Healthcare Security

RESEARCH PROJECT:

HealthShare: Engage Security, Privacy, and Trust on a Healthcare Social Network

Faculty Investigator: Dr. Chunming Gao

Email: chgao@mtu.edu

INTRODUCTION

Online E-health platforms have enabled patients to obtain useful references on the Internet, as well as to share information and exchange experiences among people with similar diseases. Such social networks also enable researchers and corporates to conduct studies or advocates on targeted groups of patients. However, to make a multi-purpose healthcare social network to be practical, concerns must be addressed regarding data security, patient privacy, accountability, and promotion for participation. While it is common practice to enforce data security in computer applications, how to protect privacy and to provide accountability at the same time still demands further research. When adding the dimension of promoting user participation, the network faces even more challenges. In this research project, we treat the above concerns as multi-trust engagement. We propose a multi-trust model to enforce security, privacy, and trust by integrating public key encryption, access control, pseudonymous authentication, and mutual rating mechanism to promote multi-trust in a multi-purpose healthcare social network.

A GENERIC MULTI-TRUST MODEL

We propose to build a generic multi-trust network model for the purpose of engaging trust among all parties on the network. The model, as illustrated in Fig. 1, includes the following components: Virtual-network Control Center (VCC), Participating Organizations (POs), Service Requesters (SRs), User Interface (UI), and users. VCC and UI are the core components of the virtual social network, where all users are kept anonymous.

VCC maintains the central database for pseudonym-based accounts and trust control. The User Interface (UI) is to provide a platform for the users to activate their virtual account on VCC and register on the virtual network, as well as to retrieve information, interact with others, and take part in the activities requested by SRs. POs hold valid user information and official user data. In order to join the network and enable its valid users to participate, POs create unique numbers (UNs) and temporary pass codes for their users. POs transfer the UNs and pass codes along with trust-necessary attributes TA(a1, a2, …, ai) to VCC. TA(a1, a2, …, ai) are the essential attributes to categorize users to serve the purpose of the virtual network, but using these attributes alone cannot reveal users’ true identities. The users then use their UN and pass code to activate and register on VCC. When registering, users will create their login name and password. They may also create their pseudonym and virtual profile according to their preference. Once registered, users can publish their TAs. TA(a1, a2, …, ai) will become published attributes PA(v1, v2, …, vj), which will be revealed to public for other users to identify similar users.

Figure 1.  Multi-trust Model    

The SRs may request survey, advocates or commercial services on VCC, targeting some specific users based on users’ TA(a1, a2, …, ai), which could only be updated by POs, and categorized by VCC. VCC is responsible to provide the targeted groups to SRs and provide the options for users to select for participation. Users will not reveal their real identities by taking such activities. VCC is responsible to manage bookkeeping of users’ participation activities using UNs and pass any rewards to the users through the UN issuers, i.e., the correspondent POs.

ENGAGEMENT OF MULTI-TRUST

In the model, multi-trust is engaged in terms of data security enforcement, user privacy control, participant credibility control, and rewarding mechanism. The main concern on data security is regarding communication channels among the network components. This could be enforced through public key encryption so that the participating parties get appropriate authentication and verification when necessary. User privacy is fully enforced through levels of access control, e.g., POs need users’ permission to create UNs; users decide to activate accounts on VCCs; users decide what attributes are revealed to public by verifying TAs(a1, a2, …, ai); users have full control on CRUD (Create, Read, Update, Delete) access operations of their activities. Only POs can link real user identities through UNs. User credibility is fundamental for SRs and other users to trust the participants and related data input. This is addressed by requiring that participants must be valid users in POs, and the users’ TA(a1, a2, …, ai) could only be updated through POs. Furthermore, users are rated by peers based on their participations and their change of pseudonyms will not affect their ratings.  A reliable reward mechanism is through VCC’s bookkeeping of user participation and inverse information passing via UNs to corresponding POs. The POs will disburse the reward to the real users through real-world channels.

 

APPLICATIONS IN HEALTHCARE SOCIAL NETWORK

In an online healthcare social network, hospitals or medical institutes serve as POs. Users could include patients, nurses, and doctors. In fact, medical schools could also serve as POs. In this case, the medical students will be special users who can gain real-time medical experience on the virtual community. When hospitals and medical schools join VCC of the healthcare social network, they specify the user status attributes accordingly. A user might have dual status, e.g., the user could be a patient and a healthcare provider, or could be a patient and a medical student. The other important attributes to be revealed to VCC are the diseases the patient once experienced. Hence, from the user’s pseudonym and the attributes one can tell if he/she is a doctor/nurse/student/patient and what diseases the person has experienced. This healthcare social network could serve multiple purposes as illustrated below.

A.    A Platform for Patients to Get Second Opinion

Using VAs(v1, v2, …, vj), patients can identify others with similar diseases and share experiences with each other. They can also consult online volunteers who are specified as medical professionals or medical students. Since users are rated, patients can put informed level of trust on the second opinion of the source.

B.    A Platform for Volunteering Medical Students

An online virtual healthcare social network will provide pre-med college students and medical school students a real-time medical environment for them to gain experience and even practice consulting. Since they participate as special users, their ratings can serve as evaluation of their study and practice, which will help them to improve knowledge and enhance skills for patient treatment.

C.    A Platform for Targeted Surveys and Commercials

Healthcare researchers could request VCC for experimental studies on targeted groups of patients or on targeted groups of people with similar diseases. Although it is in a virtual environment, TAs(a1, a2, …, ai) on VCC can assure the targeted groups are valid users. Also because of the reward-enabled feature, patient participation could be reasonably assured.

SELECTED PUBLICATIONS

  1. Gao, C., Iwane, N. “Developing a Multi-trust Model for Multi-purpose Healthcare Social Networks”,  IEEE 11th Consumer Communications & Networking Conference. (2014)
  2. Gao, C., Hendrick, E., Schooley, B. “CloudHealth: Developing a Reliable Cloud Platform for Healthcare Applications”, IEEE 2013 International Workshop on Consumer eHealth Platforms, Services and Applications. (2013)