Design and development of a secure and patient-controlled system to share healthcare data for research
Bibliographic record
Abstract
Nowadays, almost all hospitals and clinics in developed countries store patients' personal medical data in digital format taking into account appropriate security measures and in compliance with applicable legal requirements. Due to the rapid development of so-called Big Data research tools, such as artificial intelligence and machine learning, these data, if accessible, have the potential to benefit medical research in the search for new medications and improved treatment approaches. However, in Quebec as is the case in many jurisdictions, access to patient data is difficult for two main reasons. First, patient consent is required, and second, patients typically have their medical data spread across multiple source systems in multiple institutions, which makes it difficult to piece together their complete medical history. Moreover, in most contexts patients do not know who has access to their data and they cannot control access rights.Opal (opalmedapps.com) is a patient portal developed at the Research Institute of the McGill University Health Centre (RI-MUHC) that provides patients with access to some of their medical data at the MUHC. Opal's long-term roadmap calls for carefully developed infrastructure to link multiple hospitals simultaneously so that patients will be able to access their medical records that are stored in different institutions. However, the originally-designed infrastructure does not allow patients to control access to their data and contribute them for research. Therefore, in this thesis project, we explored the design and development of a new infrastructure for secure and user-controlled personal medical data sharing. Initially, we studied the architecture and workflow of the existing Opal platform. Then, we analyzed various modern decentralized tools and technologies for storing and controlling personal data. Based on the analysis and knowledge acquired, we designed and implemented a novel prototype system for controlling and sharing personal medical data with researchers using a blockchain-based infrastructure. The blockchain is a tamper-proof mechanism for storing data in an immutable way by using cryptographic and network technologies. As described in this thesis, our novel data-sharing infrastructure is designed to record (1) the permissions that each patient gives to research study personnel to access their data, (2) the data-access privileges that a "public trust" committee provides to researchers to access shared data, and (3) the data access logs of researchers who access the shared data.Thus, patients can contribute their data to a specific research study by providing electronic consent in a patient portal such as Opal and by specifying which of their data records they wish to share. The system is designed to allow patients to withdraw their consents at any time and stop further sharing of their data if they change their minds. Also, as all data access requests are automatically recorded on the blockchain, each patient has the ability to know who accessed their data and when
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.004 |
| Meta-epidemiology (narrow) | 0.000 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.006 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".