Accurate and interpretable age estimation and gender classification are essential in forensic and clinical diagnostics, particularly when using high-dimensional medical imaging data such as Cone Beam Computed Tomography (CBCT). Traditional CBCT-based approaches often suffer from high computational costs and limited interpretability, reducing their applicability in forensic investigations. This study aims to develop a multi-task deep learning framework that enhances both accuracy and explainability in CBCT-based age estimation and gender classification using attention mechanisms. We propose a multi-task learning (MTL) model that simultaneously estimates age and classifies gender using panoramic slices extracted from CBCT scans. To improve interpretability, we integrate Convolutional Block Attention Module (CBAM) and Grad-CAM visualization, highlighting relevant craniofacial regions. The dataset includes 2,426 CBCT images from individuals aged 7 to 23 years, and performance is assessed using Mean Absolute Error (MAE) for age estimation and accuracy for gender classification. The proposed model achieves a MAE of 1.08 years for age estimation and 95.3% accuracy in gender classification, significantly outperforming conventional CBCT-based methods. CBAM enhances the model's ability to focus on clinically relevant anatomical features, while Grad-CAM provides visual explanations, improving interpretability. Additionally, using panoramic slices instead of full 3D CBCT volumes reduces computational costs without sacrificing accuracy. Our framework improves both accuracy and interpretability in forensic age estimation and gender classification from CBCT images. By incorporating explainable AI techniques, this model provides a computationally efficient and clinically interpretable tool for forensic and medical applications.

Objective: Ultrasound Shear Wave Elastography (SWE) demonstrates great potential in assessing soft-tissue pathology by mapping tissue stiffness, which is linked to malignancy. Traditional SWE methods have shown promise in estimating tissue elasticity, yet their susceptibility to noise interference, reliance on limited training data, and inability to generate segmentation masks concurrently present notable challenges to accuracy and reliability. Approach: In this paper, we propose SW-ViT, a novel two-stage deep learning framework for SWE that integrates a CNN-Spatio-Temporal Vision Transformer-based reconstruction network with an efficient Transformer-based post-denoising network. The first stage uses a 3D ResNet encoder with multi-resolution spatio-temporal Transformer blocks that capture spatial and temporal features, followed by a squeeze-and-excitation attention decoder that reconstructs 2D stiffness maps. To address data limitations, a patch-based training strategy is adopted for localized learning and reconstruction. In the second stage, a denoising network with a shared encoder and dual decoders processes inclusion and background regions to produce a refined stiffness map and segmentation mask. A hybrid loss combining regional, smoothness, fusion, and Intersection over Union (IoU) components ensures improvements in both reconstruction and segmentation. Results: On simulated data, our method achieves PSNR of 32.68 dB, CNR of 46.78 dB, and SSIM of 0.995. On phantom data, results include PSNR of 21.11 dB, CNR of 42.14 dB, and SSIM of 0.936. Segmentation IoU values reach 0.949 (simulation) and 0.738 (phantom) with ASSD values being 0.184 and 1.011, respectively. Significance: SW-ViT delivers robust, high-quality elasticity map estimates from noisy SWE data and holds clear promise for clinical application.

3D medical image segmentation is vital for clinical diagnosis and treatment but is challenged by high-dimensional data and complex spatial dependencies. Traditional single-modality networks, such as CNNs and Transformers, are often limited by computational inefficiency and constrained contextual modeling in 3D settings. We introduce a novel multimodal framework that leverages Mamba and Kolmogorov-Arnold Networks (KAN) as an efficient backbone for long-sequence modeling. Our approach features three key innovations: First, an EGSC (Enhanced Gated Spatial Convolution) module captures spatial information when unfolding 3D images into 1D sequences. Second, we extend Group-Rational KAN (GR-KAN), a Kolmogorov-Arnold Networks variant with rational basis functions, into 3D-Group-Rational KAN (3D-GR-KAN) for 3D medical imaging - its first application in this domain - enabling superior feature representation tailored to volumetric data. Third, a dual-branch text-driven strategy leverages CLIP's text embeddings: one branch swaps one-hot labels for semantic vectors to preserve inter-organ semantic relationships, while the other aligns images with detailed organ descriptions to enhance semantic alignment. Experiments on the Medical Segmentation Decathlon (MSD) and KiTS23 datasets show our method achieving state-of-the-art performance, surpassing existing approaches in accuracy and efficiency. This work highlights the power of combining advanced sequence modeling, extended network architectures, and vision-language synergy to push forward 3D medical image segmentation, delivering a scalable solution for clinical use. The source code is openly available at https://github.com/yhy-whu/TK-Mamba.

