Diagnostic performance of deep learning models using ultrasound for preoperative axillary nodal status in breast cancer: a systematic review and meta-analysis
- 1Department of Ultrasound, the First Affiliated Hospital of Shihezi University
- 2Tongji Hospital, Affiliated to Tongji Medical College, Huazhong University of Science and Technology
Abstract
Background: Deep-learning (DL) models may extract nodal-risk information from breast or axillary ultrasound, but the relevant studies differ in anatomical input, preprocessing, reference standard, and validation design.
Methods: We updated PubMed, Web of Science Core Collection, and Cochrane CENTRAL, with supplementary OpenAlex and Crossref searches and repeated backward and forward citation searching through 24 August 2026. Eligible studies evaluated a deep-learning model using ultrasound as the sole imaging modality for pretreatment binary axillary or sentinel nodal status in breast cancer, used surgical pathology as the reference standard, and reported an independent patient-level validation cohort with an exact or uniquely reconstructable 2 × 2 table. One cohort per study was selected using a prespecified hierarchy. Sensitivity and specificity were synthesized with a bivariate random-effects model and an HSROC curve.
Results: Eleven studies comprising 1,027 validation patients were included. Pooled sensitivity was 0.75 (95% CI 0.66–0.81; 95% prediction interval 0.50–0.90) and pooled specificity was 0.89 (95% CI 0.83–0.93; prediction interval 0.66–0.97). The summary HSROC area was 0.81. The positive likelihood ratio was 6.60 (95% CI 4.29–10.42), the negative likelihood ratio was 0.29 (0.21–0.38), and the diagnostic odds ratio was 23.12 (12.89–41.50). Descriptive univariate I² values were 64.2% for sensitivity and 73.8% for specificity. Deeks’ test did not indicate funnel-plot asymmetry (P = 0.677).
Conclusions: Ultrasound-based DL models showed moderate sensitivity and high specificity in selected validation cohorts. Wide prediction intervals, heterogeneous anatomical inputs and preprocessing, incompletely reported definitions of nodal positivity, and the geographic concentration of studies limit immediate clinical transportability. These models should be evaluated as adjuncts to, rather than replacements for, pathological axillary staging.
Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/ , identifier CRD42024589342.
Related
- Magnetic Resonance Imaging in the Preoperative Assessment of Anal Fistula: Diagnostic Performance and Comparison of the St. James and Garg Classifications
- Diagnostic Performance of Deep Learning Using Retinal Images for Detection of Chronic Kidney Disease: A Systematic Review and Meta-Analysis
- Diagnostic Performance of Color Doppler Ultrasound for Detecting Nuchal Cord and Associated Immediate Neonatal Outcomes at Term in Uganda: A Prospective Diagnostic‐Accuracy Cohort Study
- Predicting 1-year outcomes after anterior cruciate ligament reconstruction from preoperative physical performance using machine learning: an exploratory study
- Enhancing prenatal diagnostic workflow: a deep learning-based system for multi-class fetal ultrasound classification across all gestational stages
BibTeX
@article{liu2026diagnostic,
title = {{Diagnostic performance of deep learning models using ultrasound for preoperative axillary nodal status in breast cancer: a systematic review and meta-analysis}},
author = {Tianxin Liu and Xiaowu Yuan and Jinli Wang and Yuchen He and Yan Liu and Xiaojun Yan and Chunli Cao and Ming Chen and Wenxiao Li and Jun Li and Xinwu Cui},
journal = {Frontiers in Oncology},
year = {2026},
month = oct,
doi = {10.3389/fonc.2026.1925423},
url = {https://www.frontiersin.org/articles/10.3389/fonc.2026.1925423/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Diagnostic performance of deep learning models using ultrasound for preoperative axillary nodal status in breast cancer: a systematic review and meta-analysis
AU - Tianxin Liu
AU - Xiaowu Yuan
AU - Jinli Wang
AU - Yuchen He
AU - Yan Liu
AU - Xiaojun Yan
AU - Chunli Cao
AU - Ming Chen
AU - Wenxiao Li
AU - Jun Li
AU - Xinwu Cui
PY - 2026
DA - 2026/10/23
JO - Frontiers in Oncology
DO - 10.3389/fonc.2026.1925423
UR - https://www.frontiersin.org/articles/10.3389/fonc.2026.1925423/full
AB - Background: Deep-learning (DL) models may extract nodal-risk information from breast or axillary ultrasound, but the relevant studies differ in anatomical input, preprocessing, reference standard, and validation design. Methods: We updated PubMed, Web of Science Core Collection, and Cochrane CENTRAL, with supplementary OpenAlex and Crossref searches and repeated backward and forward citation searching through 24 August 2026. Eligible studies evaluated a deep-learning model using ultrasound as the sole imaging modality for pretreatment binary axillary or sentinel nodal status in breast cancer, used surgical pathology as the reference standard, and reported an independent patient-level validation cohort with an exact or uniquely reconstructable 2 × 2 table. One cohort per study was selected using a prespecified hierarchy. Sensitivity and specificity were synthesized with a bivariate random-effects model and an HSROC curve. Results: Eleven studies comprising 1,027 validation patients were included. Pooled sensitivity was 0.75 (95% CI 0.66–0.81; 95% prediction interval 0.50–0.90) and pooled specificity was 0.89 (95% CI 0.83–0.93; prediction interval 0.66–0.97). The summary HSROC area was 0.81. The positive likelihood ratio was 6.60 (95% CI 4.29–10.42), the negative likelihood ratio was 0.29 (0.21–0.38), and the diagnostic odds ratio was 23.12 (12.89–41.50). Descriptive univariate I² values were 64.2% for sensitivity and 73.8% for specificity. Deeks’ test did not indicate funnel-plot asymmetry (P = 0.677). Conclusions: Ultrasound-based DL models showed moderate sensitivity and high specificity in selected validation cohorts. Wide prediction intervals, heterogeneous anatomical inputs and preprocessing, incompletely reported definitions of nodal positivity, and the geographic concentration of studies limit immediate clinical transportability. These models should be evaluated as adjuncts to, rather than replacements for, pathological axillary staging. Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/ , identifier CRD42024589342.
SN - 2234-943X
ER -