Design and Development of a Context-Aware Knowledge-Based Module for Identifying Relevant Information and Information Gaps in Patients With Type 1 Diabetes Self-Collected Health Data
Bibliographic record
Abstract
BACKGROUND: Patients with diabetes use an increasing number of self-management tools in their daily life. However, health institutions rarely use the data generated by these services mainly due to (1) the lack of data reliability, and (2) medical workers spending too much time extracting relevant information from the vast amount of data produced. This work is part of the FullFlow project, which focuses on self-collected health data sharing directly between patients' tools and EHRs. OBJECTIVE: The main objective is to design and implement a prototype for extracting relevant information and documenting information gaps from self-collected health data by patients with type 1 diabetes using a context-aware approach. The module should permit (1) clinicians to assess the reliability of the data and to identify issues to discuss with their patients, and (2) patients to understand the implication their lifestyle has on their disease. METHODS: The identification of context and the design of the system relied on (1) 2 workshops in which the main author participated, 1 patient with type 1 diabetes, and 1 clinician, and (2) a co-design session involving 5 patients with type 1 diabetes and 4 clinicians including 2 endocrinologists and 2 diabetes nurses. The software implementation followed a hybrid agile and waterfall approach. The testing relied on load, and black and white box methods. RESULTS: We created a context-aware knowledge-based module able to (1) detect potential errors, and information gaps from the self-collected health data, (2) pinpoint relevant data and potential causes of noticeable medical events, and (3) recommend actions to follow to improve the reliability of the data issues and medical issues to be discussed with clinicians. The module uses a reasoning engine following a hypothesize-and-test strategy built on a knowledge base and using contextual information. The knowledge base contains hypotheses, rules, and plans we defined with the input of medical experts. We identified a large set of contextual information: emotional state (eg, preferences, mood) of patients and medical workers, their relationship, their metadata (eg, age, medical specialty), the time and location of usage of the system, patient-collected data (eg, blood glucose, basal-bolus insulin), patients' goals and medical standards (eg, insulin sensitivity factor, in range values). Demonstrating the usage of the system revealed that (1) participants perceived the system as useful and relevant for consultation, and (2) the system uses less than 30 milliseconds to treat new cases. CONCLUSIONS: Using a knowledge-based system to identify anomalies concerning the reliability of patients' self-collected health data to provide information on potential information gaps and to propose relevant medical subjects to discuss or actions to follow could ease the introduction of self-collected health data into consultation. Combining this reasoning engine and the system of the FullFlow project could improve the diagnostic process in health care.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".