Ishanu Chattopadhyay PRO
ML Data Science Biomedicine Social Science Faculty
Ishanu Chattopadhyay
Assistant Professor of Medicine
Universal Early Screening For ASD
University of Chicago Medicine
Onishchenko, Dmytro, Yi Huang, James van Horne, Peter J. Smith, Michael E. Msall, and Ishanu Chattopadhyay. "Reduced false positives in autism screening via digital biomarkers inferred from deep comorbidity patterns." Science Advances 7, no. 41 (2021): eabf0354.
https://docs.google.com/document/d/15iMXe3PkDfRpP9DeBdmGahQ342Aah9OCDWvaMHf4aAQ/edit?usp=sharing
NIH Proposal 1R21MH135316-01
Autism Spectrum Disorder + AI
Literature Search: AI + Target Disease
ML To Enable Universal Screening
Takes too long, not supported by insurance, objective assessments might be difficult
Even when such tools exist they are not administered universally
MCHAT/F (ASD)
Must do pattern discovery
Discover factors that modulate risk, beyond what is already known
Must account for the possibility of non-causal spurious associations
1 in 54
ASD: Ineffective screening causes delays and incurs costs
Autistic children experience higher co-morbidities
Can we exploit these patterns to predict diagnosis?
Common Knowledge: Comorbidties Exist
ACoR:
Autism Co-morbid Risk Score
D. Onishchenko, Y. Huang, J. van Horne, P. J. Smith, M. M. Msall, I. Chattopadhyay, Reduced false positives in autism screening via digital biomarkers inferred from deep comorbidity patterns. Sci. Adv. 7, eabf0354 (2021).
Autism Co-morbid Risk (ACoR) Score
Autism Co-morbid Risk (ACoR) Score
Automatically infer Importance of different comorbidity categories
17 categories chosen:
immune | infections | endocrine | ...
Truven MarketScan (IBM) Commerical Claims & Encounters Database 2003-2018
5M Children
UCM Database 2006-2018
70K Children
ACoR Prospective Trial
Patient Flow
Aim 1. Validate reduction of false positives compared to M-CHAT/F
Aim 2. Evaluate dependency between ACoR and M-CHAT/F
Aim 3. Evaluate performance in diverse populations
Aim 4. Characterize heterogeneity of ASD from comorbidity patterns
University of Chicago Women's Board
ACoR Prospective Trial
Patient Flow
Michael Msall Peter Smith James Mitchell
Conventional AI/ML often attempts to model the physician
Universality
Universal Screening at the point of care
@ zero additional burden
Impact
Retrospective Results
Autism Co-morbid Risk (ACoR) Score
MCHAT/F
Head to head comparison with current practice
Autism Co-morbid Risk (ACoR) Score: Geospatial effect
works well everywhere
Pattern Discovery amidst Heterogeneity
Patterns are discernible in the decisions, but not easily reducible to few human-understandable intuitions
Co-morbidity Spectra
Nervous disorders
Digestive disorders
Injury & Poisoning
Neoplasms
Endocrine
Immune
Specific codes that increase the odds of true positive vs true negative
Joint Operation with MCHAT
CHOP Study allows us to see effectiveness of MCHAT in different sub-populations
Modulate sensitivity/specificity trade-offs
ACoR: Variation with Age
can track risk increase over time
Older children are easier to diagnose
Performance on Diverse Populations
Largely insensitive to race/ethnicity
How?
Deep Learning Without Neural Networks: Fractal-nets for Rare Event Modeling (Under Review Nature Machine Intelligence)
Yi Huang, James Evans, I. Chattopadhyay
Sequence Likelihood Divergence For Fast Time Series Comparison
Yi Huang, Victor Rotaru, I. Chattopadhyay
Under Review IEEE Transactions of Data and Knowledge Engineering
Abductive learning of quantized stochastic processes with probabilistic finite automata
Ishanu Chattopadhyay and Hod Lipson
2013 Phil. Trans. R. Soc. A.3712011054320110543
Longitudinal patterns matter
Immune female control
Immune female case
Endocrine female control
Endocrine female case
Cardiovascular female control
Cardiovascular female case
Secret Sauce: Leverging Temporal Patterns
Specialized HMM models from code sequences
Model control and case cohorts seprately
given a new test case, compute likelihood of sample arising from case models vs control models
sequence likelihood defect
Obvious Features Not Highly Predictive
Sanity checks
Using how often a child is sick is NOT sufficient
Uncertainty in coding does not seem to radically impact results
Good disambiguation of ASD from ID and other disorders
By Ishanu Chattopadhyay
Using EHR for screening/diagnosis of complex diseases
ML Data Science Biomedicine Social Science Faculty