Computer-Aided Screening of Autism Spectrum Disorder: Eye-Tracking Study Using Data Visualization and Deep Learning

Reducing Time to Diagnosis of Autism Spectrum Disorder Using an Integrated Community Specialty Care Model: A Retrospective Study

OBJECTIVE

To evaluate a fast-track triage model in an integrated community specialty clinic to reduce the age of diagnosis for patients with autism spectrum disorder (ASD).

STUDY DESIGN

A retrospective chart review was performed for patients seen in an integrated community specialty pediatric practice using a fast-track screening and triage model. The percentage of ASD diagnoses, age at diagnosis, and time from referral to diagnosis were evaluated. The fast-track triage model was compared with national and statewide estimates of median age of first evaluation and diagnosis.

RESULTS

From January 1, 2020, through December 31, 2021, 189 children with a mean (SD) age of 32.2 (12.4) months were screened in the integrated community specialty. Of these, 82 (43.4%) children were referred through the fast-track triage for further evaluation in the developmental and behavioral pediatrics (DBP) department, where 62 (75.6%) were given a primary diagnosis of ASD. Average wait time from referral to diagnosis using the fast-track triage model was 6 months. Mean (SD) age at diagnosis was 37.7 (13.5) months. The median age of diagnosis by the fast-track triage model was 33 months compared with the national and state median ages of diagnosis at 49 and 59 months, respectively.

CONCLUSIONS

With the known workforce shortage in fellowship-trained developmental behavioral pediatricians, the fast-track triage model is feasible and maintains quality of care while resulting in more timely diagnosis, and reducing burden on DBP by screening out cases who did not require further multidisciplinary DBP evaluation as they were appropriately managed by other areas.

Developmental Differences in Reaching-and-Placing Movement and Its Potential in Classifying Children with and without Autism Spectrum Disorder: Deep Learning Approach

Autism Spectrum Disorder (ASD) is among the most prevalent neurodevelopmental disorders, yet the current diagnostic procedures rely on behavioral analyses and interviews and lack objective screening methods. This study seeks to address this gap by integrating upper limb kinematics and deep learning methods to identify potential biomarkers that could be validated in younger age groups in the future to enhance the identification of ASD. Forty-one school-age children, with and without an ASD diagnosis (Mean age ± SE = 10.3 ± 0.4; 12 Females), participated in the study. A single Inertial Measurement Unit (IMU) was affixed to the child's wrist as they engaged in a continuous reaching and placing task. Deep learning techniques were employed to classify children with and without ASD. Our findings suggest delays in motor planning and control in school-age children compared to healthy adults. Compared to TD children, children with ASD exhibited poor motor planning and control as seen by greater number of movement units, more movement overshooting, and prolonged time to peak velocity/acceleration. Compensatory movement strategies such as greater velocity and acceleration were also seen in the ASD group. More importantly, using Multilayer Perceptron (MLP) model, we demonstrated an accuracy of ~ 78.1% in classifying children with and without ASD. These findings underscore the potential use of studying upper limb movement kinematics during goal-directed arm movements and deep learning methods as valuable tools for classifying and, consequently, aiding in the diagnosis and early identification of ASD upon further validation in younger children.

Using visual attention estimation on videos for automated prediction of autism spectrum disorder and symptom severity in preschool children

Atypical visual attention in individuals with autism spectrum disorders (ASD) has been utilised as a unique diagnosis criterion in previous research. This paper presents a novel approach to the automatic and quantitative screening of ASD as well as symptom severity prediction in preschool children. We develop a novel computational pipeline that extracts learned features from a dynamic visual stimulus to classify ASD children and predict the level of ASD-related symptoms. Experimental results demonstrate promising performance that is superior to using handcrafted features and machine learning algorithms, in terms of evaluation metrics used in diagnostic tests. Using a leave-one-out cross-validation approach, we obtained an accuracy of 94.59%, a sensitivity of 100%, a specificity of 76.47% and an area under the receiver operating characteristic curve (AUC) of 96% for ASD classification. In addition, we obtained an accuracy of 94.74%, a sensitivity of 87.50%, a specificity of 100% and an AUC of 99% for ASD symptom severity prediction.

Detection of ASD Children through Deep-Learning Application of fMRI

Autism spectrum disorder (ASD) necessitates prompt diagnostic scrutiny to enable immediate, targeted interventions. This study unveils an advanced convolutional-neural-network (CNN) algorithm that was meticulously engineered to examine resting-state functional magnetic resonance imaging (fMRI) for early ASD detection in pediatric cohorts. The CNN architecture amalgamates convolutional, pooling, batch-normalization, dropout, and fully connected layers, optimized for high-dimensional data interpretation. Rigorous preprocessing yielded 22,176 two-dimensional echo planar samples from 126 subjects (56 ASD, 70 controls) who were sourced from the Autism Brain Imaging Data Exchange (ABIDE I) repository. The model, trained on 17,740 samples across 50 epochs, demonstrated unparalleled diagnostic metrics-accuracy of 99.39%, recall of 98.80%, precision of 99.85%, and an F1 score of 99.32%-and thereby eclipsed extant computational methodologies. Feature map analyses substantiated the model's hierarchical feature extraction capabilities. This research elucidates a deep learning framework for computer-assisted ASD screening via fMRI, with transformative implications for early diagnosis and intervention.

Hybrid Techniques of Facial Feature Image Analysis for Early Detection of Autism Spectrum Disorder Based on Combined CNN Features

Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder characterized by difficulties in social communication and repetitive behaviors. The exact causes of ASD remain elusive and likely involve a combination of genetic, environmental, and neurobiological factors. Doctors often face challenges in accurately identifying ASD early due to its complex and diverse presentation. Early detection and intervention are crucial for improving outcomes for individuals with ASD. Early diagnosis allows for timely access to appropriate interventions, leading to better social and communication skills development. Artificial intelligence techniques, particularly facial feature extraction using machine learning algorithms, display promise in aiding the early detection of ASD. By analyzing facial expressions and subtle cues, AI models identify patterns associated with ASD features. This study developed various hybrid systems to diagnose facial feature images for an ASD dataset by combining convolutional neural network (CNN) features. The first approach utilized pre-trained VGG16, ResNet101, and MobileNet models. The second approach employed a hybrid technique that combined CNN models (VGG16, ResNet101, and MobileNet) with XGBoost and RF algorithms. The third strategy involved diagnosing ASD using XGBoost and an RF based on features of VGG-16-ResNet101, ResNet101-MobileNet, and VGG16-MobileNet models. Notably, the hybrid RF algorithm that utilized features from the VGG16-MobileNet models demonstrated superior performance, reached an AUC of 99.25%, an accuracy of 98.8%, a precision of 98.9%, a sensitivity of 99%, and a specificity of 99.1%.

Nonlinear features of gaze behavior during joint attention in children with autism spectrum disorder

Since children with autism spectrum disorder (ASD) might exhibit a variety of aberrant response to joint attention (RJA) behaviors, there is growing interest in identifying robust, reliable and valid eye-tracking metrics for determining differences in RJA behaviors between typically developing (TD) children and those with ASD. Previous eye-tracking studies have not been deeply investigated nonlinear features of gaze time-series during RJA. As a main motivation, this study aimed to extract three nonlinear features (i.e., complexity, long-range correlation, and local instability) of gaze time-series during RJA in children with ASD, which can be measured by fractal dimension (FD), Hurst exponent (H), and largest Lyapunov exponent (LLE), respectively. To illustrate our idea, this study adopted a publicly accessible database, including eye-tracking data collected during RJA from 19 children with ASD (7.74 ± 2.73) and 30 TD children (8.02 ± 2.89), and conducted a battery of nonparametric analysis of covariance (ANCOVA), where gender was used as covariable. Findings showed that gaze time-series during RJA in autistic children may generally have greater FD but lower H than that in TD controls. This implies that children with ASD possess more complex and unpredictable gaze behaviors during RJA than TD children. Furthermore, nonlinear metrics outperformed traditional eye-tracking metrics in obtaining higher identification performance with an accuracy of 82% and an AUC value of 0.81, distinguishing the differences between successful and failed RJA trails, and predicting the severity of ASD symptoms. Findings might bring some new insights into the understanding of the impairments in RJA behaviors for children with ASD.

