This study validates the efficiency and clinical applicability of the YOLOv11 framework for RCC subtyping and develops and validate an automated deep learning system for simultaneous tumor segmentation and histopathological subtyping using multicenter contrast-enhanced CT (CECT) imaging.
Abstract
Accurate preoperative discrimination of renal cell carcinoma (RCC) subtypes is critical for treatment stratification. We aimed to develop and validate an automated deep learning system for simultaneous tumor segmentation and histopathological subtyping using multicenter contrast-enhanced CT (CECT) imaging. To this end, we constructed a two-stage system comprising separate segmentation and classification models. The segmentation model was trained on 245 scans from Nanfang Hospital and 210 from the KiTS19 public dataset. The classification model was developed and validated on a total of 750 patients, comprising an internal cohort from Nanfang Hospital (553 patients; 328 training, 112 validation, 113 testing) and two external validation cohorts: one from Beijing Tongren Hospital (n = 111) and another combined from two other centers (n = 86). The model demonstrated strong generalizability for discriminating clear cell RCC, with AUCs of 0.878 (internal validation), 0.892 (internal testing), 0.911 (external set Ⅰ), and 0.892 (external set Ⅱ). The model's computational efficiency reached 0.24 s per file and reduced FLOPs by four times compared to conventional 3D CNNs. This study validates the efficiency and clinical applicability of the YOLOv11 framework for RCC subtyping. Future efforts should integrate prospective data and multimodal imaging to enhance sensitivity for small lesions.
The aim of the present study was to establish a preoperative prediction model of Ki-67 expression in renal cell carcinoma (RCC) by combining CT images of RCC with deep learning technology, and to evaluate its effect in clinical application. A retrospective analysis was performed on the CT images and pathological data of 137 patients with RCC who underwent renal CT plain scan plus enhancement scans and who were diagnosed pathologically from January 2019 to November 2023 at the Affiliated Hospital of Hebei University. All recruited patients were divided into 35 cases with Ki-67 ≥10% and 102 cases with Ki-67 <10% based on Ki-67 expression. Of these, 110 patients were divided into the training group and the test group in a 4:1 ratio, and the remaining 27 cases were used as a clinical validation group. Using the Mobilenetv3-large model, prediction models incorporating plain scan, arterial, venous and excretion phases were constructed using the training set data. The predictive performance of the models, including accuracy, accuracy, sensitivity, specificity, F1 score and area under the curve (AUC) values, were assessed by inputting CT images of the test group. The best-performing model was applied to the clinical validation group to compare the predicted results with the actual results of pathology; the accuracy, sensitivity, specificity and k coefficient of the model were calculated to assess its clinical efficacy. The Mobilenetv3-large model in the venous phase showed the best performance in terms of accuracy, sensitivity and F1 score, as well as relatively high precision and AUC value, indicating good robustness. In the clinical validation group, the model predicted Ki-67 expression with an accuracy of 0.814, sensitivity and specificity of 0.889 and 0.667 at low and high levels, respectively, and a k coefficient of 0.57. The Mobilenetv3-large model, when applied to venous-phase CT images, demonstrates notable clinical utility. It can proficiently forecast Ki-67 expression levels, thereby offering a precise, expedient and non-invasive aid in the formulation of tailored therapeutic strategies.
Dan Shen, Hongmei Li, Bingye Shi et al.· Oncology Letters· 0 citations
Background The preoperative differentiation of lung adenocarcinoma subtypes is critical for implementing personalized treatment but is difficult to accomplish with conventional imaging. This study aimed to develop an interpretable multimodal model integrating clinical, peritumoral, radiomic, and deep learning features to improve diagnostic accuracy. Methods A total of 3,038 patients from four hospitals were divided into training (n=1,822), test (n=608), and validation (n=608) sets. Two radiologists manually segmented two-dimensional tumor regions on computed tomography using ITK-SNAP software. After Pearson correlation analysis and least absolute shrinkage and selection operator regression, the radiomic score and deep learning score were generated. Clinical features were selected via univariate analysis, the Boruta algorithm, and recursive feature elimination (RFE). Individual logistic models were built and fused with the optimal combination selected via support vector machine-synthetic minority oversampling technique and extreme gradient boosting. Performance was evaluated in terms of the Obuchowski index, accuracy, F1-score, calibration, and decision curves, while interpretability was assessed via Shapley additive explanations (SHAP) and individual conditional expectation (ICE). Results The fused model achieved Obuchowski indices of 0.85 [95% confidence interval (CI): 0.84–0.87], 0.81 (95% CI: 0.78–0.83), and 0.79 (95% CI: 0.76–0.81) in the training, test, and validation sets, respectively outperforming the single-modality models. The F1-scores for the lepidic, acinar/papillary, and solid/micropapillary subtypes, respectively, were 0.77, 0.61, and 0.63 in the training set; 0.72, 0.57, and 0.59 in the test set; and 0.74, 0.54, and 0.52 in the validation set. Calibration and decision curve analysis confirmed the robustness and clinical utility of the model. SHAP analysis identified ResNet-101 feature as the best predictor, followed by peritumoral radiomic score, and lobulation. ICE plots revealed the linear and monotonic relationships between key features and predicted probabilities across subtypes. Conclusions The radiomics model developed in this study facilitates the accurate and interpretable preoperative classification of lung adenocarcinoma subtypes. Fusion of clinical, peritumoral, and deep learning features enhances diagnostic performance and supports clinical decision-making.
