This study aims to develop and validate a novel two-step deep learning framework for the automated detection, segmentation, and classification of bone metastases in SPECT/CT imaging, accurately distinguishing malignant from benign lesions to improve early diagnosis and facilitate personalized treatment planning. A segmentation model, BL-Seg, was developed to automatically segment lesion regions in SPECT/CT images, utilizing a multi-scale attention fusion module and a triple attention mechanism to capture metabolic variations and refine lesion boundaries. A radiomics-based ensemble learning classifier was subsequently applied to integrate metabolic and texture features for benign-malignant differentiation. The framework was trained and evaluated using a proprietary dataset of SPECT/CT images collected from our institution. Performance metrics, including Dice coefficient, sensitivity, specificity, and AUC, were compared against conventional methods. The study utilized a dataset of SPECT/CT cases from our institution, divided into training and test sets acquired on Siemens SPECT/CT scanners with minor protocol differences. BL-Seg achieved a Dice coefficient of 0.8797, surpassing existing segmentation models. The classification model yielded an AUC of 0.8502, with improved sensitivity and specificity compared to traditional approaches. The proposed framework, with BL-Seg's automated lesion segmentation, demonstrates superior accuracy in detecting, segmenting, and classifying bone metastases, offering a robust tool for early diagnosis and personalized treatment planning in metastatic bone disease.

The use of deep learning (DL) in medical image analysis has significantly improved the ability to predict lung cancer. In this study, we introduce a novel deep convolutional neural network (CNN) model, named ResNet+, which is based on the established ResNet framework. This model is specifically designed to improve the prediction of lung cancer and diseases using the images. To address the challenge of missing feature information that occurs during the downsampling process in CNNs, we integrate the ResNet-D module, a variant designed to enhance feature extraction capabilities by modifying the downsampling layers, into the traditional ResNet model. Furthermore, a convolutional attention module was incorporated into the bottleneck layers to enhance model generalization by allowing the network to focus on relevant regions of the input images. We evaluated the proposed model using five public datasets, comprising lung cancer (LC2500 $n$=3183, IQ-OTH/NCCD $n$=1336, and LCC $n$=25000 images) and lung disease (ChestXray $n$=5856, and COVIDx-CT $n$=425024 images). To address class imbalance, we used data augmentation techniques to artificially increase the representation of underrepresented classes in the training dataset. The experimental results show that ResNet+ model demonstrated remarkable accuracy/F1, reaching 98.14/98.14\% on the LC25000 dataset and 99.25/99.13\% on the IQ-OTH/NCCD dataset. Furthermore, the ResNet+ model saved computational cost compared to the original ResNet series in predicting lung cancer images. The proposed model outperformed the baseline models on publicly available datasets, achieving better performance metrics. Our codes are publicly available at https://github.com/AIPMLab/Graduation-2024/tree/main/Peng.

The limited availability of bronchoscopy images makes image synthesis particularly interesting for training deep learning models. Robust image translation across different domains -- virtual bronchoscopy, phantom as well as in-vivo and ex-vivo image data -- is pivotal for clinical applications. This paper proposes BronchoGAN introducing anatomical constraints for image-to-image translation being integrated into a conditional GAN. In particular, we force bronchial orifices to match across input and output images. We further propose to use foundation model-generated depth images as intermediate representation ensuring robustness across a variety of input domains establishing models with substantially less reliance on individual training datasets. Moreover our intermediate depth image representation allows to easily construct paired image data for training. Our experiments showed that input images from different domains (e.g. virtual bronchoscopy, phantoms) can be successfully translated to images mimicking realistic human airway appearance. We demonstrated that anatomical settings (i.e. bronchial orifices) can be robustly preserved with our approach which is shown qualitatively and quantitatively by means of improved FID, SSIM and dice coefficients scores. Our anatomical constraints enabled an improvement in the Dice coefficient of up to 0.43 for synthetic images. Through foundation models for intermediate depth representations, bronchial orifice segmentation integrated as anatomical constraints into conditional GANs we are able to robustly translate images from different bronchoscopy input domains. BronchoGAN allows to incorporate public CT scan data (virtual bronchoscopy) in order to generate large-scale bronchoscopy image datasets with realistic appearance. BronchoGAN enables to bridge the gap of missing public bronchoscopy images.

We present our solution for the Multi-Source COVID-19 Detection Challenge, which classifies chest CT scans from four distinct medical centers. To address multi-source variability, we employ the Spatial-Slice Feature Learning (SSFL) framework with Kernel-Density-based Slice Sampling (KDS). Our preprocessing pipeline combines lung region extraction, quality control, and adaptive slice sampling to select eight representative slices per scan. We compare EfficientNet and Swin Transformer architectures on the validation set. The EfficientNet model achieves an F1-score of 94.68%, compared to the Swin Transformer's 93.34%. The results demonstrate the effectiveness of our KDS-based pipeline on multi-source data and highlight the importance of dataset balance in multi-institutional medical imaging evaluation.

