RDBU| Repositório Digital da Biblioteca da Unisinos
OmniPHR : a Blockchain based interoperable architecture for personal health records
- RDBU
- →
- Rede Jesuíta Brasil
- →
- Unisinos
- →
- Produção Acadêmica
- →
- Escola Politécnica
- →
- Teses e dissertações
- →
- Doutorado
- →
- PPG Computação Aplicada
- →
- View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.
OmniPHR : a Blockchain based interoperable architecture for personal health records
Roehrs, Alex
Date:
2019-08-16
xmlui.dri2xhtml.METS-1.0.item-contributorAdvisor:
Costa, Cristiano André da
Abstract:
The advances in the Health Information Technology (HIT) brought many benefits to the health care area, especially to the digital storage of patients’ health records. However, it is still a challenge to have a unified viewpoint of patients’ health history, because typically, health data is scattered among different health organizations. Furthermore, there are several standards for these records, some of them open and others proprietary. Usually, health records are stored in databases within health organizations and generally do not have external access. This situation applies mainly to cases where health care providers maintain patients’
data, known as EHR (Electronic Health Record). In the case of PHR (Personal Health Record), in which patients by definition can manage their health records, they usually have no control over their data stored in health care providers’ databases. Even with adopted standards, patients often need to explain over and over their health information when they are taken care at different locations. This problem hinders the adoption of PHR. OBJECTIVE: Thereby, we envision two main challenges regarding PHR context: first, how patients could have a unified view of their scattered health records, and second, how health care providers can access up-to-date data regarding their patients, even though changes occurred elsewhere. The scientific contribution is to propose an architectural model based on Blockchain to support a distributed PHR, where patients can maintain their health history in a unified viewpoint, from any device anywhere. Likewise, the scientific contribution for health care providers seeks to promote the possibility of having their patients’ data interconnected among health organizations. The methodology consists in proposing and prototyping an application model named OmniPHR (’Omni’ comes from omnipresent) as a distributed model to integrate PHRs. The method to evaluate the model includes assessing the network performance, interoperability, and semantic integration of different health standards, using a real database from anonymized patients. The evaluations demonstrate the feasibility of the model in maintaining health records distributed in an architecture model that promotes a unified view of PHR with the scalability of the solution. As a result, we evaluated the health data processed in different standards, represented by openEHR and HL7/FHIR. OmniPHR demonstrated the feasibility to provide semantic interoperability through a standard ontology and machine learning with NLP (Natural Language Processing). Although 12% of health records still required manual intervention in conversion, we present a way to obtain the original data from different standards on a single format. We evaluated our model implementation using the data set of more than 40,000 adult patients anonymized from two hospital databases. We tested the distribution and reintegration of the data to compose a single view of health records. Moreover, we profiled the model by evaluating a scenario with ten superpeers and thousands of concurrent sessions transacting operations on health records simultaneously, resulting in an average response time below 500 ms. The Blockchain implemented in our prototype achieved 98% availability. As contribution, OmniPHR presents a unified, semantic, and up-to-date vision of PHR for patients and health providers. The results were promising and demonstrated the possibility of subsidizing the creation of inferences rules about possible patients’ health problems and preventing future problems.
Files in this item
This item appears in the following Collection(s)
-
PPG Computação Aplicada [30]
PPG Computação Aplicada