Machine Learning to Identify Autism Risk in Infancy: A study in the Danish National Birth Cohort and Cork Baseline Birth Cohort

Project Area: Population Health Research

Project Summary

Routine neonatal screening for autism could significantly reduce time to diagnosis and facilitate early intervention. There is firm evidence that early intervention can improve clinical outcome. Despite the known benefits, early intervention is limited by difficulties in diagnosing autism at an early age, and at present diagnosis relies solely on specialist behavioural and psychiatric assessment. Machine Learning is increasingly employed to produce prognostic models in psychiatric disorders, employing genetic, clinical, and OMICS data as predictors with excellent success.

The collaborative project is aimed at using big datasets with a complex set of sequential and related autism clinical risk factors, in combination with multi-OMICS data, to aid the development of a prediction model for autism in infancy.

Data from two internationally renowned birth cohorts; the Danish National Birth Cohort; ( the Cork BASELINE Birth Cohort (

The SPHeRE Scholar will employ routinely collected health records, interview data, and blood based biomarker measurements to aid the development of a prediction model for autism in infancy.

With advances in digital health care, this study will highlight the potential to leverage accessible health record data to empower clinical decision-making and advance the field of translational psychiatry. The project has the potential to translate knowledge on prenatal exposures in autism for clinical benefit, and advance early intervention strategies in psychiatry.

International placement
The successful candidate will have the opportunity to take up an international placement with collaborators Prof’s Bodil Hammer Beck and Tine Brink Henriksen, Aarhus University, Denmark.

Research skills
BSc or MSc in Computational Biology/ Data Science/ Bioinformatics is highly desirable. Experience in Machine Learning, OMICS data analysis or modelling in molecular biology would be an advantage.

Translational Medicine, Neuroscience, AI, Machine Learning, Prediction Modelling, Autism, Neurodevelopmental Disorders, Biomarkers.

Supervisory Team:

Dr Jane English (Principal Investigator), Department of Anatomy & Neuroscience, PI at INFANT, UCC.

Dr Ali Khashan, Epidemiology & Public Health, UCC, and PI at INFANT, UCC.

External Supervisor: Prof Bodil Beck, Aarhus University, Denmark.

Prof Barry O’Sullivan Director, Insight Centre for Data Analytics, School of Computer Science & IT, UCC.

This project will be based in UCC.