Background Type 2 diabetes mellitus (T2DM) is one of the most pressing global public health challenges. Machine learning (ML) combined with Explainable Artificial Intelligence (XAI) is increasingly applied to T2DM risk prediction, yet, to the best of our knowledge, no systematic quantitative overview of this landscape has been conducted.

Methods We present a bibliometric analysis of 1,933 Scopus documents (collected 29 March 2026), employing keyword analysis, thematic clustering, and publication trend analysis.

Results The field grew substantially — from 35 documents (2020) to 828 documents (2025). SHAP (SHapley Additive exPlanations) and LIME are the most prevalent XAI methods; XGBoost and Random Forest dominate the ML landscape. KG/GNN (Knowledge Graph/Graph Neural Network) gap: only 7 of 1,933 documents (∼0.36%) — a 71:1 disparity versus the XAI group. A selective review of 15 highly cited papers confirmed that 0/15 studies combine all three components ML + XAI + KG in T2DM risk prediction.

Contribution This study identifies the clinical interpretability gap — where statistical explanations from SHAP/LIME are not connected to structured clinical pathways — as the central challenge of XAI in T2DM risk prediction, and proposes a three-layer conceptual framework integrating KG as a semantic layer to address it. The framework has direct implications for clinical decision support systems (CDSS) and population-level public health screening, where explainability aligned with medical reasoning is essential for responsible deployment. The study provides a transferable methodology and quantified research gap analysis to guide future work combining machine learning, XAI, and structured medical knowledge.

The authors have declared no competing interest.

The author(s) received no specific funding for this work.

I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.

Yes

The details of the IRB/oversight body that provided approval or exemption for the research described are given below:

No IRB approval was required. This study is a bibliometric analysis of publicly available publication metadata retrieved from Scopus and does not involve human participants, identifiable data, specimens, or tissue.

I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.

Yes

I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).

Yes

I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.

Yes