|
|
|
|
|
|
|
Cognitive impairment (CI) is often underdiagnosed in primary care settings, but machine learning (ML) analysis of patient-physician dialogue recordings may help identify speech patterns signaling CI.
In this JAMA Neurology Author Interview, Joseph T. Colonel, PhD, a postdoctoral research fellow at the Icahn School of Medicine at Mount Sinai, describes his acoustic analysis research. Colonel and his colleagues analyzed audio recordings of conversations during primary care visits using ML models trained on acoustic features that identified pitch, timing, and variability features as key correlates of cognitive impairment. They found these signs in 20% of previously undiagnosed patients.
Listen now on Spotify | Apple Podcasts | YouTube | JAMA.com.
Editor’s Picks in this week’s JAMA+ AI Update:
- In a diagnostic study of 288 emergency department encounters, applying different large language models (LLMs) to screen for missed opportunities for diagnosis (MOD) yielded meaningful differences in default threshold behavior. The findings suggest that LLMs may be more useful for clinician-facing prescreening than as stand-alone MOD adjudicators. (JAMA Network Open)
- Generative AI in clinical decision support should complement deterministic rules and conventional clinical natural language processing, not replace them, write the authors of this JAMA Perspective article. Large language models may help synthesize relevant information from notes, portal messages, and dispensing records, but their outputs should be grounded in curated, versioned sources with source attribution and deterministic verification. (JAMA)
- Clinician-facing AI medical knowledge platforms could degrade over time because the financial pressure on technology companies to maximize profits may conflict with quality goals. To prevent this outcome, clinicians can advocate for independent benchmarks, regulation and antitrust enforcement, and public and nonprofit platform alternatives to help ensure AI platforms truly enable better patient care. (JAMA Internal Medicine)
|
|
Original Investigation
JAMA
Acoustic Analysis of Primary Care Patient–Clinician Conversations to Screen for Cognitive Impairment
Joseph T. Colonel, PhD; Jacqueline Becker, PhD; Lili Chan, MD; et al
Original Investigation
JAMA Network Open
Screening for Missed Opportunities for Diagnosis in the ED Using eTriggers and Large Language Models
Clifford M. Marks, MD, MBA; Sean Gibney, MD; Bryan Stenson, MD; et al
Perspective
JAMA
How Generative AI Should Transform Clinical Decision Support
Siru Liu, PhD; Josh F. Peterson, MD, MPH; Adam Wright, PhD
Viewpoint
JAMA Internal Medicine
Artificial Intelligence and the Financialization of Medical Knowledge
Teva D. Brender, MD; Allan S. Detsky, MD, PhD; Lindsey Yourman, MD
AUDIO
 Acoustic Analysis of Primary Care Patient–Clinician Conversations to Screen for Cognitive Impairment
|
|
|
|
|
|
|
Thank you for subscribing to JAMA Network email alerts. This message was sent to buiduytam1@gmail.com by updates@email.jamanetwork.com.
To update your contact information, change your email preferences, or unsubscribe, click here.
To ensure you always receive JAMA Network emails, add the email address updates@email.jamanetwork.com to your address book.
To unsubscribe by mail, contact:
JAMA Network
AMA Plaza
330 N Wabash Ave
Chicago, IL 60611
Or call (800) 621-8335.
|
      |
|
©2026 American Medical Association. All rights reserved.
|
|
 |
|
Advertisement
|
|
|
