A Multi-Modal Explainable Deep Learning Model for Rheumatoid Arthritis Diagnosis Using Multi-Omics and Clinical Data
Bibliographic record
Abstract
Rheumatoid arthritis (RA) is a long-term autoimmune syndrome that causes inflammation and tenderness in the joints. Because of the wide range of progressing symptoms and patient diversity, rheumatoid arthritis therapy necessitates thorough examination of available remedies. Rheumatic arthritis factors- anti-cyclic citrullinated peptide (anti-CCP) antibodies, erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) are systematically indicators that have historically been used in the confirmation of RA, along with medical tests such as stiffness during the morning and swelling around the joints. Advanced technologies such as artificial intelligence, machine learning as well deep learning provide the possibility of reducing delays in diagnosis and enhancing treatment outcomes in the case of rheumatoid arthritis, a chronic inflammatory disease that necessitates immediate treatment. This study suggests a unique multi-modal attention based ensemble (MAE) model for earliest RA detection which incorporates imaging information, clinical markers, transcriptomics, and metabolomics analysis. This research builds a reliable framework with attention mechanisms and SHAP-based interpretation using huge freely accessible datasets, including Metabolomics Workbench- MT BLS3744,$\mathbf{n} \boldsymbol{=} \mathbf{2 2 0}$, GEO- GSE93272,$\mathbf{n} \boldsymbol{=} \mathbf{2 3 2}$, and OAI clinical data, where$\mathbf{n} \boldsymbol{=} \mathbf{3 0 0}$. According to initial analyses, the suggested MAE model outperforms current EBM as well as LightGBM techniques in terms of accuracy such as$\text{AUC}=0943$, sensitivity =0.921, specificity =0.89, Brier Score$=0.096$. The finding clears the path for the use of comprehensible, multi-modal AI methodologies in precise rheumatology and show that they are feasible to implement in RA detection.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 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 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".