Close Menu
Healthtost
  • News
  • Mental Health
  • Men’s Health
  • Women’s Health
  • Skin Care
  • Sexual Health
  • Pregnancy
  • Nutrition
  • Fitness
  • Recommended Essentials
What's Hot

PSA-based tool improves decision-making for prostate cancer screening and treatment

January 17, 2026

Why it still makes sense to limit saturated fat

January 17, 2026

The FDA-cleared ADHD device is not effective in reducing symptoms

January 16, 2026
Facebook X (Twitter) Instagram
  • About Us
  • Contact Us
  • Privacy Policy
  • Terms and Conditions
  • Disclaimer
Facebook X (Twitter) Instagram
Healthtost
SUBSCRIBE
  • News

    PSA-based tool improves decision-making for prostate cancer screening and treatment

    January 17, 2026

    The FDA-cleared ADHD device is not effective in reducing symptoms

    January 16, 2026

    Incretin-based diabetes drugs show possible protective effects against dementia

    January 16, 2026

    Does night work increase the risk of osteoporosis?

    January 15, 2026

    Scientists uncover promising therapeutic target for autoimmune disease that affects the brain

    January 15, 2026
  • Mental Health

    How to apply for a fully funded PhD in the UK

    January 8, 2026

    9 Secrets on How to Stop Procrastinating

    January 6, 2026

    Setting boundaries for self-care in 2026

    January 4, 2026

    In a world of digital money, what is the proper etiquette for splitting the bill with friends?

    January 1, 2026

    Rest is essential during the holidays, but it can mean getting active, not crashing on the couch

    December 26, 2025
  • Men’s Health

    Why it still makes sense to limit saturated fat

    January 17, 2026

    Escape Gym Groundhog Day: Why your workout takes seasons

    January 16, 2026

    What is Blue Collar Guilt?

    January 14, 2026

    Glucose stability in diabetes is enhanced by natural daylight

    January 10, 2026

    I reconcile my father’s anger and mine: some hills don’t deserve to die

    January 10, 2026
  • Women’s Health

    What really works? – Vuvatech

    January 16, 2026

    What is mental wellness and how does it differ from mental health?

    January 14, 2026

    Beyond weight loss: How GLP-1 ‘miracle drugs’ are revolutionizing whole-body health

    January 14, 2026

    8 Simple Food Swaps to Improve Your Health

    January 13, 2026

    Ways Omega-3s Benefit Women Specifically

    January 13, 2026
  • Skin Care

    An OUMERE Scientific and Regul – OUMERE

    January 16, 2026

    Collagen Induction Treatments in Rittenhouse Square

    January 15, 2026

    🥜⚠️ Why nut allergies are on the rise—and what it means for its future

    January 14, 2026

    Postnatal massage: Benefits, timing and what to book

    January 13, 2026

    Prepare your lips for the cold with TNW Lip Balm – The Natural Wash

    January 12, 2026
  • Sexual Health

    Your ultimate guide to climax and orgasm control

    January 16, 2026

    Stillbirths may be more common in US than previously known—Study

    January 14, 2026

    COVID-19 heightens vulnerabilities for women asylum seekers and refugee women in South Africa < SRHM

    January 14, 2026

    What does an unclear test result mean?

    January 13, 2026

    Relationship diversity, conflict, and why it matters for sex counselor certification — Sexual Health Alliance

    January 12, 2026
  • Pregnancy

    Weighing in: How GLP-1s fit into your pregnancy plans

    January 15, 2026

    5 foods, 4 habits, 3 reminders

    January 14, 2026

    I’m pregnant… Now what? 13 Things You Should Do Right Now

    January 14, 2026

    Which vitamins and minerals are important to consume during pregnancy?

    January 12, 2026

    Supporting emotional development in our children and ourselves – Podcast Ep 194

    January 11, 2026
  • Nutrition

    Sliced ​​meatballs | The Nutritionist Reviews

    January 16, 2026

    5-ingredient skillet dinner recipe

    January 15, 2026

    Family sheet-Gnocchi pan with vegetables and chicken sausage (30 minutes!)