Feng-Juan Tian, Jing Ding, Zhen-Yu Cao et al.· Quantitative Imaging in Medi...· 0 citations
BACKGROUND
Pediatric posterior fossa tumors vary in malignancy, treatment, and prognosis across tumor types and molecular subtypes, yet noninvasive preoperative differentiation remains challenging.
PURPOSE
To develop a deep learning (DL) pipeline using T2-weighted (T2w) MR images to automatically segment pediatric posterior fossa tumors, differentiate tumor types (medulloblastoma [MB], ependymoma [EP], pilocytic astrocytoma [PA]), and classify molecular subtypes of MB and EP.
STUDY TYPE
Retrospective and prospective.
POPULATION
1305 patients (M/F: 828/477; 490 MB, 327 EP, and 488 PA) from three centers. For tumor segmentation and classification, PF-nnU-Net was developed on the training set (n = 880) and validated on a validation set (n = 220), an internal prospective test set (n = 90), and two external independent test sets (n = 68, n = 47). MB-nnU-Net was trained on 338 patients and tested on 63 patients for MB subtyping; a prior developed EP-nnU-Net was tested on 38 patients for EP subtyping.
FIELD STRENGTH/SEQUENCE
1.5 T or 3 T MRI, axial T2w images (turbo spin echo).
ASSESSMENT
Three nnU-Net-based models: PF-nnU-Net and MB-nnU-Net for development, EP-nnU-Net for validation. Five-fold cross-validation was performed on training sets, followed by testing on independent test sets.
STATISTICAL TESTS
Dice similarity coefficient for segmentation. Accuracy, sensitivity, specificity, the area under the receiver operating characteristic curve (AUC), and Cohen's kappa for classification. A two-sided p < 0.05 was considered significant.
RESULTS
PF-nnU-Net achieved Dice scores of 0.94-0.96 and overall classification accuracy of 0.824-0.918 (multiclass Cohen's kappa: 0.722-0.873). MB-nnU-Net attained an overall accuracy of 0.794 (multiclass Cohen's kappa: 0.605), and EP-nnU-Net achieved an accuracy of 0.789 (Cohen's kappa: 0.538).
DATA CONCLUSION
A fully automated DL pipeline was developed and validated to accurately segment pediatric posterior fossa tumors, differentiate tumor types (MB, EP, PA), and classify MB and EP molecular subtypes.
EVIDENCE LEVEL
3.
TECHNICAL EFFICACY
Stage 2.
Ying Jin, Yang-Yang Li, Ren-Long Zhang et al.· Journal of Magnetic Resonanc...· 0 citations
Accurate segmentation of non-small cell lung cancer (NSCLC) on positron emission tomography/computed tomography (PET/CT) is an essential prerequisite for automated metabolic tumor volume (MTV) quantification and staging. Although deep learning models achieve high performance on large-scale datasets, their generalization across different clinical domains is limited by variations in imaging protocols and patient demographics. This study aims to evaluate several deep learning architectures and investigate a transfer learning strategy to mitigate domain shift. Three architectures, including ResNet-backbone 3D U-Net, nnU-Net v2, and Swin UNETR, were benchmarked from scratch and compared with a fine-tuned nnU-Net initialized with AutoPET II weights. Results on the internal dataset showed that the fine-tuned nnU-Net achieved a Dice similarity coefficient (DSC) of 83.4 ± 6.5%, a 95% Hausdorff distance (HD95) of 5.1 ± 3.6 mm, and a precision of 89.6 ± 8.2%. Compared to the nnU-Net v2, the fine-tuned nnU-Net improved the absolute DSC by 6.8% while reducing local training time by 37.5% by bypassing the initial feature-learning phase. The fine-tuned nnU-Net model also demonstrated a high correlation between the MTV and the ground truth (Pearson r = 0.96, p < 0.001), indicating its potential as a reliable automated approach for quantitative MTV extraction and NSCLC prognostic-related analysis.
Q. Hồ, Ngoc Ha Bui, T. Tran et al.· Journal of Imaging· 0 citations
A hybrid architecture in which ResNet50 is employed for localized spatial feature extraction, while Vision Transformer enables global contextual learning to automatically classify kidney tumors into multiple classes is proposed.
Hivi Kamal, W. M. Abduallah· passer of basic and applied...· 0 citations
Breast cancer is currently one of the leading malignancies and mortality rates among women globally, creating an urgent need for accurate and efficient automated diagnostic tools. This study proposes and systematically compares convolutional neural network (CNN) architectures, including VG16, ResNet18, ResNet50, DenseNet121, EfficientNet-B0, and Swin-Transformer, applied to whole slide image (WSI) histopathology data for breast cancer prediction and classification. The models were trained and evaluated on the same WSI dataset with consistent image preprocessing techniques, allowing for objective comparison of performance. Experimental results showed that the performance of the models varied depending on the training strategy. In the non-fine-tune setting, ResNet18 achieved the best results with a C-index of 0.6587 and a Mean time-dependent AUC of 0.7317. When performing fine-tuning, ResNet50 outperformed the other architectures with a C-index of 0.6877 and a Mean AUC of 0.7738. Meanwhile, in the non-fine-tuned setup with combined clinical testing, Swin-Transformer achieved the highest performance with a C-index of 0.6793 and a Mean AUC of 0.7619.
Cuu-Duong Dang, T. Le, An-Thai Vo et al.· 2026 International Workshop...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
