Emotion recognition across visual and auditory modalities in autism spectrum disorder: A systematic review and meta-analysis

2022 ◽  
Vol 63 ◽  
pp. 101000
Author(s):  
Florence Yik Nam Leung ◽  
Jacqueline Sin ◽  
Caitlin Dawson ◽  
Jia Hoong Ong ◽  
Chen Zhao ◽  
...  
2019 ◽  
Author(s):  
Sun Jae Moon ◽  
Jin Seub Hwang ◽  
Rajesh Kana ◽  
John Torous ◽  
Jung Won Kim

BACKGROUND Over the recent years, machine learning algorithms have been more widely and increasingly applied in biomedical fields. In particular, its application has been drawing more attention in the field of psychiatry, for instance, as diagnostic tests/tools for autism spectrum disorder. However, given its complexity and potential clinical implications, there is ongoing need for further research on its accuracy. OBJECTIVE The current study aims to summarize the evidence for the accuracy of use of machine learning algorithms in diagnosing autism spectrum disorder (ASD) through systematic review and meta-analysis. METHODS MEDLINE, Embase, CINAHL Complete (with OpenDissertations), PsyINFO and IEEE Xplore Digital Library databases were searched on November 28th, 2018. Studies, which used a machine learning algorithm partially or fully in classifying ASD from controls and provided accuracy measures, were included in our analysis. Bivariate random effects model was applied to the pooled data in meta-analysis. Subgroup analysis was used to investigate and resolve the source of heterogeneity between studies. True-positive, false-positive, false negative and true-negative values from individual studies were used to calculate the pooled sensitivity and specificity values, draw SROC curves, and obtain area under the curve (AUC) and partial AUC. RESULTS A total of 43 studies were included for the final analysis, of which meta-analysis was performed on 40 studies (53 samples with 12,128 participants). A structural MRI subgroup meta-analysis (12 samples with 1,776 participants) showed the sensitivity at 0.83 (95% CI-0.76 to 0.89), specificity at 0.84 (95% CI -0.74 to 0.91), and AUC/pAUC at 0.90/0.83. An fMRI/deep neural network (DNN) subgroup meta-analysis (five samples with 1,345 participants) showed the sensitivity at 0.69 (95% CI- 0.62 to 0.75), the specificity at 0.66 (95% CI -0.61 to 0.70), and AUC/pAUC at 0.71/0.67. CONCLUSIONS Machine learning algorithms that used structural MRI features in diagnosis of ASD were shown to have accuracy that is similar to currently used diagnostic tools.


2021 ◽  
pp. 116856
Author(s):  
Frédéric Dutheil ◽  
Aurélie Comptour ◽  
Roxane Morlon ◽  
Martial Mermillod ◽  
Bruno Pereira ◽  
...  

2021 ◽  
Vol 11 (6) ◽  
pp. 488
Author(s):  
Daniel A Rossignol ◽  
Richard E Frye

Autism spectrum disorder (ASD) is a neurodevelopmental disorder affecting approximately 2% of children in the United States. Growing evidence suggests that immune dysregulation is associated with ASD. One immunomodulatory treatment that has been studied in ASD is intravenous immunoglobulins (IVIG). This systematic review and meta-analysis examined the studies which assessed immunoglobulin G (IgG) concentrations and the therapeutic use of IVIG for individuals with ASD. Twelve studies that examined IgG levels suggested abnormalities in total IgG and IgG 4 subclass concentrations, with concentrations in these IgGs related to aberrant behavior and social impairments, respectively. Meta-analysis supported possible subsets of children with ASD with low total IgG and elevated IgG 4 subclass but also found significant variability among studies. A total of 27 publications reported treating individuals with ASD using IVIG, including four prospective, controlled studies (one was a double-blind, placebo-controlled study); six prospective, uncontrolled studies; 2 retrospective, controlled studies; and 15 retrospective, uncontrolled studies. In some studies, clinical improvements were observed in communication, irritability, hyperactivity, cognition, attention, social interaction, eye contact, echolalia, speech, response to commands, drowsiness, decreased activity and in some cases, the complete resolution of ASD symptoms. Several studies reported some loss of these improvements when IVIG was stopped. Meta-analysis combining the aberrant behavior checklist outcome from two studies demonstrated that IVIG treatment was significantly associated with improvements in total aberrant behavior and irritability (with large effect sizes), and hyperactivity and social withdrawal (with medium effect sizes). Several studies reported improvements in pro-inflammatory cytokines (including TNF-alpha). Six studies reported improvements in seizures with IVIG (including patients with refractory seizures), with one study reporting a worsening of seizures when IVIG was stopped. Other studies demonstrated improvements in recurrent infections, appetite, weight gain, neuropathy, dysautonomia, and gastrointestinal symptoms. Adverse events were generally limited but included headaches, vomiting, worsening behaviors, anxiety, fever, nausea, fatigue, and rash. Many studies were limited by the lack of standardized objective outcome measures. IVIG is a promising and potentially effective treatment for symptoms in individuals with ASD; further research is needed to provide solid evidence of efficacy and determine the subset of children with ASD who may best respond to this treatment as well as to investigate biomarkers which might help identify responsive candidates.


2019 ◽  
Vol 40 (6) ◽  
pp. 1421-1454 ◽  
Author(s):  
Tamar Kalandadze ◽  
Valentina Bambini ◽  
Kari-Anne B. Næss

AbstractIndividuals with autism spectrum disorder (ASD) often experience difficulty in comprehending metaphors compared to individuals with typical development (TD). However, there is a large variation in the results across studies, possibly related to the properties of the metaphor tasks. This preregistered systematic review and meta-analysis (a) explored the properties of the metaphor tasks used in ASD research, and (b) investigated the group difference between individuals with ASD and TD on metaphor comprehension, as well as the relationship between the task properties and any between-study variation. A systematic search was undertaken in seven relevant databases. Fourteen studies fulfilled our predetermined inclusion criteria. Across tasks, we detected four types of response format and a great variety of metaphors in terms of familiarity, syntactic structure, and linguistic context. Individuals with TD outperformed individuals with ASD on metaphor comprehension (Hedges’ g = −0.63). Verbal explanation response format was utilized in the study showing the largest effect size in the group comparison. However, due to the sparse experimental manipulations, the role of task properties could not be established. Future studies should consider and report task properties to determine their role in metaphor comprehension, and to inform experimental paradigms as well as educational assessment.


Sign in / Sign up

Export Citation Format

Share Document