Biparametric MRI-Based Habitat Analysis Integrated With Deep Learning for Predicting Clinically Significant Prostate Cancer in PI-RADS Category 3 Lesions.
[BACKGROUND] Detection of clinically significant prostate cancer (csPCa) within PI-RADS category 3 lesions remains a major diagnostic challenge.
- 표본수 (n) 328
APA
Deng S, Hu J, et al. (2026). Biparametric MRI-Based Habitat Analysis Integrated With Deep Learning for Predicting Clinically Significant Prostate Cancer in PI-RADS Category 3 Lesions.. Journal of magnetic resonance imaging : JMRI, 63(4), 1165-1176. https://doi.org/10.1002/jmri.70205
MLA
Deng S, et al.. "Biparametric MRI-Based Habitat Analysis Integrated With Deep Learning for Predicting Clinically Significant Prostate Cancer in PI-RADS Category 3 Lesions.." Journal of magnetic resonance imaging : JMRI, vol. 63, no. 4, 2026, pp. 1165-1176.
PMID
41398999
Abstract
[BACKGROUND] Detection of clinically significant prostate cancer (csPCa) within PI-RADS category 3 lesions remains a major diagnostic challenge.
[PURPOSE] To develop and validate a biparametric MRI (bpMRI)-based habitat analysis model integrating deep learning features for predicting csPCa in PI-RADS 3 lesions using dual-center data.
[STUDY TYPE] Retrospective.
[POPULATION] This study included 551 patients with MRI-identified PI-RADS category 3 lesions and histopathological confirmation. A total of 439 patients from Center 1 were randomly assigned to a training set (n = 328) and an internal validation (in-vad) set (n = 111), while an external validation (ex-vad) set (n = 112) was obtained from Center 2.
[FIELD STRENGTH/SEQUENCE] 3 T/1.5 T. T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) sequences.
[ASSESSMENT] Lesions were manually segmented on preoperative T2WI and DWI, and tumor subregions were determined using k-means clustering. Deep learning features were obtained from each habitat subregion, and habitat-based models were built based on selected features. A habitat whole-tumor (Habitat W) model was subsequently derived by integrating all subregions. Recursive feature elimination (RFE) was applied to select the optimal predictors from the clinical and habitat-derived features; the clinical model was constructed using the selected clinical features, while the combined model incorporated all selected features.
[STATISTICAL TESTS] Student's t-test, Mann-Whitney U tests, Chi-squared tests, LASSO, areas under the curve (AUC), decision curve analysis (DCA), calibration curves, RFE, SHapley Additive exPlanations (SHAP). Statistical significance was defined as p-value < 0.05.
[RESULTS] In the training, in-vad and ex-vad sets, the clinical model demonstrated AUC values of 0.893, 0.844, and 0.837, respectively. The habitat models (habitat 1, 2,3 and -W) achieved AUCs ranging from 0.857 to 0.952. The combined model yielded AUCs of 0.959, 0.963, and 0.949, respectively.
[DATA CONCLUSION] The bpMRI-based deep learning Habitat W and combined model enables accurate assessment of csPCa in PI-RADS 3 lesions.
[TECHNICAL EFFICACY STAGE] 3.
[PURPOSE] To develop and validate a biparametric MRI (bpMRI)-based habitat analysis model integrating deep learning features for predicting csPCa in PI-RADS 3 lesions using dual-center data.
[STUDY TYPE] Retrospective.
[POPULATION] This study included 551 patients with MRI-identified PI-RADS category 3 lesions and histopathological confirmation. A total of 439 patients from Center 1 were randomly assigned to a training set (n = 328) and an internal validation (in-vad) set (n = 111), while an external validation (ex-vad) set (n = 112) was obtained from Center 2.
[FIELD STRENGTH/SEQUENCE] 3 T/1.5 T. T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) sequences.
[ASSESSMENT] Lesions were manually segmented on preoperative T2WI and DWI, and tumor subregions were determined using k-means clustering. Deep learning features were obtained from each habitat subregion, and habitat-based models were built based on selected features. A habitat whole-tumor (Habitat W) model was subsequently derived by integrating all subregions. Recursive feature elimination (RFE) was applied to select the optimal predictors from the clinical and habitat-derived features; the clinical model was constructed using the selected clinical features, while the combined model incorporated all selected features.
[STATISTICAL TESTS] Student's t-test, Mann-Whitney U tests, Chi-squared tests, LASSO, areas under the curve (AUC), decision curve analysis (DCA), calibration curves, RFE, SHapley Additive exPlanations (SHAP). Statistical significance was defined as p-value < 0.05.
[RESULTS] In the training, in-vad and ex-vad sets, the clinical model demonstrated AUC values of 0.893, 0.844, and 0.837, respectively. The habitat models (habitat 1, 2,3 and -W) achieved AUCs ranging from 0.857 to 0.952. The combined model yielded AUCs of 0.959, 0.963, and 0.949, respectively.
[DATA CONCLUSION] The bpMRI-based deep learning Habitat W and combined model enables accurate assessment of csPCa in PI-RADS 3 lesions.
[TECHNICAL EFFICACY STAGE] 3.
MeSH Terms
Humans; Male; Prostatic Neoplasms; Deep Learning; Retrospective Studies; Middle Aged; Aged; Magnetic Resonance Imaging; Reproducibility of Results; Prostate; Image Interpretation, Computer-Assisted
같은 제1저자의 인용 많은 논문 (5)
- Antibody-drug conjugates: A new twist to overcome EGFR-TKIs resistance in non-small cell lung cancer.
- Cancer-associated fibroblasts in non-small cell lung cancer therapy: Challenges and opportunities.
- Repurposing of the Macrolide Antibiotic Clarithromycin for the Prevention of Lung Cancer.
- Decoding the Plasticity of Cancer-Associated Fibroblasts: Mechanistic Insights and Precision Targeting Strategies in Gastric Cancer Progression and Therapeutic Resistance.
- SLC25A3 promotes hepatocellular carcinoma progression by inducing mitochondrial dysfunction and chromatin remodeling.