Blockchain-Powered Dynamic Segmentation in Personal Health Record

Abstract

Sensitive patient data must be handled securely in modern oncology care to protect patient privacy, regulatory compliance, and clinical decision-making integrity. Conventional centralized healthcare systems pose serious threats to patient confidentiality and treatment continuity because of their susceptibility to data breaches, illegal access, and single points of failure. This study suggests a blockchain-based decentralized cancer healthcare security solution. The proposed system is based on the idea of layers; each layer has a distinct function that gives it a feature that sets it apart from the others. A hierarchical clustering algorithm (HCA) is used in the first layer to determine the patient's cancer type. The second layer follows, which is mostly founded on the dynamic segmentation (DS) principle. To guarantee high security, patients are isolated on the basis of their kind in various segments. The data are then compressed via the snappy algorithm (SA) at the compression layer. To prevent hacking, the smart contracts layer then applies conditions to each of those segments. These conditions are based on the named entity recognition (NER) algorithm's decision, which is crucial in deciding whether to use blockchain technology (BCT), and ChaCha20 encryption is used to protect the confidentiality of distributed transactions in BCT. Data analysis revealed that, in comparison with earlier methods, the suggested system offers good security, and the high-performance speed of system interaction did not surpass 0.18 ms. Additionally, we were able to acquire a segmentation time of 0.25 ms, a discrimination time of 1.5 ms, and a smart contract verification time of 0.45 ms.