    January 15, 2026

    3 Easy, Nutritious Salads – JSHealth

    January 13, 2026

    What to Eat During Weeks 2-4 on GLP-1: Simple Protein Plan | glp-1, weight loss, medical weight loss and more

    January 13, 2026
  • Fitness

    Not sure your multivitamin is working? 3 ways the signal could be missing

    January 16, 2026

    Barbell RDL: Proper Form & Benefits

    January 15, 2026

    Lazy high protein dinners that I make when I don’t feel like cooking

    January 15, 2026

    Behavioral health 101: What it means and why it matters

    January 14, 2026

    Snack smarter this New Year: 5 healthy low-calorie options

    January 13, 2026
  • Recommended Essentials
Healthtost
Home»News»GPT-4 demonstrates high accuracy in parsing multilingual medical notes
News

GPT-4 demonstrates high accuracy in parsing multilingual medical notes

healthtostBy healthtostJanuary 6, 2025No Comments6 Mins Read
Facebook Twitter Pinterest LinkedIn Tumblr Reddit WhatsApp Email
Gpt 4 Demonstrates High Accuracy In Parsing Multilingual Medical Notes
Share
Facebook Twitter LinkedIn Pinterest WhatsApp Email

The study evaluates the GPT-4’s ability to process medical notes in English, Spanish and Italian, achieving physician agreement 79% of the time.

Study: The ability of Generative Pre-trained Transformer 4 (GPT-4) to analyze medical notes in three different languages: a retrospective model evaluation study. Image credit: SuPatMaN/Shutterstock.com

In a recent study published in Lancet Digital Healtha group of researchers evaluated the ability of Generative Pre-trained Transformer 4 (GPT-4) to answer predefined questions based on medical notes written in three languages ​​(English, Spanish and Italian).

Background

Medical notes contain valuable clinical knowledge, yet their unstructured narrative form poses challenges for automated analysis.

Large language models (LLMs) such as GPT-4 show promise in extracting explicit details such as medications, but often struggle with implicit understanding of contexts, crucial for nuanced medical decision making. Variability in documentation styles between providers adds to the complexity.

Existing research demonstrates the potential of LLMs for free-text medical data processing, including decoding abbreviations and extracting social determinants of health, however these studies mainly focus on English language notes.

Further research is vital to enhance the ability of LLMs to handle complex tasks, improve contextual reasoning, and assess performance in multiple languages ​​and settings.

About the study

The present retrospective model evaluation study involved eight university hospitals from four countries: the United States of America (USA), Colombia, Singapore, and Italy.

The participating institutions were part of the 4CE Consortium. They included Boston Children’s Hospital, the University of Michigan, the University of Wisconsin, the National University of Singapore, the University of Kansas Medical Center, the University of Pittsburgh Medical Center, the Universidad de Antioquia, and the Istituti Clinici Scientifici Maugeri.

The Department of Biomedical Informatics at Harvard University served as the coordinating center. Each site contributed seven de-identified medical notes written between February 1, 2020 and June 1, 2023, resulting in a total of 56 medical notes, with six sites submitting notes in English, one in Spanish, and one in Italian.

Participating sites selected notes based on proposed criteria, including patients aged 18–65 years with a diagnosis of obesity and coronavirus disease 2019 (COVID-19) at admission. Compliance with these criteria was optional.

Notes submitted included admission, progress and consultation notes, but not discharge summaries. The notes were removed in accordance with the guidelines of the US Health Insurance Portability and Accountability Act, regardless of country of origin.

The study used the GPT-4 API in Python to analyze medical notes through a predefined question-answer framework. Parameters such as temperature, top-p and frequency penalty were adjusted to optimize performance.

Physicians rated the free-text responses and indicated whether they agreed with the GPT-4 responses. They were masked in each other’s ratings but not in the GPT-4 responses.

Statistical analyzes were performed to assess agreement between the GPT-4 and physicians, exploring instances of disagreement and categorizing errors as issues of derivation, inference, or hallucinations.

Subgroup analyzes and sensitivity analyzes addressed variations in accuracy, such as differences in language and specific inclusion criteria.

The study highlighted the ability of GPT-4 to process medical notes in multiple languages, but noted challenges in inference based on context and variability in documentation styles. Data analyzes were performed in RStudio and no external funding supported the study.

Study results

A total of 56 medical records were collected from eight sites in four countries: USA, Colombia, Singapore and Italy. Of these, 42 (75%) notes were in English, seven (13%) in Italian and seven (13%) in Spanish. For each note, the GPT-4 generated responses to 14 predefined questions, resulting in 784 responses.