A face image classification method of autistic children based on the two-phase transfer learning

Autism spectrum disorder (ASD) is a neurodevelopmental disorder, which seriously affects children's normal life. Screening potential autistic children before professional diagnose is helpful to early detection and early intervention. Autistic children have some different facial features from non-autistic children, so the potential autistic children can be screened by taking children's facial images and analyzing them with a mobile phone. The area under curve (AUC) is a more robust metrics than accuracy in evaluating the performance of a model used to carry out the two-category classification, and the AUC of the deep learning model suitable for the mobile terminal in the existing research can be further improved. Moreover, the size of an input image is large, which is not fit for a mobile phone. A deep transfer learning method is proposed in this research, which can use images with smaller size and improve the AUC of existing studies. The proposed transfer method uses the two-phase transfer learning mode and the multi-classifier integration mode. For MobileNetV2 and MobileNetV3-Large that are suitable for a mobile phone, the two-phase transfer learning mode is used to improve their classification performance, and then the multi-classifier integration mode is used to integrate them to further improve the classification performance. A multi-classifier integrating calculation method is also proposed to calculate the final classification results according to the classifying results of the participating models. The experimental results show that compared with the one-phase transfer learning, the two-phase transfer learning can significantly improve the classification performance of MobileNetV2 and MobileNetV3-Large, and the classification performance of the integrated classifier is better than that of any participating classifiers. The accuracy of the integrated classifier in this research is 90.5%, and the AUC is 96.32%, which is 3.51% greater than the AUC (92.81%) of the previous studies.

Global trends and hotspots in the digital therapeutics of autism spectrum disorders: a bibliometric analysis from 2002 to 2022

Introduction

Autism spectrum disorder (ASD) is a severe neurodevelopmental disorder that has become a major cause of disability in children. Digital therapeutics (DTx) delivers evidence-based therapeutic interventions to patients that are driven by software to prevent, manage, or treat a medical disorder or disease. This study objectively analyzed the current research status of global DTx in ASD from 2002 to 2022, aiming to explore the current global research status and trends in the field.

Methods

The Web of Science database was searched for articles about DTx in ASD from January 2002 to October 2022. CiteSpace was used to analyze the co-occurrence of keywords in literature, partnerships between authors, institutions, and countries, the sudden occurrence of keywords, clustering of keywords over time, and analysis of references, cited authors, and cited journals.

Results

A total of 509 articles were included. The most productive country and institution were the United States and Vanderbilt University. The largest contributing authors were Warren, Zachary, and Sarkar, Nilanjan. The most-cited journal was the Journal of Autism and Developmental Disorders. The most-cited and co-cited articles were Brian Scarselati (Robots for Use in Autism Research, 2012) and Ralph Adolphs (Abnormal processing of social information from faces in autism, 2001). "Artificial Intelligence," "machine learning," "Virtual Reality," and "eye tracking" were common new and cutting-edge trends in research on DTx in ASD.

Discussion

The use of DTx in ASD is developing rapidly and gaining the attention of researchers worldwide. The publications in this field have increased year by year, mainly concentrated in the developed countries, especially in the United States. Both Vanderbilt University and Yale University are very important institutions in the field. The researcher from Vanderbilt University, Warren and Zachary, his dynamics or achievements in the field is also more worth our attention. The application of new technologies such as virtual reality, machine learning, and eye-tracking in this field has driven the development of DTx on ASD and is currently a popular research topic. More cross-regional and cross-disciplinary collaborations are recommended to advance the development and availability of DTx.

