Pixel-level Radiomics and Deep Learning for Predicting Ki-67 Expression in Breast Cancer Based on Dual-modal Ultrasound Images.
1/5 보강
PICO 자동 추출 (휴리스틱, conf 2/4)
유사 논문P · Population 대상 환자/모집단
63 patients were prospectively enrolled for further validation.
I · Intervention 중재 / 시술
추출되지 않음
C · Comparison 대조 / 비교
추출되지 않음
O · Outcome 결과 / 결론
SHAP analysis provided visual interpretability of the model's decision-making process. [CONCLUSION] The V-MURC model based on pixel-level RFMs can accurately predict Ki-67 expression in BC and may serve as a valuable tool for individualized treatment decision-making in clinical practice.
ℹ️ 이 논문은 무료 전문이 아직 없습니다. 코퍼스 전체의 43.7%는 무료 가능 (통계 →) · 🏥 기관 EZproxy로 시도
[RATIONALE AND OBJECTIVES] This study aimed to develop a deep learning model using a novel pixel-level radiomics approach based on two-dimensional (2D) and strain elastography (SE) ultrasound images t
- 표본수 (n) 616
- p-value P < 0.05
- 95% CI 0.929 - 0.975
APA
Wei W, Xia F, et al. (2026). Pixel-level Radiomics and Deep Learning for Predicting Ki-67 Expression in Breast Cancer Based on Dual-modal Ultrasound Images.. Academic radiology, 33(3), 900-917. https://doi.org/10.1016/j.acra.2025.12.047
MLA
Wei W, et al.. "Pixel-level Radiomics and Deep Learning for Predicting Ki-67 Expression in Breast Cancer Based on Dual-modal Ultrasound Images.." Academic radiology, vol. 33, no. 3, 2026, pp. 900-917.
PMID
41539915 ↗
Abstract 한글 요약
[RATIONALE AND OBJECTIVES] This study aimed to develop a deep learning model using a novel pixel-level radiomics approach based on two-dimensional (2D) and strain elastography (SE) ultrasound images to predict Ki-67 expression in breast cancer (BC).
[METHODS] This multicenter study included 1031 BC patients, who were divided into training (n = 616), internal validation (n = 265), and external test (n = 150) cohorts. An additional 63 patients were prospectively enrolled for further validation. The deep learning model, termed Vision-Mamba, predicts Ki67 expression by integrating ultrasound (2D and SE) images with pixel-level radiomics feature maps (RFMs). A combined model was subsequently constructed by incorporating independent clinical predictors. Model performance was assessed using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). SHapley Additive exPlanations (SHAP) were applied to enhance interpretability.
[RESULTS] We developed a Vision-Mamba-US-RFMs-Clinical (V-MURC) model that integrates ultrasound images, RFMs, and clinical data for accurate prediction of Ki-67 expression in BC. The area under the ROC curve (AUC) values for the internal validation, external test, and prospective validation cohorts were 0.954 (95% CI, 0.929 - 0.975), 0.941 (95% CI, 0.903 - 0.975), and 0.945 (95% CI, 0.883 - 0.989), respectively, demonstrating excellent discrimination and calibration. Compared with individual models, the V-MURC model achieved significantly superior performance across all datasets (Delong test, P < 0.05). Calibration curves and DCA further supported its clinical applicability. SHAP analysis provided visual interpretability of the model's decision-making process.
[CONCLUSION] The V-MURC model based on pixel-level RFMs can accurately predict Ki-67 expression in BC and may serve as a valuable tool for individualized treatment decision-making in clinical practice.
[METHODS] This multicenter study included 1031 BC patients, who were divided into training (n = 616), internal validation (n = 265), and external test (n = 150) cohorts. An additional 63 patients were prospectively enrolled for further validation. The deep learning model, termed Vision-Mamba, predicts Ki67 expression by integrating ultrasound (2D and SE) images with pixel-level radiomics feature maps (RFMs). A combined model was subsequently constructed by incorporating independent clinical predictors. Model performance was assessed using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). SHapley Additive exPlanations (SHAP) were applied to enhance interpretability.
[RESULTS] We developed a Vision-Mamba-US-RFMs-Clinical (V-MURC) model that integrates ultrasound images, RFMs, and clinical data for accurate prediction of Ki-67 expression in BC. The area under the ROC curve (AUC) values for the internal validation, external test, and prospective validation cohorts were 0.954 (95% CI, 0.929 - 0.975), 0.941 (95% CI, 0.903 - 0.975), and 0.945 (95% CI, 0.883 - 0.989), respectively, demonstrating excellent discrimination and calibration. Compared with individual models, the V-MURC model achieved significantly superior performance across all datasets (Delong test, P < 0.05). Calibration curves and DCA further supported its clinical applicability. SHAP analysis provided visual interpretability of the model's decision-making process.
[CONCLUSION] The V-MURC model based on pixel-level RFMs can accurately predict Ki-67 expression in BC and may serve as a valuable tool for individualized treatment decision-making in clinical practice.
🏷️ 키워드 / MeSH 📖 같은 키워드 OA만
- Humans
- Breast Neoplasms
- Female
- Ki-67 Antigen
- Deep Learning
- Middle Aged
- Ultrasonography
- Mammary
- Aged
- Adult
- Elasticity Imaging Techniques
- Prospective Studies
- Image Interpretation
- Computer-Assisted
- Radiomics
- Breast cancer
- Deep learning
- Ki-67 expression
- Radiomics feature maps
- SHapley Additive exPlanations
같은 제1저자의 인용 많은 논문 (5)
- Deciphering MFAP5+ Fibroblasts in Pancreatic Cancer Progression via Multi-Regional Single-Cell RNA Sequencing With Experimental Validation.
- polysaccharides inhibit castration-resistant prostate cancer progression possibly via dual modulation of PIK3R1 and PGC-1α: a preclinical evaluation.
- A Longitudinal Study on Development and Validation of the Prostate Cancer Scale Among the System of Quality of Life Instruments for Patients with Cancer (QLICP-PR V2.0) Based on Classical Test Theory and Generalizability Theory.
- A new paradigm for treating lung cancer by targeting the intratumoral microbiome.
- Ionizing radiation: molecular mechanisms, biological effects, and therapeutic targets.
🏷️ 같은 키워드 · 무료전문 — 이 논문 MeSH/keyword 기반
- A Phase I Study of Hydroxychloroquine and Suba-Itraconazole in Men with Biochemical Relapse of Prostate Cancer (HITMAN-PC): Dose Escalation Results.
- Self-management of male urinary symptoms: qualitative findings from a primary care trial.
- Clinical and Liquid Biomarkers of 20-Year Prostate Cancer Risk in Men Aged 45 to 70 Years.
- Diagnostic accuracy of Ga-PSMA PET/CT versus multiparametric MRI for preoperative pelvic invasion in the patients with prostate cancer.
- Association of patient health education with the postoperative health related quality of life in low- intermediate recurrence risk differentiated thyroid cancer patients.
- Early local immune activation following intra-operative radiotherapy in human breast tissue.