Explainable classification of autism in children with a convolutional neural network
Bibliographic record
Abstract
Autism is a complex neurodevelopmental condition that influences how individuals interact, communicate, and behave. Although its prevalence is high, it can be challenging to make an early diagnosis, mainly due to the gradual onset patterns of its symptoms. Artificial intelligence (AI) models using magnetic resonance imaging (MRI) can support the diagnosis of autism in children by detecting complex disease-related brain patterns that are not obvious to human experts. The purpose of this study was to develop and evaluate an explainable deep learning (DL) model to support the diagnosis of autism in children and to identify the most important brain regions for the classification task. For the development and evaluation of the proposed DL model, we used 452 T1-weighted structural magnetic resonance images of individuals aged 9 to 11 years from the Autism Brain Imaging Data Exchange I and II (ABIDE I and II) databases. Using this data sample, a convolutional neural network was trained to classify neurologically typical children and autistic children (360 used for training / 46 images for validation / 46 images for testing). The results based on the images used as the test set showed that the proposed deep learning method achieves an overall accuracy of 71.74%, with a sensitivity of 73.91% and a specificity of 70.83%. The corresponding saliency voxel attribution maps were computed, which hihglighted the left transverse temporal gyrus, the left lateral ventricle, the left VI-VII vermal lobules, and the left thalamus as the most important regions for the classification task. These brain regions are consistent with previous studies that identified differences in these areas in autistic individuals. To our knowledge, this is the first study that aims to classify autism in children aged 9-11 years using a deep learning approach based on structural MRI data in combination with artificial intelligence explainability techniques to identify relevant brain regions for this task.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".