Both CNN-based and Transformer-based methods have achieved remarkable success in medical image segmentation tasks. However, CNN-based methods struggle to effectively capture global contextual information due to the inherent limitations of convolution operations. Meanwhile, Transformer-based methods suffer from insufficient local feature modeling and face challenges related to the high computational complexity caused by the self-attention mechanism. To address these limitations, we propose a novel hybrid CNN-Transformer architecture, named MSLAU-Net, which integrates the strengths of both paradigms. The proposed MSLAU-Net incorporates two key ideas. First, it introduces Multi-Scale Linear Attention, designed to efficiently extract multi-scale features from medical images while modeling long-range dependencies with low computational complexity. Second, it adopts a top-down feature aggregation mechanism, which performs multi-level feature aggregation and restores spatial resolution using a lightweight structure. Extensive experiments conducted on benchmark datasets covering three imaging modalities demonstrate that the proposed MSLAU-Net outperforms other state-of-the-art methods on nearly all evaluation metrics, validating the superiority, effectiveness, and robustness of our approach. Our code is available at https://github.com/Monsoon49/MSLAU-Net.

This study investigates the effectiveness of synthetic training data in predicting 2D landmarks for 3D intraoral reconstruction in cleft lip and palate patients. We take inspiration from existing landmark prediction and 3D reconstruction techniques for faces and demonstrate their potential in medical applications. We generated both real and synthetic datasets from intraoral scans and videos. A convolutional neural network was trained using a negative-Gaussian log-likelihood loss function to predict 2D landmarks and their corresponding confidence scores. The predicted landmarks were then used to fit a statistical shape model to generate 3D reconstructions from individual images. We analyzed the model's performance on real patient data and explored the dataset size required to overcome the domain gap between synthetic and real images. Our approach generates satisfying results on synthetic data and shows promise when tested on real data. The method achieves rapid 3D reconstruction from single images and can therefore provide significant value in day-to-day medical work. Our results demonstrate that synthetic training data are viable for training models to predict 2D landmarks and reconstruct 3D meshes in patients with cleft lip and palate. This approach offers an accessible, low-cost alternative to traditional methods, using smartphone technology for noninvasive, rapid, and accurate 3D reconstructions in clinical settings.

Medical image segmentation is vital for clinical diagnosis, yet current deep learning methods often demand extensive expert effort, i.e., either through annotating large training datasets or providing prompts at inference time for each new case. This paper introduces a zero-shot and automatic segmentation pipeline that combines off-the-shelf vision-language and segmentation foundation models. Given a medical image and a task definition (e.g., "segment the optic disc in an eye fundus image"), our method uses a grounding model to generate an initial bounding box, followed by a visual prompt boosting module that enhance the prompts, which are then processed by a promptable segmentation model to produce the final mask. To address the challenges of domain gap and result verification, we introduce a test-time adaptation framework featuring a set of learnable adaptors that align the medical inputs with foundation model representations. Its hyperparameters are optimized via Bayesian Optimization, guided by a proxy validation model without requiring ground-truth labels. Our pipeline offers an annotation-efficient and scalable solution for zero-shot medical image segmentation across diverse tasks. Our pipeline is evaluated on seven diverse medical imaging datasets and shows promising results. By proper decomposition and test-time adaptation, our fully automatic pipeline performs competitively with weakly-prompted interactive foundation models.

Accurate prediction of prognosis and postoperative radiotherapy response is critical for personalized treatment in head and neck squamous cell carcinoma (HNSCC). We developed a multimodal deep learning model (MDLM) integrating computed tomography, whole-slide images, and clinical features from 1087 HNSCC patients across multiple centers. The MDLM exhibited good performance in predicting overall survival (OS) and disease-free survival in external test cohorts. Additionally, the MDLM outperformed unimodal models. Patients with a high-risk score who underwent postoperative radiotherapy exhibited prolonged OS compared to those who did not (P = 0.016), whereas no significant improvement in OS was observed among patients with a low-risk score (P = 0.898). Biological exploration indicated that the model may be related to changes in the cytochrome P450 metabolic pathway, tumor microenvironment, and myeloid-derived cell subpopulations. Overall, the MDLM effectively predicts prognosis and postoperative radiotherapy response, offering a promising tool for personalized HNSCC therapy.

