Predicting autism traits from baby wellness records: A machine learning approach.

LAY ABSTRACT: Timely identification of autism spectrum conditions is a necessity to enable children to receive the most benefit from early interventions. Emerging technological advancements provide avenues for detecting subtle, early indicators of autism from routinely collected health information. This study tested a model that provides a likelihood score for autism diagnosis from baby wellness visit records collected during the first 2 years of life. It included records of 591,989 non-autistic children and 12,846 children with autism. The model identified two-thirds of the autism spectrum condition group (boys 63% and girls 66%). Sex-specific models had several predictive features in common. These included language development, fine motor skills, and social milestones from visits at 12-24 months, mother's age, and lower initial growth but higher last growth measurements. Parental concerns about development or hearing impairment were other predictors. The models differed in other growth measurements and birth parameters. These models can support the detection of early signs of autism in girls and boys by using information routinely recorded during the first 2 years of life.

A Prediction Model of Autism Spectrum Diagnosis from Well-Baby Electronic Data Using Machine Learning.

Early detection of autism spectrum disorder (ASD) is crucial for timely intervention, yet diagnosis typically occurs after age three. This study aimed to develop a machine learning model to predict ASD diagnosis using infants' electronic health records obtained through a national screening program and evaluate its accuracy. A retrospective cohort study analyzed health records of 780,610 children, including 1163 with ASD diagnoses. Data encompassed birth parameters, growth metrics, developmental milestones, and familial and post-natal variables from routine wellness visits within the first two years. Using a gradient boosting model with 3-fold cross-validation, 100 parameters predicted ASD diagnosis with an average area under the ROC curve of 0.86 (SD < 0.002). Feature importance was quantified using the Shapley Additive explanation tool. The model identified a high-risk group with a 4.3-fold higher ASD incidence (0.006) compared to the cohort (0.001). Key predictors included failing six milestones in language, social, and fine motor domains during the second year, male gender, parental developmental concerns, non-nursing, older maternal age, lower gestational age, and atypical growth percentiles. Machine learning algorithms capitalizing on preventative care electronic health records can facilitate ASD screening considering complex relations between familial and birth factors, post-natal growth, developmental parameters, and parent concern.

An escalating continuum of learning and attention difficulties from premutation to full mutation in female carriers of FMR1 expansion.

Objective: Carriers of Fragile X premutation may have associated medical comorbidities, such as Fragile X-associated tremor and ataxia (FXTAS) and Fragile X-associated premature ovarian insufficiency (FXPOI). We examined the Fragile X premutation effect on cognition, and we assumed that there is a direct correlation between the continuous spectrum of specific learning and attention deficits to the number of CGG repeats on the FMR1 gene. Methods: A total of 108 women were referred to our center due to a related Fragile X syndrome (FXS) patient, 79 women carried a premutation of 56-199 repeats, and 19 women carried a full mutation of more than 200 CGG repeats on FMR1 gene. Genetic results of CGG repeats, demographic information, structured questionnaires for ADHD, learning disabilities of language and mathematics, and independence level were analyzed in women carrying the FMR1 premutation and compared to the group carrying the full mutation. Women with FXS and FXTAS were excluded. Results: When analyzed as a continuum, there was a significant increase in the following complaints which were associated with a higher number of repeats: specific daily function skills such as driving a car, writing checks, disorientation in directions, and also specific learning difficulties such as spelling and math difficulties. Additionally, when tested as a categorical independent variable, we observe that women with the full mutation were more likely to have ADHD or other learning disability diagnoses in the past than during premutation (<200 CGG repetitions). Conclusion: Specific learning and attention difficulties and resulting daily function difficulties correlate with an increased number of CGG repeats and are more likely to be associated as a common feature of premutation and full mutation in a female premutation carrier. Despite evidence of learning and attention difficulties, it is encouraging that most female carriers of the premutation and full mutation function well in most areas. Nevertheless, they face significant difficulties in specific areas of functioning such as driving, and confusion in times and schedules. Those daily function skills are mostly impacted by dyscalculia, right and left disorientation, and attention difficulties. This may aid to design specific interventions to address specific learning deficits in order to improve daily function skills and quality of life.

The myth of vaccination and autism spectrum.

