<p>The healthcare industry has witnessed significant advancements with the adoption of smart wearable devices, enabling remote patient monitoring and treatment. However, these technologies introduce critical security risks, including session hijacking, data manipulation, and spoofing attacks, which can compromise patient privacy and safety. This study proposes a secure framework integrating a ridgelet neural network optimized with the addax optimization algorithm (AOA) to detect and mitigate threats in wearable healthcare systems. The proposed three-layered framework consists of (1) an analytics layer, which utilizes a ridgelet neural network for classifying wearable device data as malicious or non-malicious, (2) a blockchain layer, ensuring the integrity and secure storage of verified patient data, and (3) a user layer, facilitating authorized access to healthcare providers. The framework demonstrates superior performance in detecting malicious data, achieving an accuracy ranging from 98.5 to 99.5%. The blockchain layer ensures data transparency, immutability, and security, significantly reducing the risks of cyber threats. The proposed approach enhances the security of wearable healthcare technology by effectively classifying threats and safeguarding patient data, offering a trusted and resilient health data management solution.</p>