P3‐059: Does Synthetic Data Oversampling in Feature Selection Improve the Classification Rate in Alzheimer's Disease?
Bibliographic record
Abstract
Large-scale trials are sampled with unequal proportions since the complex pathological conformations have its own most probability of occurrence. This produces unbalanced classes in the overall dataset leading to ill posed classification problem. Here we ask the question if an oversampling technique can increase the classification rate for an Alzheimer disease classification problem between three classes (CN 352, MCI 847, AD 131). Is well-established that biomarkers combined are more effective, in order to find the best combination of biomarkers that maximizes the separation between classes we'll use a wrapper feature selection. The dataset was provided by ADNI (Alzheimer's Disease Neuroimaging Initiative). In the training phase was applied kNN (k-Nearest Neighbors) preprocessed with decimal normalization and SMOTE (Synthetic Minority Over-sampling TechniquE). The feature selection was performed using leave-one-out-cross-validation to select the combinations of the following biomarkers: ADAS-cog (Alzheimer's Disease Assessment Scale-cognitive), Rey Auditory-Verbal Learning Test (RAVLT), Aβ1-42 (ABETA), Total Tau Protein (TAU), Phosphorylated Tau Protein (PTAU), Fludeoxyglucose (FDG), Mini Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), Florbetapir (18F-AV-45). For the validation phase was separated 20% of the original dataset, in order to compare between SMOTE and undersampling. Using SMOTE for feature selection to separate classes, we had a mean improvement of 12.01% when compared with undersampling, although feature selection using all data can amplify noise in the training phase. Other options include borderline SMOTE to improve Alzheimer classification taking into account early and late MCI.
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 | 0.002 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| 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".