BACKGROUND: Among all of the studied potential causes of autism, vaccines have received some of the most scrutiny and have been the topic of many evidence-based studies. These efforts have led the great majority of scientists, physicians, and public health researchers to refute causation between vaccines and autism. RATIONALE: This presumed association and concern has been a major contributor to parents' refusal to immunize their children and has become a major threat to public health in secluded populations over the last two decades, even prior to the COVID-19 pandemic. With the emergence of COVID-19 immunizations, sentiments towards this topic were addressed as a public health concern that may influence the ability to overcome the Corona virus worldwide. SCIENTIFIC REVIEW OF DATA: Despite the overwhelming data demonstrating that there is no link between vaccines and autism, many parents are hesitant to immunize their children because of the alleged association. Other contributing factors to the myths and conspiracy theories surrounding the association between vaccines and autism include the fact that the diagnosis of autism is typically made after the age of receiving the main childhood immunizations, as well as the occasional occurrence of regression after the age of first year vaccinations. In spite of vast evidence that the main contribution to the increase in incidence is from improvement of the diagnostic process, this rapid and publicized rise in autism diagnoses feeds parental concerns regarding any medical intervention that may be associated with the health of their children. RECOMMENDATIONS: It is plausible that with more evidence-based studies linking autism to specific etiologies the myth will diminish and disappear eventually. In an era where conspiracy theories are prevalent on social media, it is critical that evidence-based studies relating autism to specific etiologies be made public, and that information concerning autism diagnosis and causes be made more readily available through social media and parental organizations.

The Weak Link: Hypotonia in Infancy and Autism Early Identification.

Background: Presenting symptoms and age specific differential diagnosis of Autism Spectrum Disorder (ASD), determine the age of initial assessment and the age of a definite diagnosis. The AAP recommends screening all children for ASD at 18 and 24 months followed by a comprehensive evaluation for children with developmental concerns. More recently it has been recommended that the evaluation should be performed at a younger age, with a diagnosis being made as early as the beginning of the second year of life resulting in earlier intensive intervention. Objective: To assess early developmental milestones in a cohort of children diagnosed with Autism Spectrum Disorder (ASD) in order to find an objective and reliable early marker. We suggest that low muscle tone- hypotonia, is a sign that meets the above criteria of consistency and reliability and may be related to early diagnosis. Methods: We compared age distributions of ASD diagnosis in the presence of hypotonia in a dataset of 5,205 children diagnosed at Keshet Center. One thousand, one hundred eighty-two children (953 males) were diagnosed with ASD and compared to other developmental diagnoses. Within the ASD cohort we further analyzed for gender and pre-maturity differences. Results: In the presence of hypotonia, the mean age for ASD diagnosis was significantly lower (by 1.5 years for males and females) and this effect increased in children born at term as compared to pre-maturity. Conclusions: Hypotonia is a recognizable marker of ASD and may serve as a "red flag" to prompt earlier recognition and neurodevelopmental evaluation toward an autism diagnosis.

Autism risk linked to prematurity is more accentuated in girls.

INTRODUCTION: Prematurity has been identified as a risk factor for Autism Spectrum Disorder (ASD). The link between Autism Spectrum Disorder (ASD) and birth-week has not been strongly evidenced. We evaluated the correlation between the degree of prematurity and the incidence of autism in a cohort of 871 children born prematurely and followed from birth. The cohort was reduced to 416 premature infants born between 2011-2017 who were followed for 2-14 years, and analyzed according to birth week (degree of prematurity), and according to gender. RESULTS: 43 children (10.3%) received a definite diagnosis of ASD. There was a significant correlation between birth week and the risk of ASD, with 22.6% of children diagnosed with ASD when born at 25 weeks, versus 6% of ASD diagnoses at 31 weeks of prematurity. For children born after 32 weeks, the incidence decreased to 8-12.5%. A strong link was found between earlier birth week and increased autism risk; the risk remained elevated during near-term prematurity in boys. A correlation between early birth week and an elevated risk for ASD was seen in all children, but accentuated in females, gradually decreasing as birth week progresses; in males the risk for ASD remains elevated for any birth week. CONCLUSION: A statistically significant increase in rates of autism was found with each additional week of prematurity. Females drove this direct risk related to degree of prematurity, while males had an elevated risk throughout prematurity weeks, even at near-term. We recommend including ASD screening in follow up of infants born prematurely, at all levels of prematurity.