Rendimiento de los estímulos visuales en el diagnóstico del TEA por Eye Tracking: Revisión Sistemática

El eye-tracking es una herramienta diagnóstica que tiene como   fin el estudio del comportamiento de la mirada a través del escaneo de ojos para observar el seguimiento ocular, cómo se distribuye la mirada y la precisión de los movimientos oculares. Este sistema se ha utilizado con niños/as del Trastorno del Espectro Autista. El objetivo de esta revisión sistemática ha sido analizar el rendimiento de los estímulos visuales en el diagnóstico del TEA por método eye tracking. Para ello, se siguió la metodología PRISMA, realizando una búsqueda en las bases de datos PubMed, Science Direct y Scopus, así como, Reseach Gate. Se seleccionaron 22 artículos que cumplían los criterios de inclusión con experimentos unifactoriales, experimentales factoriales y cuasiexperimentales. Todos los experimentos han tenido un grupo control compuesto de muestra con participantes con desarrollo normotípico y de un grupo de caso compuesto de muestra con participantes TEA. Esta revisión sintetiza en tres categorías en base a las características del estímulo usado en el diagnóstico (estímulos sociales, no sociales y por confrontación), el análisis del rendimiento de los estímulos visuales, de manera que los estímulos sociales y los estímulos por confrontación van a ser eficaces para establecer un diagnóstico preciso de TEA puesto que permiten realizar un cribado de ambos grupos y establecer un riesgo temprano del trastorno.

Eye-movement analysis on facial expression for identifying children and adults with neurodevelopmental disorders

Experienced psychiatrists identify people with autism spectrum disorder (ASD) and schizophrenia (Sz) through interviews based on diagnostic criteria, their responses, and various neuropsychological tests. To improve the clinical diagnosis of neurodevelopmental disorders such as ASD and Sz, the discovery of disorder-specific biomarkers and behavioral indicators with sufficient sensitivity is important. In recent years, studies have been conducted using machine learning to make more accurate predictions. Among various indicators, eye movement, which can be easily obtained, has attracted much attention and various studies have been conducted for ASD and Sz. Eye movement specificity during facial expression recognition has been studied extensively in the past, but modeling taking into account differences in specificity among facial expressions has not been conducted. In this paper, we propose a method to detect ASD or Sz from eye movement during the Facial Emotion Identification Test (FEIT) while considering differences in eye movement due to the facial expressions presented. We also confirm that weighting using the differences improves classification accuracy. Our data set sample consisted of 15 adults with ASD and Sz, 16 controls, and 15 children with ASD and 17 controls. Random forest was used to weight each test and classify the participants as control, ASD, or Sz. The most successful approach used heat maps and convolutional neural networks (CNN) for eye retention. This method classified Sz in adults with 64.5% accuracy, ASD in adults with up to 71.0% accuracy, and ASD in children with 66.7% accuracy. Classifying of ASD result was significantly different (p<.05) by the binomial test with chance rate. The results show a 10% and 16.7% improvement in accuracy, respectively, compared to a model that does not take facial expressions into account. In ASD, this indicates that modeling is effective, which weights the output of each image.

Machine learning based on eye-tracking data to identify Autism Spectrum Disorder: A systematic review and meta-analysis

BACKGROUND

Machine learning has been widely used to identify Autism Spectrum Disorder (ASD) based on eye-tracking, but its accuracy is uncertain. We aimed to summarize the available evidence on the performances of machine learning algorithms in classifying ASD and typically developing (TD) individuals based on eye-tracking data.

METHODS

We searched Medline, Embase, Web of Science, Scopus, Cochrane Library, IEEE Xplore Digital Library, Wan Fang Database, China National Knowledge Infrastructure, Chinese BioMedical Literature Database, VIP Database for Chinese Technical Periodicals, from database inception to December 24, 2021. Studies using machine learning methods to classify ASD and TD individuals based on eye-tracking technologies were included. We extracted the data on study population, model performances, algorithms of machine learning, and paradigms of eye-tracking. This study is registered with PROSPERO, CRD42022296037.