Low-quality pseudo labels pose a significant obstacle in semi-supervised medical image segmentation (SSMIS), impeding consistency learning on unlabeled data. Leveraging vision-language model (VLM) holds promise in ameliorating pseudo label quality by employing textual prompts to delineate segmentation regions, but it faces the challenge of cross-modal alignment uncertainty due to multiple correspondences (multiple images/texts tend to correspond to one text/image). Existing VLMs address this challenge by modeling semantics as distributions but such distributions lead to semantic degradation. To address these problems, we propose Alignment-Multiplicity Aware Vision-Language Model (AMVLM), a new VLM pre-training paradigm with two novel similarity metric strategies. (i) Cross-modal Similarity Supervision (CSS) proposes a probability distribution transformer to supervise similarity scores across fine-granularity semantics through measuring cross-modal distribution disparities, thus learning cross-modal multiple alignments. (ii) Intra-modal Contrastive Learning (ICL) takes into account the similarity metric of coarse-fine granularity information within each modality to encourage cross-modal semantic consistency. Furthermore, using the pretrained AMVLM, we propose a pioneering text-guided SSMIS network to compensate for the quality deficiencies of pseudo-labels. This network incorporates a text mask generator to produce multimodal supervision information, enhancing pseudo label quality and the model's consistency learning. Extensive experimentation validates the efficacy of our AMVLM-driven SSMIS, showcasing superior performance across four publicly available datasets. The code will be available at: https://github.com/QingtaoPan/AMVLM.

Glaucoma is a group of serious eye diseases that can cause incurable blindness. Despite the critical need for early detection, over 60% of cases remain undiagnosed, especially in less developed regions. Glaucoma diagnosis is a costly task and some models have been proposed to automate diagnosis based on images of the retina, specifically the area known as the optic cup and the associated disc where retinal blood vessels and nerves enter and leave the eye. However, diagnosis is complicated because both normal and glaucoma-affected eyes can vary greatly in appearance. Some normal cases, like glaucoma, exhibit a larger cup-to-disc ratio, one of the main diagnostic criteria, making it challenging to distinguish between them. We propose a deep learning model with domain features (DLMDF) to combine unstructured and structured features to distinguish between glaucoma and physiologic large cups. The structured features were based upon the known cup-to-disc ratios of the four quadrants of the optic discs in normal, physiologic large cups, and glaucomatous optic cups. We segmented each cup and disc using a fully convolutional neural network and then calculated the cup size, disc size, and cup-to-disc ratio of each quadrant. The unstructured features were learned from a deep convolutional neural network. The average precision (AP) for disc segmentation was 98.52%, and for cup segmentation it was also 98.57%. Thus, the relatively high AP values enabled us to calculate the 15 reliable features from each segmented disc and cup. In classification tasks, the DLMDF outperformed other models, achieving superior accuracy, precision, and recall. These results validate the effectiveness of combining deep learning-derived features with domain-specific structured features, underscoring the potential of this approach to advance glaucoma diagnosis.

Whole Slide Images (WSIs) in histopathology present a significant challenge for large-scale medical image analysis due to their high resolution, large size, and complex tile relationships. Existing Multiple Instance Learning (MIL) methods, such as Graph Neural Networks (GNNs) and Transformer-based models, face limitations in scalability and computational cost. To bridge this gap, we propose the WSI-GMamba framework, which synergistically combines the relational modeling strengths of GNNs with the efficiency of Mamba, the State Space Model designed for sequence learning. The proposed GMamba block integrates Message Passing, Graph Scanning & Flattening, and feature aggregation via a Bidirectional State Space Model (Bi-SSM), achieving Transformer-level performance with 7* fewer FLOPs. By leveraging the complementary strengths of lightweight GNNs and Mamba, the WSI-GMamba framework delivers a scalable solution for large-scale WSI analysis, offering both high accuracy and computational efficiency for slide-level classification.