Interpretable deep learning radiomics from F-FDG PET/CT for differentiating diffuse large B-cell lymphoma and follicular lymphoma.
1/5 보강
PICO 자동 추출 (휴리스틱, conf 2/4)
유사 논문P · Population 대상 환자/모집단
250 patients from two centers.
I · Intervention 중재 / 시술
추출되지 않음
C · Comparison 대조 / 비교
추출되지 않음
O · Outcome 결과 / 결론
The proposed SHAP-based interpretable model offers superior diagnostic accuracy and potential clinical utility. [SUPPLEMENTARY INFORMATION] The online version contains supplementary material available at 10.1186/s12880-026-02253-y.
[OBJECTIVE] To develop and validate interpretable models integrating standardized uptake value (SUV), radiomics (Rad), and deep learning (DL) features from F-FDG PET/CT for differentiating diffuse lar
APA
Liu C, Zhang H, et al. (2026). Interpretable deep learning radiomics from F-FDG PET/CT for differentiating diffuse large B-cell lymphoma and follicular lymphoma.. BMC medical imaging, 26(1). https://doi.org/10.1186/s12880-026-02253-y
MLA
Liu C, et al.. "Interpretable deep learning radiomics from F-FDG PET/CT for differentiating diffuse large B-cell lymphoma and follicular lymphoma.." BMC medical imaging, vol. 26, no. 1, 2026.
PMID
41832418 ↗
Abstract 한글 요약
[OBJECTIVE] To develop and validate interpretable models integrating standardized uptake value (SUV), radiomics (Rad), and deep learning (DL) features from F-FDG PET/CT for differentiating diffuse large B-cell lymphoma (DLBCL) and follicular lymphoma (FL).
[METHODS] This retrospective study included 250 patients from two centers. Volumes of interest (VOIs) were delineated on PET images for SUV, Rad, and DL features extraction. Feature selection was performed using the Mann–Whitney U test, random forest–based recursive feature elimination, and the least absolute shrinkage and selection operator (LASSO). Seven machine learning classifiers were applied to construct diagnostic models, and fused Rad and DL features were further integrated to construct deep learning radiomics (DLR) models. Model interpretability was assessed using SHapley Additive exPlanations (SHAP). Model performance was evaluated in terms of discrimination, calibration, and clinical applicability.
[RESULTS] The DLR model achieved the best diagnostic performance, with an area under the curve (AUC) of 0.905 and an accuracy of 0.813 in the testing cohort. SHAP analysis identified the Rad feature “original_Maximum” as the most influential predictor for differentiating DLBCL from FL. Calibration curve and decision curve analyses further supported the superiority of the DLR model.
[CONCLUSION] Rad and DL features derived from F-FDG PET/CT enable effective differentiation between DLBCL and FL. The proposed SHAP-based interpretable model offers superior diagnostic accuracy and potential clinical utility.
[SUPPLEMENTARY INFORMATION] The online version contains supplementary material available at 10.1186/s12880-026-02253-y.
[METHODS] This retrospective study included 250 patients from two centers. Volumes of interest (VOIs) were delineated on PET images for SUV, Rad, and DL features extraction. Feature selection was performed using the Mann–Whitney U test, random forest–based recursive feature elimination, and the least absolute shrinkage and selection operator (LASSO). Seven machine learning classifiers were applied to construct diagnostic models, and fused Rad and DL features were further integrated to construct deep learning radiomics (DLR) models. Model interpretability was assessed using SHapley Additive exPlanations (SHAP). Model performance was evaluated in terms of discrimination, calibration, and clinical applicability.
[RESULTS] The DLR model achieved the best diagnostic performance, with an area under the curve (AUC) of 0.905 and an accuracy of 0.813 in the testing cohort. SHAP analysis identified the Rad feature “original_Maximum” as the most influential predictor for differentiating DLBCL from FL. Calibration curve and decision curve analyses further supported the superiority of the DLR model.
[CONCLUSION] Rad and DL features derived from F-FDG PET/CT enable effective differentiation between DLBCL and FL. The proposed SHAP-based interpretable model offers superior diagnostic accuracy and potential clinical utility.
[SUPPLEMENTARY INFORMATION] The online version contains supplementary material available at 10.1186/s12880-026-02253-y.
🏷️ 키워드 / MeSH 📖 같은 키워드 OA만
같은 제1저자의 인용 많은 논문 (5)
- Novel roles of SETD2 in tumor metabolism and immunotherapy: a systematic review and meta-analysis.
- Motion-robust magnetic resonance fingerprinting (MR-MRF) for quantitative liver cancer imaging.
- Effect of whole-course nutrition management on skeletal muscle mass in patients with gastric cancer undergoing neoadjuvant treatment.
- Luffa cylindrica flower extract induces apoptosis and autophagy in breast cancer cells.
- Regulation of G protein-coupled receptor kinase 2 and its role in tumors.
🏷️ 같은 키워드 · 무료전문 — 이 논문 MeSH/keyword 기반
- Microfluidic-based patient-derived organoids recapitulate thyroid cancer heterogeneity and reveal NF-κB-driven maturation for precision therapy.
- Association of Increased Nasal Fluids-Serum Concordance of Protein Profile with Prognosis in Nasal Polyps.
- Three Steps Hybrid Treatment: En Bloc Resection, Endovascular Treatment and Microsurgical Reconstruction. Case Report of a Giant Dorsum-Abdominal-Pelvic Arteriovenous Malformation.
- Clinical Characteristics and Prognostic Prediction of Secondary Solid Malignancies in Patients With Diffuse Large B-Cell Lymphoma and Follicular Lymphoma.
- Exosome membrane-biomimetic nanomedicine targets the pre-metastatic niche NF-κB inhibition to suppress breast cancer lung metastasis.
- Integration of Ki67 and Pan-Immune-Inflammation Value (PIV) into a predictive nomogram for pathologic complete response in triple-negative breast cancer : (Ki67 and inflammation in triple-negative breast cancer).