Among them, both physicians agreed with the GPT-4 in 622 (79%) responses, one physician agreed in 82 (11%) responses, and neither physician agreed in 80 (10%) responses. When the National University of Singapore data were excluded, agreement rates remained similar: 534 (78%) responses had double agreement, 82 (12%) had partial agreement, and 70 (10%) had no agreement.

Physicians were more likely to agree with the GPT-4 for Spanish (86/98, 88%) and Italian (82/98, 84%) notes than for English notes (454/588, 77%).

Note type or length did not affect agreement rates. In cases where only one physician agreed with the GPT-4 (82 responses), 59 (72%) disagreements arose from issues of inference, such as different interpretations of implicit information.

In one case, a physician concluded that a patient did not have COVID-19 based on a “recent infection with COVID-19” note, while the GPT-4 left the condition as undetermined. Extraction problems accounted for 8 (10%) of these disagreements, such as a physician overlooking a documented medical history that identified GPT-4. Differences in level of agreement accounted for the remaining 15 (18%) cases.

In responses where both clinicians disagreed with the GPT-4 (80 responses), inference issues were most frequent (47/80, 59%), followed by inference errors (23/80, 29%) and hallucinations (10/ 80, 13% ).

For example, GPT-4 sometimes failed to link complications such as multisystem inflammatory syndrome as related to COVID-19, a connection made by both doctors. Delusion issues included information about the construction of the GPT-4 that was not in the notes, such as falsely claiming that a patient had COVID-19 when it was not reported.

When evaluating the ability of the GPT-4 to select patients for hypothetical study enrollment based on four inclusion criteria (age, obesity, COVID-19 status, and admission note type), its sensitivity varied. GPT-4 showed high sensitivity for obesity (97%), COVID-19 (96%) and age (94%), but lower specificity for admission notes (22%).

When the acceptance note criterion was excluded, the GPT-4 accurately identified all three remaining criteria 90% of the time.

conclusions

In summary, the study showed that the GPT-4 accurately analyzed medical notes in English, Italian, and Spanish, even without a direct technique.

Surprisingly, it performed better on Italian and Spanish notes than on English, possibly due to the greater complexity of US medical notes, although note length did not affect performance. GPT-4 efficiently extracted explicit information, but its main limitation was extracting implicit details.

This aligns with previous findings that models optimized for medical tasks can overcome such challenges. While the GPT-4 excelled at identifying explicit study inclusion criteria such as age and obesity, it struggled to classify admission notes, likely due to reliance on indirect constructs.

accuracy demonstrates GPT4 high medical multilingual Notes parsing
bhanuprakash.cg
healthtost
  • Website

Related Posts

PSA-based tool improves decision-making for prostate cancer screening and treatment

January 17, 2026

The FDA-cleared ADHD device is not effective in reducing symptoms

January 16, 2026

Incretin-based diabetes drugs show possible protective effects against dementia

January 16, 2026

Leave A Reply Cancel Reply

Don't Miss
News

PSA-based tool improves decision-making for prostate cancer screening and treatment

By healthtostJanuary 17, 20260

Prostate cancer is the second leading cause of cancer death in American men. About 1…

Why it still makes sense to limit saturated fat

January 17, 2026

The FDA-cleared ADHD device is not effective in reducing symptoms

January 16, 2026

Escape Gym Groundhog Day: Why your workout takes seasons

January 16, 2026
Stay In Touch
  • Facebook
  • Twitter
  • Pinterest
  • Instagram
  • YouTube
  • Vimeo
TAGS
Baby benefits body brain cancer care Day Diet disease exercise finds Fitness food Guide health healthy heart Improve Life Loss Men mental Natural Nutrition Patients People Pregnancy protein research reveals risk routine sex sexual Skin study Therapy Tips Top Training Treatment ways weight women Workout
About Us
About Us

Welcome to HealthTost, your trusted source for breaking health news, expert insights, and wellness inspiration. At HealthTost, we are committed to delivering accurate, timely, and empowering information to help you make informed decisions about your health and well-being.

Latest Articles

PSA-based tool improves decision-making for prostate cancer screening and treatment

January 17, 2026

Why it still makes sense to limit saturated fat

January 17, 2026

The FDA-cleared ADHD device is not effective in reducing symptoms

January 16, 2026
New Comments
    Facebook X (Twitter) Instagram Pinterest
    • About Us
    • Contact Us
    • Privacy Policy
    • Terms and Conditions
    • Disclaimer
    © 2026 HealthTost. All Rights Reserved.

    Type above and press Enter to search. Press Esc to cancel.