Deep learning for preoperative MRI-based endometrial cancer staging prediction.

Predicting ki-67 expression in breast cancer via transformer and multiple instance learning on DCE-MRI

Accurate assessment of Ki-67 expression levels in breast cancer is crucial for determining prognosis and making informed treatment decisions. Current immunohistochemical methods relying on needle biopsy introduce sampling errors due to tumor spatial heterogeneity, making the development of non-invasive, precise preoperative prediction methods of significant clinical importance. This study aims to explore and compare advanced deep learning models based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for noninvasive assessment of Ki-67 expression. This retrospective study analyzed preoperative DCE-MRI data from 308 patients with histologically confirmed breast cancer. Adjacent slices centered on the tumor’s most significant cross-section were obtained to create a 2.5-dimensional (2·5D) dataset. We innovatively developed two deep learning models using the same dataset (1): a Multi-Instance Learning (MIL) model that combines slice-level predictive features with Predictive Likelihood Histogram (PLH) and Bag-of-Words (BoW) techniques (2); a Transformer-based fusion model that directly captures global contextual relationships between slices via self-attention mechanisms. The predictive performance of both models was systematically compared with traditional radiomics and clinical models. On an independent test set, the Transformer fusion model demonstrated optimal predictive performance with an area under the curve (AUC) of 0.875, achieving accuracy, sensitivity, and specificity of 0.839, 0.848, and 0.833, respectively. The MIL model ranked second (AUC = 0.825), with both models significantly outperforming traditional radiomics models (AUC = 0.698) and clinical models (AUC = 0.648). Deep learning models based on 2·5D DCE-MRI, especially Transformer models that achieve global feature fusion through self-attention mechanisms, can effectively and non-invasively predict Ki-67 expression status in breast cancer, surpassing traditional methods. This model shows potential as a reliable tool to help clinicians accurately assess tumor proliferation activity before surgery.

EQS-net for feature-optimized cervical cancer diagnosis using multi-layer fusion & dimensionality reduction

Cervical cancer remains a leading cause of cancer-related mortality among women worldwide, yet its progression is largely preventable through timely and accurate diagnosis. Conventional Pap smear screening pipelines, however, are constrained by subjective interpretation, diagnostic complexity, and limited throughput — barriers that impede scalable deployment in resource-limited settings. To address these critical gaps, this study presents EQS-NET, a novel, lightweight, and feature-optimized Computer-Aided Diagnosis (CAD) framework that leverages a heterogeneous ensemble of compact CNNs — ShuffleNet, SqueezeNet, and EfficientNet — to extract complementary deep representations via transfer learning. EQS-NET employs a Multi-Layer Deep Feature Fusion strategy aggregating discriminative features across the final three convolutional layers of each network, refined through mRMR-based feature selection. The framework eliminates the need for hand-crafted feature engineering, image segmentation, and cytology-specific preprocessing, relying only on minimal standard preprocessing. Validated under stratified 5-fold cross-validation on two benchmark datasets, EQS-NET achieves 98.5% ± 0.18% on SIPaKMeD and 99.98% ± 0.04% on Mendeley LBC (p < 0.001 against single-CNN baselines), outperforming existing state-of-the-art methods across sensitivity, specificity, and AUC — establishing it as a scalable, efficient, and clinically viable solution for automated cervical cancer screening.

[Medical prior-guided TabMap deep learning model for ovarian cancer prediction and interpretability analysis].

CLNAD-Net: A Multi-Task Learning Framework for Cervical Lymph Node Computer-Aided Diagnosis Network.

Deep Learning-Based Lung Cancer Segmentation and Volumetric Analysis Using CT Imaging: A Comprehensive Survey.

Although deep learning-based volumetric analysis has shown great potential in addressing the shortcomings of linear tumor assessment, several challenges still impede clinical implementation, including limited data availability, variability in annotation, sensitivity to scanners and acquisition protocols, poor interpretability, and multimodal integration.

Explainable Hybrid Deep Learning Approach for Multi-Modality Uterine Fibroid Detection and Clinical Decision Support

Uterine fibroids are one of the most common gynaecological tumours, but they can be hard to diagnose because their appearance varies and the images may have poor contrast. Existing computer-aided detection (CAD) methods often use features that were handcrafted or use only a single modality, which limits their reliability and interpretability. An Explainable Hybrid Deep Learning (E-HDL) framework is introduced in this study. It includes a self-improving preprocessing pipeline, adaptive multi-branch feature fusion, and attention-based interpretability for reliably detecting uterine fibroids on both ultrasound and MRI. To make sure that the input quality is the same, the process uses bias-field correction, anisotropic diffusion filtering, CLAHE contrast normalisation, and adaptive spatial alignment. Spatial deep features from a CNN are combined dynamically with contextual embeddings from Transformers and temporal representations from RNNs using an attention-based feature-scaling layer. Extensive experiments on the Mendeley Uterine Fibroid Dataset and cross-dataset evaluation on the HIFU-MRI and UFID-2023 benchmarks show that E-HDL works better than recent state-of-the-art models like Swin-Transformer, MedT, ConvNeXt, and EfficientNetV2. The suggested model is more accurate than previous combination and transformer-based systems, with a 97.1% F1-score and 0.989 AUC. Grad-CAM, SHAP, and attention heatmaps improve interpretability in clinical settings. The study shows that E-HDL is a high-performance, scalable, and easy-to-understand diagnostic aid for detecting uterine tumours.