RESULTS

261 articles were identified, of which 24 studies with sample sizes ranging from 28 to 141 were included (n = 1396 individuals). Machine learning based on eye-tracking yielded the pooled classified accuracy of 81 % (I² = 73 %), specificity of 79 % (I² = 61 %), and sensitivity of 84 % (I² = 61 %) in classifying ASD and TD individuals. In subgroup analysis, the accuracy was 88 % (95 % CI: 85-91 %), 79 % (95 % CI: 72-84 %), 71 % (95 % CI: 59-91 %) for preschool-aged, school-aged, and adolescent-adult group. Eye-tracking stimuli and machine learning algorithms varied widely across studies, with social, static, and active stimuli and Support Vector Machine and Random Forest most commonly reported. Regarding the model performance evaluation, 15 studies reported their final results on validation datasets, four based on testing datasets, and five did not report whether they used validation datasets. Most studies failed to report the information on eye-tracking hardware and the implementation process.

CONCLUSION

Using eye-tracking data, machine learning has shown potential in identifying ASD individuals with high accuracy, especially in preschool-aged children. However, the heterogeneity between studies, the absence of test set-based performance evaluations, the small sample size, and the non-standardized implementation of eye-tracking might deteriorate the reliability of results. Further well-designed and well-executed studies with comprehensive and transparent reporting are needed to determine the optimal eye-tracking paradigms and machine learning algorithms.

Autism screening at 18 months of age: a comparison of the Q-CHAT-10 and M-CHAT screeners

Background

Autism screening is recommended at 18- and 24-month pediatric well visits. The Modified Checklist for Autism in Toddlers—Revised (M-CHAT-R) authors recommend a follow-up interview (M-CHAT-R/F) when positive. M-CHAT-R/F may be less accurate for 18-month-olds than 24-month-olds and accuracy for identification prior to two years is not known in samples that include children screening negative. Since autism symptoms may emerge gradually, ordinally scoring items based on the full range of response options, such as in the 10-item version of the Quantitative Checklist for Autism in Toddlers (Q-CHAT-10), might better capture autism signs than the dichotomous (i.e., yes/no) items in M-CHAT-R or the pass/fail scoring of Q-CHAT-10 items. The aims of this study were to determine and compare the accuracy of the M-CHAT-R/F and the Q-CHAT-10 and to describe the accuracy of the ordinally scored Q-CHAT-10 (Q-CHAT-10-O) for predicting autism in a sample of children who were screened at 18 months.

Methods

This is a community pediatrics validation study with screen positive (n = 167) and age- and practice-matched screen negative children (n = 241) recruited for diagnostic evaluations completed prior to 2 years old.

Clinical diagnosis of autism was based on results of in-person diagnostic autism evaluations by research reliable testers blind to screening results and using the Autism Diagnostic Observation Schedule—Second Edition (ADOS-2) Toddler Module and Mullen Scales of Early Learning (MSEL) per standard guidelines.

Results

While the M-CHAT-R/F had higher specificity and PPV compared to M-CHAT-R, Q-CHAT-10-O showed higher sensitivity than M-CHAT-R/F and Q-CHAT-10.

Limitations

Many parents declined participation and the sample is over-represented by higher educated parents. Results cannot be extended to older ages.

Conclusions

Limitations of the currently recommended two-stage M-CHAT-R/F at the 18-month visit include low sensitivity with minimal balancing benefit of improved PPV from the follow-up interview. Ordinal, rather than dichotomous, scoring of autism screening items appears to be beneficial at this age. The Q-CHAT-10-O with ordinal scoring shows advantages to M-CHAT-R/F with half the number of items, no requirement for a follow-up interview, and improved sensitivity. Yet, Q-CHAT-10-O sensitivity is less than M-CHAT-R (without follow-up) and specificity is less than the two-stage procedure. Such limitations are consistent with recognition that screening needs to recur beyond this age.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13229-021-00480-4.