Blockchain technology generally is associated with financial applications where it serves the role of maintaining records. However, such a tamper-resistant distributed ledger can be used for fashioning applications in the healthcare domain. Harnessing the potential of blockchain as a data store for health records would ensure that they would be secure as multiple copies of them are preserved in a decentralized manner to ensure data redundancy and security. With this technology in effect, the medical sector could leverage blockchain tech to allow any authorized hospitals to securely communicate, share information/records independent from a central figure. Information shared would only contain relevant data related to the query between stakeholders with appropriate permission attached to their roles. This chapter delineates the architecture of such a system explaining its benefits and limitations as a medical data-storage architecture.
TopIntroduction
In the present scenario, bulk Electronic Medical Records (EMRs) handling is dependent on the Health Information Systems (HISs) (Haux, R., 2006). These systems are centralized and provide multiple functionalities such as patient scheduling, admission/discharge/transfer logs, and billing information. The principal disadvantage of this centralized infrastructure is the lack of transparency, increased risk of data leaks, and subsequently higher repair costs (Menachemi, N., & Collum, 2011). The HIS usually handles Electronic Medical Records (EMRs) from a single hospital. It means that the patient's cumulative report from multiple hospitals and doctors is not part of the EMR. This information can be pivotal for providing detailed insight in identifying symptoms and envisaging responses for the patient. An Electronic Health Record (EHR) is, therefore, a better alternative. EHRs give a holistic, long-term view of the patient's health (Heart, T., Ben-Assuli, O., & Shabtai, I. 2017).
The word blockchain seems quite self-explanatory since one would initially think of it as a collection of blocks linked together linearly. But in the real world, blockchains can be intricate. On the ground level, it is just a database storing information. But the structure and the concepts used in this database make it so unique (Pilkington, M., 2016). With the advent of cryptocurrencies in the last decade, blockchain has gained much popularity. It forms the heart of any cryptocurrency or a decentralized system (Eyal, I. 2017). Talking about decentralization, it is the process by which activities of a system are distributed and are not in direct control of a centralized authority.
Our work proposes a Consortium Blockchain (Hasselgren, A., Kralevska, K., Gligoroski, D., Pedersen, S. A., & Faxvaag, A., 2020) to support transparent Electronic Health Records (EHRs) and enhance the disease prediction system. The proposed architecture ensures transparency while maintaining anonymity by utilizing the principle of Smart Contracts. Some key highlights of the proposal include organizing multiple hospitals under Hospital Clusters (HCs). The Hospital Clusters (HCs) aggregate multiple hospitals according to seven parameters. These parameters are specialization, location, threshold rating, average cost of treatment, capacity, fatalities, and recovery rate. The combination of these HCs forms a decentralized network that shares the consortium blockchain. A chatbot tuned according to each Hospital Cluster (HC) is then entwined with the user interface. The architecture encapsulates the following primary objectives:
- •
Channelized rule-based patient information gathering.
- •
Transparent and perspicacious Electronic Health Records maintenance.
- •
Secure storage of appropriate information on the ledger.
Communicating with customers through live chat interfaces has become an increasingly popular means to provide real-time customer service in e-commerce and medical services domains. Customers use these chat services to obtain information, seek assistance, or consult for reservations. The real-time nature of chat services has transformed customer service into two-way communication (Mero, J., 2018). The addition of a chatbot to gather user inputs can further enhance the proposed system.
At least four different structures are the basis of the functioning of a chatbot (Davenport, T., & Kalakota, R., 2019):
- •
Natural language processing: To make sense of the demands.
- •
Knowledge management: To process the relevant details and provide an answer.
- •
Deep learning helps the chatbot improve its response to each interaction.
- •
Sentiment analysis: To detect the frustration and transfer them to a human.
The idea is that the user enters their symptoms as the input in the input section of the chatbot. Then the chatbot finds the keywords from the given text and updates the state of the dialogue using the state tracking system (Noroozi, V., Zhang, Y., Bakhturina, E., & Kornuta, T., 2020). After the user has provided all their inputs, the chatbot converts the results into a natural language and gives the output back in a form that the user (human) can understand.