Mobile detection of autism through machine learning on home video: A development and prospective validation study

Abstract

BackgroundThe standard approaches to diagnosing autism spectrum disorder (ASD) evaluate between 20 and 100 behaviors and take several hours to complete. This has in part contributed to long wait times for a diagnosis and subsequent delays in access to therapy. We hypothesize that the use of machine learning analysis on home video can speed the diagnosis without compromising accuracy. We have analyzed item-level records from 2 standard diagnostic instruments to construct machine learning classifiers optimized for sparsity, interpretability, and accuracy. In the present study, we prospectively test whether the features from these optimized models can be extracted by blinded nonexpert raters from 3-minute home videos of children with and without ASD to arrive at a rapid and accurate machine learning autism classification.Methods and findingsWe created a mobile web portal for video raters to assess 30 behavioral features (e.g., eye contact, social smile) that are used by 8 independent machine learning models for identifying ASD, each with >94% accuracy in cross-validation testing and subsequent independent validation from previous work. We then collected 116 short home videos of children with autism (mean age = 4 years 10 months, SD = 2 years 3 months) and 46 videos of typically developing children (mean age = 2 years 11 months, SD = 1 year 2 months). Three raters blind to the diagnosis independently measured each of the 30 features from the 8 models, with a median time to completion of 4 minutes. Although several models (consisting of alternating decision trees, support vector machine [SVM], logistic regression (LR), radial kernel, and linear SVM) performed well, a sparse 5-feature LR classifier (LR5) yielded the highest accuracy (area under the curve [AUC]: 92% [95% CI 88%–97%]) across all ages tested. We used a prospectively collected independent validation set of 66 videos (33 ASD and 33 non-ASD) and 3 independent rater measurements to validate the outcome, achieving lower but comparable accuracy (AUC: 89% [95% CI 81%–95%]). Finally, we applied LR to the 162-video-feature matrix to construct an 8-feature model, which achieved 0.93 AUC (95% CI 0.90–0.97) on the held-out test set and 0.86 on the validation set of 66 videos. Validation on children with an existing diagnosis limited the ability to generalize the performance to undiagnosed populations.ConclusionsThese results support the hypothesis that feature tagging of home videos for machine learning classification of autism can yield accurate outcomes in short time frames, using mobile devices. Further work will be needed to confirm that this approach can accelerate autism diagnosis at scale.

Identifying predictive features of autism spectrum disorders in a clinical sample of adolescents and adults using machine learning

“…all subtasks) and consequently the time saved by coding only a smaller number of behavioral features is marginal. In contrast to this view, however, recent findings have shown that ASD classification of children can be achieved by applying the ADOS items to shorter, unstructured social interactions [54][55][56] and even by relying solely on written extracts of children´s medical and educational records 57,58 , thereby suggesting that a reduction in time associated with ASD detection might be feasible. For example, Fusaro and colleagues 54 have assessed the feasibility of applying all of the ADOS Module 1 codes, but not the behavior observation exam, to short (<10 minutes) and unstructured home-videos collected from YouTube.…”

“…These findings suggest that short, unstructured interactions can provide sufficient information to rate ADOS codes and to detect ASD. Building upon this work, Tariq and colleagues 55 recently investigated how the reduced feature subsets from the ADOS modules 1 to 3 that were identified in previous machine learning experiments 38,41,42 could be translated into clinical practice. For this purpose, they created a mobile web portal and asked video raters to assess the previously identified minimal feature subsets in short home videos (<5 minutes) of children with and without ASD.…”

Exploration in the Presence of Mother in Typically and Non-typically Developing Pre-walking Human Infants

“…Despite these limitations, we have, strictly speaking, found a candidate referral tool for child development specialists examining young pre-walking infants. Once having proved the replicability of this prodrome, machine learning classification using the relevant features could be extracted from videos of infants taken in an unforeseen natural environment, alerting for ASD risk in minutes (Tariq et al, 2018(Tariq et al, , 2019.…”

The Use of Artificial Intelligence in Screening and Diagnosis of Autism Spectrum Disorder: A Literature Review

“…Third, there remains a lack of clear understanding of the technique that is being used. A number of studies used multiple algorithm approaches and report on the highest predictive value [32,33,36,37,43,44,48,49]. Before arguing on the best algorithm to use, it would be important to understand why there are such differences in the results and the reason as to which approach would be most appropriate depending on the characteristics of the dataset and what sort of an output one is trying to achieve.…”