Risk factors and drug discovery for cognitive impairment in type 2 diabetes mellitus using artificial intelligence interpretation and graph neural networks
Bibliographic record
Abstract
Background: Among the 382 million diabetic patients worldwide, approximately 30% experience neuropathy, and one-fifth of these patients eventually develop diabetes cognitive impairment (CI). However, the mechanism underlying diabetes CI remains unknown, and early diagnostic methods or effective treatments are currently not available. Objective: This study aimed to explore the risk factors for CI in patients with type 2 diabetes mellitus (T2DM), screen potential therapeutic drugs for T2DM-CI, and provide evidence for preventing and treating T2DM-CI. Methods: This study focused on the T2DM population admitted to the First Affiliated Hospital of Hunan College of Traditional Chinese Medicine and the First Affiliated Hospital of Hunan University of Chinese Medicine. Sociodemographic data and clinical objective indicators of T2DM patients admitted from January 2018 to December 2022 were collected. Based on the Montreal Cognitive Assessment (MoCA) Scale scores, 719 patients were categorized into two groups, the T2DM-CI group with CI and the T2DM-N group with normal cognition. The survey content included demographic characteristics, laboratory serological indicators, complications, and medication information. Six machine learning algorithms were used to analyze the risk factors of T2DM-CI, and the Shapley method was used to enhance model interpretability. Furthermore, we developed a graph neural network (GNN) model to identify potential drugs associated with T2DM-CI. Results: Our results showed that the T2DM-CI risk prediction model based on Catboost exhibited superior performance with an area under the receiver operating characteristic curve (AUC) of 0.95 (specificity of 93.17% and sensitivity of 78.58%). Diabetes duration, age, education level, aspartate aminotransferase (AST), drinking, and intestinal flora were identified as risk factors for T2DM-CI. The top 10 potential drugs related to T2DM-CI, including Metformin, Liraglutide, and Lixisenatide, were selected by the GNN model. Some herbs, such as licorice and cuscutae semen, were also included. Finally, we discovered the mechanism of herbal medicine interventions in gut microbiota. Conclusion: The method based on Interpreting AI and GNN can identify the risk factors and potential drugs associated with T2DM-CI.
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 | 0.000 | 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 | 0.000 |
| 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".