以AVL Tree探討個人健康記錄於醫學中心之存取權限分級分類管控機制

Investigating the Accessing Authority Categorized Level Management Agent of Personal Health Records in Healthcare Enterprise using AVL Tree

在資訊化的生活中，許多資料都已數位化，使得隱私保護成為一項越來越受重視的課題。什麼人可瀏覽特定人物的資料，誰又可對這些資料進行更新，都需要有明確的規範。因此要做到權限分級分類控制管理，以限制彼此間對於資料的存取權限，乃成為迫切需要解決的問題之一。而在這許多不同的情境中，由於醫院人員之職位身份非常多元化，且每種身分的存取權限各有不同，一位員工可能同時擔任兩種以上的職位，相對造成權限分級管理的複雜度，而醫院系統的使用者必須在不同身份權限之間來回切換，不僅麻煩且沒有效率，故至今還沒有高效率的機制來解決此問題。本研究就是以此為目的，首先透過改良先前以B-Tree為設計基礎之版本，改以AVL Tree為基礎的資料結構來儲存職員節點，作為權限判斷之依據。並設計出一個能指定身份權限給使用者，且於使用者進行存取時可自動判斷權限的系統，同時輔以Circular Doubly Linked List來串連相同的使用者節點，加速搜尋功能。我們將大多數使用者所可能遇到的情境，加以進行分類，建立出不同的權限集合，對應到每種情境，使系統可自動對使用者的存取動作做權限判斷。並期望這樣的架構往後可被自由套用至相關的情境下，而不侷限於醫療機構之內。

In this informational life, many data have been digitalized and made privacy protection became a more and more important issue. For example, who can browse or update specific person's data, it needs specific definition. So that to achieve the privilege level management for limiting each other's accessing privilege is one of the urgent problems that need to be solved. And in so many cases, take healthcare enterprise for instance. Due to the various identities in such environments and each of the identities have different privileges. One staff might act as more than one identity at the same time. Make the privilege management more complexity. And it still doesn't have an efficient solution to solve the problem. It is also the purpose of our research. We firstly improved the previous B-Tree-based data structure version with AVL Tree to store the staff nodes for authority determination basis. And design a system which can assign privileges to the user and can automatically judge the authority permission. Be supplemented by circular doubly linked list to connect the same user nodes in different trees and to accelerating the search function. We categorized most situations that users might encounter, and created different privilege sets that corresponding to each scenario. Let the system could automatically make permission decisions regards to users' accessing operations. We expect such system architecture can be adapted to any situation, not only in the healthcare enterprise.