Validation of machine learning models to detect neuropathologies
Bibliographic record
Abstract
BACKGROUND: White matter hyperintensities (WMHs) are increasingly recognized for their role in cognitive decline and the progression of neurodegenerative conditions including Alzheimer's disease (AD). Despite advances in imaging technologies, the exact contribution of WMHs to disease processes remains a subject of ongoing research. This study aims to apply machine learning approaches to determine critical features of AD-related neuropathologies in vivo. METHODS: A total of 65 participants (17 females, mean age = 79.0) from the Alzheimer's Disease Neuroimaging Initiative (ADNI) were included. In the ADNI dataset, machine learning models were applied towards feature selection of MRI, clinical, and demographic data to identify the best performing set of variables that could predict neuropathology outcomes [e.g., Braak neurofibrillary tangle stage, Consortium to Establish a Registry for AD (CERAD) neuritic plaque, etc.]. The best-performing neuropathology predictors using the top seven MRI, clinical, and demographic features were selected. For continuous measures, gradient boosting, bagging, support vector regression, and linear regression were implemented. For binary outcomes, logistic regression, gradient boosting, support vector machine, and bagging classifiers were utilized. RESULTS: Four machine learning models applying feature ranking methods using similar information criteria consistently ranked WMHs as important features in predicting all neuropathology measures. In the ADNI dataset, prediction accuracy was highest for Braak stage, CERAD neuritic palques, and diffuse plaques (i.e., cross-validated correlation between actual measures and predictions was above 0.8). The best-performing model achieved over r = 0.85 correlation in predicting Braak. CONCLUSION: These results highlight the importance of WMHs as core features of AD and the benefits of using machine learning models that incorporate WMH burden in predicting AD-related neuropathologies. The use of machine learning may prove beneficial in early detection of AD.
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 | 0.025 | 0.035 |
| Meta-epidemiology (narrow) | 0.003 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.004 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.002 | 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".