Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: May 29, 2026

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

Adaptive Fast-Slow Large Language Model Framework for Multidimensional Classification of Prenatal Ultrasound Reports:

Wei Zhong¹, Huihui Yan², Yifan Liu²

¹Department of Medical Genetics, Beijing Obstetrics and Gynecology Hospital, Capital Medical University. Beijing Maternal and Child Health Care Hospital, Beijing, China.

Journal of Medical Internet Research

|May 28, 2026

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Open-Source Large Language Models Distilled DeepSeek-R1 Pose Challenges for On-Premises Clinical Deployment in Medical Diagnosis: A Comparative Study of Performance.

Journal of medical systems·2026

Same author

Maternal Haemoglobin and Haematocrit During Pregnancy and Stillbirth Risk: A Prospective, Population-Based Birth Cohort Study.

Paediatric and perinatal epidemiology·2026

Same author

Performance of ChatGPT-4o and Four Open-Source Large Language Models in Generating Diagnoses Based on China's Rare Disease Catalog: Comparative Study.

Journal of medical Internet research·2025

Same author

IL-33 Has a Protective Effect Against Perinatal Listeriosis by Regulating Mitophagy of CD8<sup>+</sup> T Lymphocytes.

Journal of interferon & cytokine research : the official journal of the International Society for Interferon and Cytokine Research·2025

Same author

Enhancing the Accuracy of Human Phenotype Ontology Identification: Comparative Evaluation of Multimodal Large Language Models.

Journal of medical Internet research·2025

Same author

Analysis of a Series of 26 Cases With Prenatal Skeletal Dysplasia via Multiplatform Genetic Detection.

Molecular genetics & genomic medicine·2025

Same journal

Comparative Effectiveness of AI-Assisted Telerehabilitation, Telerehabilitation, In-Person Care, and Usual Care for Chronic Nonspecific Low Back Pain: Bayesian Network Meta-Analysis.

Journal of medical Internet research·2026

Same journal

Effectiveness of WeChat Public Account Intervention Based on the Information-Motivation-Behavioral Skills Model Among College Students With Internet Addiction: Randomized Controlled Trial.

Journal of medical Internet research·2026

Same journal

Are Traditional Registries Becoming Obsolete in the Modern Digital Health Ecosystem?

Journal of medical Internet research·2026

Same journal

Detecting and Preventing Fraudulent Participation in Qualitative Research: Content Analysis of Two Multisite Studies.

Journal of medical Internet research·2026

Same journal

Patient Perceptions of Artificial Intelligence-Supported Shared Decision-Making in UK Primary Care for Multiple Long-Term Conditions: Qualitative Study.

Journal of medical Internet research·2026

Same journal

Impact of Telemedicine-Enhanced Integrated Management of Gestational Diabetes on Pregnancy Outcomes and Glycemic Control: Real-World Study Using TangMama App.

Journal of medical Internet research·2026

See all related articles

This summary is machine-generated.

Large language models (LLMs) can classify prenatal ultrasound reports for diagnosis. Chain-of-thought reasoning (CoT) excels in subjective assessments, improving accuracy for fetal anomaly diagnosis.

Area of Science:

Medical Informatics
Artificial Intelligence in Healthcare
Genetics and Genomics

Background:

Phenotype-driven prenatal diagnosis links ultrasound findings with genetic outcomes.
Unstructured clinical ultrasound reports hinder accurate correlation.
Large language models (LLMs) offer potential solutions, but their application is underexplored.

Purpose of the Study:

To develop an LLM framework for classifying prenatal ultrasound reports.
To evaluate DeepSeek-V3.2 models with retrieval-augmented generation (RAG) and chain-of-thought (CoT) reasoning.
To assess performance on factual and subjective classification dimensions.

Main Methods:

Evaluated DeepSeek-V3.2-B (base) and V3.2-R (reasoning-enhanced) on 254 fetal anomaly reports.

Keywords:

DeepSeek chain-of-thought large language models phenotype-driven diagnosis prenatal ultrasound retrieval-augmented generation

Related Experiment Videos

Last Updated: May 29, 2026

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

Assessed 5 classification dimensions: primary, terminology, anatomy, count, and severity.

Validated against amniocentesis-derived genetic outcomes.

Main Results:

V3.2-B achieved >90% accuracy for factual tasks but 56.6% for subjective severity.
CoT-enhanced V3.2-R showed 86% accuracy on external data without RAG; RAG degraded performance.
Accurate phenotypic profiles correlated significantly with pathogenic genetic risks.

Conclusions:

Base LLMs are efficient for factual data; CoT is crucial for subjective assessments.
CoT outperformed RAG in subjective tasks within this study's constraints.
An adaptive LLM framework can accelerate phenotype-genotype research and guide diagnostic decisions.