Neurodevelopmental Risk Detection System Using Computer Vision and ML
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current screening practices for Autism Spectrum Disorder (ASD) in children are inefficient, leading to excessive false positives and false negatives, particularly in lower socioeconomic status groups, girls, and racial/ethnic minorities, resulting in delayed diagnosis and inadequate treatment.
Innovation Solution
A system and method for detecting ASD risk using a combination of electronically-administered screening surveys, direct observation via computer vision analysis, electronic health records, and genetic testing, with machine learning algorithms to derive a risk level and provide next-step guidance for referrals and treatment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If parent questionnaires and parent interviews are used for screening, then screening can be performed, but excessive false positives occur causing undue anxiety and increased wait times
Solution Approach 1:
The screening process is segmented into multiple independent components: initial parent questionnaire, automated risk level derivation, and conditional follow-up questions. This segmentation allows the system to process only high-risk cases through full evaluation, reducing wait times while maintaining accuracy.
Solution Approach 2:
A computerized system acts as an intermediary between parent questionnaires and specialized evaluation. The system derives risk levels automatically and determines which cases require follow-up, reducing false positives and optimizing the workflow between screening and specialized assessment.
2Reliability
If parent questionnaires and parent interviews are used for screening, then screening can be performed, but excessive false positives occur causing undue anxiety to parents
Solution Approach 1:
The screening process is segmented into multiple independent components: initial parent questionnaire, automated risk level derivation, and conditional follow-up questions. This segmentation allows the system to process only high-risk cases through full evaluation, reducing wait times while maintaining accuracy.
Solution Approach 2:
The system provides automated feedback through risk level derivation and actionable guidance, giving parents clear information about their child's screening status and next steps. This reduces uncertainty and anxiety compared to traditional methods where parents receive no feedback from positive screens.
3Reliability
If current screening practices are used, then screening can be performed, but inflated false negatives occur among lower maternal socioeconomic status groups, girls, and racial/ethnic minority populations
Solution Approach 1:
The screening system uses universal objective measures (computer vision, automated analysis) that function consistently across all populations regardless of socioeconomic status, gender, or ethnicity. This universal approach reduces false negatives in marginalized groups while maintaining overall screening accuracy.
4Measurement precision
If multiple screening methods are combined, then detection accuracy improves, but system complexity increases
Solution Approach 1:
Multiple screening methods (parent questionnaires, computer vision analysis, automated risk derivation) are merged into a single integrated system. The system combines these methods while managing complexity through automated processing and unified risk level derivation, achieving high detection accuracy without proportionally increasing operational complexity.
Data Source
AI summary
The present disclosure describes methods and systems for risk detection and intervention for neurodevelopmental disorders. The method includes assessment of risk level, guidance and treatment recommendations and strategies, and longitudinal monitoring of patients with neurodevelopmental disorders. The assessments and monitoring can be integrated into the patient's health care program and electronic health record (EHR).