Benchmarking the AI-based diagnostic potential of plasma proteomics for neurodegenerative disease in 17,187 people
Bibliographic record
Abstract
Co-pathology is a common feature of neurodegenerative diseases that complicates diagnosis, treatment and clinical management. However, sensitive, specific and scalable biomarkers for in vivo pathological diagnosis are not available for most neurodegenerative neuropathologies. Here, we present ProtAIDe-Dx, a deep joint-learning model trained on 17,170 patients and controls that uses plasma proteomics to provide simultaneous probabilistic diagnosis across six conditions associated with dementia in aging. ProtAIDe-Dx achieves cross-validated balanced classification accuracy of 69%-96% and AUCs > 79% across all conditions. The model's diagnostic probabilities highlighted subgroups of patients with co-pathologies, and were associated with pathology-specific biomarkers in an external sample, even among cognitively unimpaired people. Model interpretation revealed a suite of protein networks marking shared and specific biological processes across diseases, and identified novel and previously described proteins discriminating each diagnosis. ProtAIDe-Dx significantly improved biomarker-based differential diagnosis in a memory clinic sample, pinpointing proteins leading to diagnostic decisions at an individual level. Together, this work highlights the promise of plasma proteomics to improve patient-level diagnostic work-up with a single blood draw.
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 | 0.002 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".