Estimating brain age from structural MRI has emerged as a powerful tool for characterizing normative and pathological aging. In this work, we explore contrastive learning as a scalable and robust alternative to supervised approaches for brain age estimation. We introduce a novel contrastive loss function, $\mathcal{L}^{exp}$, and evaluate it across multiple public neuroimaging datasets comprising over 20,000 scans. Our experiments reveal four key findings. First, scaling pre-training on diverse, multi-site data consistently improves generalization performance, cutting external mean absolute error (MAE) nearly in half. Second, $\mathcal{L}^{exp}$ is robust to site-related confounds, maintaining low scanner-predictability as training size increases. Third, contrastive models reliably capture accelerated aging in patients with cognitive impairment and Alzheimer's disease, as shown through brain age gap analysis, ROC curves, and longitudinal trends. Lastly, unlike supervised baselines, $\mathcal{L}^{exp}$ maintains a strong correlation between brain age accuracy and downstream diagnostic performance, supporting its potential as a foundation model for neuroimaging. These results position contrastive learning as a promising direction for building generalizable and clinically meaningful brain representations.

Vision Transformers (ViTs) have gained significant popularity in the natural image domain but have been less successful in 3D medical image segmentation. Nevertheless, 3D ViTs are particularly interesting for large medical imaging volumes due to their efficient self-supervised training within the masked autoencoder (MAE) framework, which enables the use of imaging data without the need for expensive manual annotations. intracranial arterial calcification (IAC) is an imaging biomarker visible on routinely acquired CT scans linked to neurovascular diseases such as stroke and dementia, and automated IAC quantification could enable their large-scale risk assessment. We pre-train ViTs with MAE and fine-tune them for IAC segmentation for the first time. To develop our models, we use highly heterogeneous data from a large clinical trial, the third International Stroke Trial (IST-3). We evaluate key aspects of MAE pre-trained ViTs in IAC segmentation, and analyse the clinical implications. We show: 1) our calibrated self-supervised ViT beats a strong supervised nnU-Net baseline by 3.2 Dice points, 2) low patch sizes are crucial for ViTs for IAC segmentation and interpolation upsampling with regular convolutions is preferable to transposed convolutions for ViT-based models, and 3) our ViTs increase robustness to higher slice thicknesses and improve risk group classification in a clinical scenario by 46%. Our code is available online.

Vision transformers (ViTs) have rapidly gained prominence in medical imaging tasks such as disease classification, segmentation, and detection due to their superior accuracy compared to conventional deep learning models. However, due to their size and complex interactions via the self-attention mechanism, they are not well understood. In particular, it is unclear whether the representations produced by such models are semantically meaningful. In this paper, using a projected gradient-based algorithm, we show that their representations are not semantically meaningful and they are inherently vulnerable to small changes. Images with imperceptible differences can have very different representations; on the other hand, images that should belong to different semantic classes can have nearly identical representations. Such vulnerability can lead to unreliable classification results; for example, unnoticeable changes cause the classification accuracy to be reduced by over 60\%. %. To the best of our knowledge, this is the first work to systematically demonstrate this fundamental lack of semantic meaningfulness in ViT representations for medical image classification, revealing a critical challenge for their deployment in safety-critical systems.

Low-dose computed tomography (LDCT) imaging employed in lung cancer screening (LCS) programs is increasing in uptake worldwide. LCS programs herald a generational opportunity to simultaneously detect cancer and non-cancer-related early-stage lung disease. Yet these efforts are hampered by a shortage of radiologists to interpret scans at scale. Here, we present TANGERINE, a computationally frugal, open-source vision foundation model for volumetric LDCT analysis. Designed for broad accessibility and rapid adaptation, TANGERINE can be fine-tuned off the shelf for a wide range of disease-specific tasks with limited computational resources and training data. Relative to models trained from scratch, TANGERINE demonstrates fast convergence during fine-tuning, thereby requiring significantly fewer GPU hours, and displays strong label efficiency, achieving comparable or superior performance with a fraction of fine-tuning data. Pretrained using self-supervised learning on over 98,000 thoracic LDCTs, including the UK's largest LCS initiative to date and 27 public datasets, TANGERINE achieves state-of-the-art performance across 14 disease classification tasks, including lung cancer and multiple respiratory diseases, while generalising robustly across diverse clinical centres. By extending a masked autoencoder framework to 3D imaging, TANGERINE offers a scalable solution for LDCT analysis, departing from recent closed, resource-intensive models by combining architectural simplicity, public availability, and modest computational requirements. Its accessible, open-source lightweight design lays the foundation for rapid integration into next-generation medical imaging tools that could transform LCS initiatives, allowing them to pivot from a singular focus on lung cancer detection to comprehensive respiratory disease management in high-risk populations.

Accurate preoperative assessment of lymph node metastasis (LNM) and overall survival (OS) status is essential for patients with locally advanced gastric cancer receiving neoadjuvant chemotherapy, providing timely guidance for clinical decision-making. However, current approaches to evaluate LNM and OS have limited accuracy. In this study, we used longitudinal CT images from 1,021 patients with locally advanced gastric cancer to develop and validate a multitask deep learning model, named co-attention tri-oriented spatial Mamba (CTSMamba), to simultaneously predict LNM and OS. CTSMamba was trained and validated on 398 patients, and the performance was further validated on 623 patients at two additional centers. Notably, CTSMamba exhibited significantly more robust performance than a clinical model in predicting LNM across all of the cohorts. Additionally, integrating CTSMamba survival scores with clinical predictors further improved personalized OS prediction. These results support the potential of CTSMamba to accurately predict LNM and OS from longitudinal images, potentially providing clinicians with a tool to inform individualized treatment approaches and optimized prognostic strategies. CTSMamba is a multitask deep learning model trained on longitudinal CT images of neoadjuvant chemotherapy-treated locally advanced gastric cancer that accurately predicts lymph node metastasis and overall survival to inform clinical decision-making. This article is part of a special series: Driving Cancer Discoveries with Computational Research, Data Science, and Machine Learning/AI.