Background And Objectives: Recent advances in multimodal large language models (MLLMs) have shown promise in medical image interpretation, yet their utility in surgical contexts remains unexplored. This study evaluates six MLLMs' performance in interpreting diverse imaging modalities for laryngeal cancer surgery.

Methods: We analyzed 169 images (X-rays, CT scans, laryngoscopy, and pathology findings) from 50 patients using six state-of-the-art MLLMs. Model performance was assessed across 1084 clinically relevant questions by two independent physicians.

Results: Claude 3.5 Sonnet achieves the highest accuracy (79.43%, 95% CI: 77.02%-81.84%). Performance varied significantly across imaging modalities and between commercial and open-source models, with a 19-percentage point gap between the best commercial and open-source solutions.

Conclusion: Advanced MLLMs show promising potential as clinical decision support tools in laryngeal cancer surgery, while performance variations suggest the need for specialized model development and clinical workflow integration. Future research should focus on developing specialized MLLMs trained on large-scale multi-center laryngeal cancer datasets.

Download full-text PDF

Source
http://dx.doi.org/10.1097/JS9.0000000000002234DOI Listing

Publication Analysis

Top Keywords

laryngeal cancer
16
multimodal large
8
large language
8
language models
8
cancer surgery
8
image interpretation
8
imaging modalities
8
commercial open-source
8
models
4
models address
4

Similar Publications

Preoperative Nutritional Status Influences Enteral Nutrition Weaning 6 Months Post-Surgery in Patients with Esophageal Cancer.

Asian Pac J Cancer Prev

January 2025

Department of Anesthesiology and Resuscitology, Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University, Okayama, Japan.

Objective: This study aimed to clarify whether nutritional status at admission affects enteral nutrition weaning 6 months after surgery in patients with esophageal cancer.

Methods: This was a retrospective study of 81 patients who underwent subtotal esophageal cancer resection between April 2014 and February 2016. The survey items were as follows: 1) sex, 2) age, 3) presence or absence of family members living together, 4) clinical stage, 5) surgical procedure, 6) reconstructed organs, 7) nutritional status at admission, 8) presence or absence of postoperative complications (anastomotic leakage, chylothorax, and recurrent laryngeal nerve paralysis), and 9) presence or absence of treatment other than surgery (chemo- or radiotherapy).

View Article and Find Full Text PDF

Background: Airway obstruction secondary to chyle leak is an exceptionally rare phenomenon. Here, we describe this complication in a patient with anaplastic thyroid carcinoma (ATC) undergoing consolidative surgery after BRAF-targeted therapy.

Methods: A 55-year-old man presented with a rapidly enlarging neck mass.

View Article and Find Full Text PDF

Objective: TRIB3 has been confirmed to participate in and regulate biological metabolic activities in head and neck tumors such as nasopharyngeal carcinoma and oropharyngeal carcinoma, so the purpose of this study was to explore whether there is a correlation between TRIB3 and Laryngeal Squamous Cell Carcinoma (LSCC) and to preliminarily explore the biological characteristics of TRIB3 in LSCC.

Methods: TRIB3 expression in the LSCC was analyzed based on The Cancer Genome Atlas (TCGA) database. CCK-8 assay, Colony Formation Assay, wound healing assay, and Transwell assay were performed to investigate the roles of TRIB3 in the proliferation, invasion and metastasis of LSCC.

View Article and Find Full Text PDF

Background And Objectives: Recent advances in multimodal large language models (MLLMs) have shown promise in medical image interpretation, yet their utility in surgical contexts remains unexplored. This study evaluates six MLLMs' performance in interpreting diverse imaging modalities for laryngeal cancer surgery.

Methods: We analyzed 169 images (X-rays, CT scans, laryngoscopy, and pathology findings) from 50 patients using six state-of-the-art MLLMs.

View Article and Find Full Text PDF

Malignant neoplasms arise within a region of chronic inflammation caused by tissue injuries. Inflammation is a key factor involved in all aspects of tumorigenesis including initiation, proliferation, invasion, angiogenesis, and metastasis. Interleukin-1 (IL-1) plays critical functions in tumor development with influencing the tumor microenvironment and promoting cancer progression.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!