Skip to content
Log in
Open access

TFE3‐DualNet: An Interpretable Foundation Model‐Based Deep Learning Ensemble for Diagnosing TFE3‐Rearranged Renal Cell Carcinoma From Whole‐Slide Images in a Two‐Center Cohort

Yu-Hang Chen Quanhui Xu Haohua Yao Ke-Zhi Liu Chengpeng Gui L. Fu Ying-han Wang Jiang Zhu Jun-Cai Li Minyu Chen Kang-Bo Huang Han-cui Lin B. Liao Yun Cao Jin Wei Peng-Ju Li Jun-Hang Luo Jiazheng Cao
Aug 2026 · Cancer Medicine · Vol 15 · 0 citations · 32 references
Medicine

Abstract

TFE3‐rearranged renal cell carcinoma (TFE3‐rRCC) is a rare, aggressive subtype that predominantly affects adolescents and young adults. Its marked morphologic heterogeneity can delay recognition and downstream confirmatory testing.

DOI PubMed Semantic Scholar
Read PDF

Similar papers

Open access Jul 2026

MD-Mamba: A multi-scale dilated state-space network for breast cancer histopathology image classification

Highlights • MD-Mamba integrates state-space modeling with multi-scale dilated convolutions.• Dual-path attention improves interpretability and focuses on diagnostic tissue regions.• Achieves 96.25% accuracy with perfect malignant classification on BACH dataset.• Enables efficient, interpretable image biomarkers for breast cancer pathology.

Gengxun Liu, Shengquan Luo, Can Wu et al. · 0 citations
Open access Aug 2026

Real‐World Validation of the Clinical Distinctiveness of IDH‐Mutant Glioblastoma in the SEER Transition Era: A Population‐Based Study Integrating Machine Learning

The 2021 WHO classification reclassified “IDH‐mutant glioblastoma (GBM)” as “Astrocytoma, IDH‐mutant, grade 4.” This study aims to provide real‐world validation of this reclassification using the specific ICD‐O‐3 code (9445/3) from the Surveillance, Epidemiology, and End Results (SEER) “Transition Era” (2018–2022) and develop a machine learning (ML)‐based prognostic model.

De-Wei Du, Amu Jike, Dongnan Yu et al. · 0 citations
Open access Aug 2026

An Interpretable MRI-Based Machine Learning Model for Preoperative Identification of the Vascular Dissemination Phenotype in Hepatocellular Carcinoma: A Dual-Center Study.

Objective To develop and externally validate an interpretable MRI-based machine-learning model for preoperative identification of the vascular dissemination phenotype in hepatocellular carcinoma (HCC), defined by vessels encapsulating tumor clusters (VETC) and/or microvascular invasion (MVI). Materials and Methods This dual-center retrospective study included 642 patients with surgically confirmed HCC who underwent preoperative contrast-enhanced MRI. Patients from Institution I (n = 435) and Institution II (n = 207) formed the training and independent external validation cohorts, respectively. Clinical and conventional MRI predictors were selected using univariable logistic regression, collinearity assessment, and recursive feature elimination. Nine machine-learning models were developed and externally validated. Performance was assessed using area under the curve (AUC), sensitivity, specificity, calibration, decision-curve analysis, and subgroup analyses. SHAP was used for model interpretation, and transcriptomic analysis was performed in 30 patients. Results Among the nine machine-learning models, XGBoost achieved the highest observed AUCs in the training cohort (0.85; 95% CI: 0.81, 0.88) and external validation cohort (0.82; 95% CI: 0.75, 0.88), with higher AUCs than logistic regression in both cohorts (p < 0.001 and p = 0.047, respectively). In external validation, sensitivity and specificity were 71.1% and 85.5%, respectively. Calibration and decision-curve analyses supported model performance. Subgroup discrimination remained acceptable for tumors ≤ 5.0 cm and BCLC stage 0 or A disease, although sensitivity was lower for tumors ≤ 5.0 cm. SHAP identified intratumoral artery, nonsimple nodular growth type, and necrosis or severe ischemia as leading contributors. Transcriptomic analysis suggested exploratory enrichment of cell-cycle and metabolic pathways. Conclusion The interpretable MRI-based XGBoost model showed favorable performance for identifying the vascular dissemination phenotype in HCC, with SHAP-based interpretability and exploratory transcriptomic context.

Junhan Pan, Cong Zhang, Yi-Tian Wu et al. · 0 citations
Open access Aug 2026

A multimodal radiomics, deep learning, and pathomics signature for predicting the prognosis in central conventional chondrosarcoma

Highlights • Prognostic tools remain limited for central conventional chondrosarcoma.• A multimodal signature integrated CT radiomics, deep learning, and pathomics.• The signature predicted progression-free survival in independent cohorts.• The signature may support postoperative risk stratification.

Qiushi Su, Jie Wu, Ben Li et al. · 0 citations
Aug 2026

High-dimensional overfitting and noisy diagnostic labels: reconsidering the role of BRAF copy number in primary dermal melanoma.

We address a fundamental ground-truth fallacy in the study's design: utilizing admittedly imperfect clinical diagnoses as the absolute standard to train classification models creates a circular logic. We propose that BRAF amplification in these cases more likely serves as a diagnostic corrective marker for misclassified dermal metastases, rather than a true prognostic indicator for isolated PDM.

Jingyang Zhang, Hai-Wen Wang, Qin Zheng et al. · 0 citations