhibited reduced execution time and block access time, significantly enhancing intrusion detection performance.</p>