ViT-CB: Integrating hybrid Vision Transformer and CatBoost to enhanced brain tumor detection with SHAP

Biomedical Signal Processing and Control, Год журнала: 2024, Номер 100, С. 107027 - 107027

Опубликована: Окт. 24, 2024

Язык: Английский

---

Development and Validation of an AI-Based Multimodal Model for Pathological Staging of Gastric Cancer Using CT and Endoscopic Images

Academic Radiology, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

Язык: Английский

---

Differentiation of Citri Reticulatae Pericarpium varieties via HPLC fingerprinting of polysaccharides combined with machine learning

Food Chemistry, Год журнала: 2025, Номер unknown, С. 143053 - 143053

Опубликована: Янв. 1, 2025

Язык: Английский

---

​Multi-Parameter MRI Radiomics Differential Diagnosis of Medulloblastoma and Ependymoma in Children

Опубликована: Янв. 1, 2025

Язык: Английский

---

MRI-based deep learning with clinical and imaging features to differentiate medulloblastoma and ependymoma in children

Frontiers in Molecular Biosciences, Год журнала: 2025, Номер 12

Опубликована: Апрель 28, 2025

Medulloblastoma (MB) and ependymoma (EM) in children share similarities terms of age group, tumor location, clinical presentation, which makes it challenging to clinically diagnose distinguish them. The present study aims explore the effectiveness T2-weighted magnetic resonance imaging (MRI)-based deep learning (DL) combined with features for differentiating MB from EM. Axial MRI sequences obtained 201 patients across three centers were used model training testing. regions interest manually delineated by an experienced neuroradiologist supervision a senior radiologist. We developed DL classifier using pretrained AlexNet architecture that was fine-tuned on our dataset. To mitigate class imbalance, we implemented data augmentation employed K-fold cross-validation enhance generalizability. For patient classification, two voting strategies: hard strategy majority prediction selected individual image slices; soft scores averaged slices threshold 0.5. Additionally, multimodality fusion constructed integrating features. performance assessed 7:3 random split dataset validation, respectively. key metrics like sensitivity, specificity, positive predictive value, negative F1 score, area under receiver operating characteristic curve (AUC), accuracy calculated, statistical comparisons performed DeLong test. Thereafter, classified as positive, while EM negative. achieved AUC values 0.712 (95% confidence interval (CI): 0.625-0.797) set 0.689 CI: 0.554-0.826) test set. In contrast, demonstrated superior 0.987 0.974-0.996) 0.889 0.803-0.949) indicated statistically significant improvement compared (p < 0.001), highlighting its enhanced discriminative ability. MRI-based multimodal can be effectively differentiate children. Thus, structure decision tree is expected greatly assist clinicians daily practice.

Язык: Английский

---

Development and Validation of Multiparametric MRI-based Interpretable Deep Learning Radiomics Fusion Model for Predicting Lymph Node Metastasis and Prognosis in Rectal Cancer: A Two-center Study

Academic Radiology, Год журнала: 2024, Номер unknown

Опубликована: Дек. 1, 2024

Язык: Английский

---