Skip to main content

Transparent deep learning to identify autism spectrum disorders (ASD) in EHR using clinical notes.

Leroy G, Andrews JG, KeAlohi-Preece M, Jaswani A, Song H, Galindo MK, Rice SA. J Am Med Inform Assoc. 2024 May 20;31(6):1313-1321. doi: 10.1093/jamia/ocae080. PMID: 38626184; PMCID: PMC11105145.

Read the Abstract

Presenter

Gondy Leroy, PhD
University of Arizona’s Eller College of Management

Statement of Purpose

This project is an interdisciplinary project that combines a focus on machine learning and autism spectrum disorders. Overall, our goal is to support clinicians with much or little mental health expertise to diagnose children with autism early. The age of diagnosis is later than 4 years old while 3 years old or younger would be much better. And even though the prevalence of autism is going up, the age of diagnosis is not coming down. Machine learning has been used by several different research groups to help diagnose autism. However, most of these projects have created a black-box approach where an entire record, or equivalent, is used by the model to produce a single label, e.g., autism or not autism.

Our project is different. We focus on the intermediary steps in recognizing autism. We aim to label individual behaviors with the relevant diagnostic criteria which allows us to propose a diagnosis. Our labels are based on the DSM5. By using this approach, we have created a transparent approach: it is easy to see why a label of autism or not (or any variant) is assigned to a case. The approach also makes it easy to correct a decision. Finally, the approach is also very robust and achieves very high performance.

Learning Outcomes

  • Explain the difference between transparent and black box machine learning.
  • Understand the different machine learning models and how quickly progess is made in the field.
  • Gain the ability to diagnose autism using machine learning.

Format

35-minute presentation by article author(s) considering salient features of the published study and its potential impact on practice

25-minute discussion of questions submitted by listeners via the webinar tools and moderated by JAMIA Student Editorial Board members.

CME Credit

The American Medical Informatics Association is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians.

The American Medical Informatics Association designates this live activity for a maximum of 1.0 AMA PRA Category 1™ credits. Physicians should claim only the credit commensurate with the extent of their participation in the activity.

CNE Credit

The American Medical Informatics Association is accredited as a provider of nursing continuing professional development by the American Nurses Credentialing Center's Commission on Accreditation.

  • Approved Contact Hours: 1.0 total
  • Nurse Planner: Jenna Thate, PhD, RN, CNE

 

Dates and Times: -
Type: Webinar
Course Format(s): On Demand
Price: Free
Share