From model performance to clinical action: artificial intelligence in perioperative gallbladder cancer care

Perioperative gallbladder cancer care requires linked decisions across diagnosis, staging, surgery, and postoperative management. Artificial intelligence (AI) and machine-learning systems have been evaluated as clinical decision support at each point, yet most studies emphasize discrimination rather than how model outputs would alter care. This review examines AI along the perioperative clinical pathway and interprets the available evidence in relation to its intended use. Diagnostic studies currently provide comparatively stronger evidence: prospective ultrasound cohorts, institution-level external validation and an offline assisted-reader study have shown that clinically relevant signals can be detected beyond the development dataset. By comparison, models for staging, resectability and postoperative risk remain largely retrospective, with small temporal or institution-level external validation cohorts, limited reporting of quantitative calibration and no established link between a model threshold and a clinical action. Some multimodal models have shown higher discrimination in matched comparisons, although their added value must be balanced against missing inputs, assay or imaging requirements, turnaround time and cost. Across these applications, live human–AI evaluation and comparative evidence of clinical impact remain limited. Prognostic risk and treatment response must also be distinguished from treatment benefit, which requires comparison between alternative treatment strategies. The current literature therefore supports AI as an investigational aid to clinical decision-making at clearly defined points in gallbladder cancer care, rather than as an autonomous decision maker. Progress toward clinical use will depend chiefly on evaluation of frozen systems across institutions, workflow-based human–AI studies and evidence of downstream clinical impact.

Authors

Institutions

Publication Details

Journal
Frontiers in Surgery
Published
2026-11-17
DOI
https://doi.org/10.3389/fsurg.2026.1953724
Primary Topic
Cholangiocarcinoma and Gallbladder Cancer Studies
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

From model performance to clinical action: artificial intelligence in perioperative gallbladder cancer care

Dongming Chen, Shan-Shan Jin, Jia-Wei Zhang, Hao-Yuan Xiong et al.
3 links
Frontiers in Surgery
Cholangiocarcinoma and Gallbladder Cancer Studies
article

From model performance to clinical action: artificial intelligence in perioperative gallbladder cancer care

Dongming Chen, Shan-Shan Jin, Jia-Wei Zhang, Hao-Yuan Xiong, Yong-Jun Jiang, Tao Wang, Tian-Tian Chen
article en

Abstract

Perioperative gallbladder cancer care requires linked decisions across diagnosis, staging, surgery, and postoperative management. Artificial intelligence (AI) and machine-learning systems have been evaluated as clinical decision support at each point, yet most studies emphasize discrimination rather than how model outputs would alter care. This review examines AI along the perioperative clinical pathway and interprets the available evidence in relation to its intended use. Diagnostic studies currently provide comparatively stronger evidence: prospective ultrasound cohorts, institution-level external validation and an offline assisted-reader study have shown that clinically relevant signals can be detected beyond the development dataset. By comparison, models for staging, resectability and postoperative risk remain largely retrospective, with small temporal or institution-level external validation cohorts, limited reporting of quantitative calibration and no established link between a model threshold and a clinical action. Some multimodal models have shown higher discrimination in matched comparisons, although their added value must be balanced against missing inputs, assay or imaging requirements, turnaround time and cost. Across these applications, live human–AI evaluation and comparative evidence of clinical impact remain limited. Prognostic risk and treatment response must also be distinguished from treatment benefit, which requires comparison between alternative treatment strategies. The current literature therefore supports AI as an investigational aid to clinical decision-making at clearly defined points in gallbladder cancer care, rather than as an autonomous decision maker. Progress toward clinical use will depend chiefly on evaluation of frozen systems across institutions, workflow-based human–AI studies and evidence of downstream clinical impact.

Frontiers in SurgeryVol. 13
North Sichuan Medical University (CN), Chengdu Military General Hospital (CN)
Good health and well-being
Cholangiocarcinoma and Gallbladder Cancer Studies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon