Non-equivalence of sub-tasks of the Rey-Osterrieth Complex Figure Test with convolutional neural networks to discriminate mild cognitive impairment
Bibliographic record
Abstract
BACKGROUND: The Rey-Osterrieth Complex Figure Test (RCFT) is a tool to evaluate cognitive function. Despite its usefulness, its scoring criteria are as complicated as its figure, leading to a low reliability. Therefore, this study aimed to determine the feasibility of using the convolutional neural network (CNN) model based on the RCFT as a screening tool for mild cognitive impairment (MCI) and investigate the non-equivalence of sub-tasks of the RCFT. METHODS: A total of 354 RCFT images (copy and recall conditions) were obtained from 103 healthy controls (HCs) and 74 patients with amnestic MCI (a-MCI). The CNN model was trained to predict MCI based on the RCFT-copy and RCFT-recall images. To evaluate the CNN model's performance, accuracy, sensitivity, specificity, and F1-score were measured. To compare discriminative power, the area under the curve (AUC) was calculated by the receiver operating characteristic (ROC) curve analysis. RESULTS: The CNN model based on the RCFT-recall was the most accurate in discriminating a-MCI (accuracy: RCFT-copy = 0.846, RCFT-recall = 0.872, MoCA-K = 0.818). Furthermore, the CNN model based on the RCFT could better discriminate MCI than the MoCA-K (AUC: RCFT-copy = 0.851, RCFT-recall = 0.88, MoCA-K = 0.848). The CNN model based on the RCFT-recall was superior to the RCFT-copy. CONCLUSION: These findings suggest the feasibility of using the CNN model based on the RCFT as a surrogate for a conventional screening tool for a-MCI and demonstrate the superiority of the CNN model based on the RCFT-recall to the RCFT-copy.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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".