A Video-Based Autism Detection In Children Using MLP Classifier and Computer Vision Technique
Bibliographic record
Abstract
This article examines ways machine learning and computer vision can be used to detect the symptoms of autism in children at an early stage.The moment of diagnosis is important, both to the healthcare and to the development of the child, hence automating it makes a lot of sense.In this case, MediaPipe is used to extract information concerning the facial expression and the body movements.OpenCV takes care of face capture of uploaded videos as well as live video streams.After acquiring these features, we save them in the form of CSV files and then subject them to an MLP classifier, so as to determine whether the videos depict autistic or non autistic behavior. The results of the MLP classifier are high accuracy after number crunching. It can be used with both pre recorded videos and live footage and the system even records the live results automatically that can be reviewed later.Experiments indicate that MediaPipes feature extraction with deep learning with MLP would actually be effective.This would make the process a viable tool of early screening and behavioral evaluation. To clinicians, this would save time on manual observation and a higher chance of detecting autism in its early stages, thus children would obtain help earlier. On balance, this study advances the automated autism detection, which will combine computer vision and machine learning into the tangible healthcare solutions.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 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 | 0.000 |
| 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".