Breast cancer is the most frequently diagnosed malignancy among women worldwide and a leading cause of cancer-related mortality. Dynamic contrast-enhanced magnetic resonance imaging plays a central role in tumor characterization and treatment monitoring, particularly in patients receiving neoadjuvant chemotherapy. However, existing artificial intelligence models for breast magnetic resonance imaging are often developed using single-center data and evaluated using aggregate performance metrics, limiting their generalizability and obscuring potential performance disparities across demographic subgroups. The MAMA-MIA Challenge was designed to address these limitations by introducing a large-scale benchmark that jointly evaluates primary tumor segmentation and prediction of pathologic complete response using pre-treatment magnetic resonance imaging only. The training cohort comprised 1,506 patients from multiple institutions in the United States, while evaluation was conducted on an external test set of 574 patients from three independent European centers to assess cross-continental and cross-institutional generalization. A unified scoring framework combined predictive performance with subgroup consistency across age, menopausal status, and breast density. Twenty-six international teams participated in the final evaluation phase. Results demonstrate substantial performance variability under external testing and reveal trade-offs between overall accuracy and subgroup fairness. The challenge provides standardized datasets, evaluation protocols, and public resources to promote the development of robust and equitable artificial intelligence systems for breast cancer imaging.

Automated segmentation of liver lesions on non-contrast computed tomography (NCCT) is clinically important but fundamentally challenging, particularly in low-resource settings across Africa and Asia where contrast agents are frequently unavailable. Progress has been limited by the absence of annotated NCCT benchmarks. Here we describe the TriALS challenge for automated liver lesion segmentation under contrast-limited conditions, supported by a multi-centre dataset of 150 cases with four-phase CT acquisitions (600 volumes) from Egyptian and Chinese institutions. Algorithms were evaluated on 70 cases from three institutions, including an independent external cohort. The top-performing method achieved a mean venous-phase Dice of 0.754, consistent with human-level performance, yet dropped to 0.57 on NCCT. On external validation, the leading method outperformed off-the-shelf models by up to 28

Contrast-enhanced mammography (CEM) offers improved breast cancer diagnosis by enhancing vascular contrast uptake. However, the development of reliable deep learning-based computer-aided detection (CAD) systems for CEM is hindered by limited data availability. This paper introduces ELK (Enhanced Learning through cross-modal Knowledge transfer), a deep learning pipeline designed to adapt large pre-trained models into a target limited data-volume population by leveraging synthetic data augmentation. Specifically, we adapt a detection model pretrained on digital breast tomosynthesis (DBT) and digital mammography data into a target CEM population using diffusion models to generate high-resolution, realistic synthetic lesions, preserving the visual integrity of CEM images. To assess the efficacy of our synthetic lesions, we compare the detection performance of a pretrained Faster R-CNN detector fine-tuned using only real images, synthetic images, and a combination of both. Our approach improves mean sensitivity by 4 https://github.com/Likalto4/CEM-Detect .

X-ray phase contrast is a promising breast image modality. This technique is capable of simultaneously providing three types of images: absorption, differential phase contrast (DPC) and dark-field (DF) images, allowing to obtain complementary information from each one. However, the Talbot-Lau interferometer, the device used to acquire this type of images, can yield Moiré artifacts in the corresponding images. The aim of this work is to introduce a deep learning approach, using a generative adversarial network, in particular the pix2pix neural network, to reduce Moiré artifacts efficiently. Our approach was tested using simulated DPC and DF images obtained from the INbreast dataset. Moiré and mammography-based images are fused using a novel approach which aims to eliminate the bias yielded by the traditional one. Results shows a significant image quality improvement for the DF dataset, reaching a structural similarity (SSIM) index of SSIM=0.96± 0.02 , in average, after applying the neural network. However, under the same training conditions, the denoised DPC images do not show such a clear improvement, yielding checkerboard and discontinuity artifacts.

Breast cancer is one of the most common causes of death among women worldwide. Early detection helps in reducing the number of deaths. Automated 3D Breast Ultrasound (ABUS) is a newer approach for breast screening, which has many advantages over handheld mammography such as safety, speed, and higher detection rate of breast cancer. Tumor detection, segmentation, and classification are key components in the analysis of medical images, especially challenging in the context of 3D ABUS due to the significant variability in tumor size and shape, unclear tumor boundaries, and a low signal-to-noise ratio. The lack of publicly accessible, well-labeled ABUS datasets further hinders the advancement of systems for breast tumor analysis. Addressing this gap, we have organized the inaugural Tumor Detection, Segmentation, and Classification Challenge on Automated 3D Breast Ultrasound 2023 (TDSC-ABUS2023). This initiative aims to spearhead research in this field and create a definitive benchmark for tasks associated with 3D ABUS image analysis. In this paper, we summarize the top-performing algorithms from the challenge and provide critical analysis for ABUS image examination. We offer the TDSC-ABUS challenge as an open-access platform at https://tdsc-abus2023.grand-challenge.org/ to benchmark and inspire future developments in algorithmic research.

Purpose:Accurate simulation of breast tissue deformation is essential for reliable image registration between 3D imaging modalities and 2D mammograms, where compression significantly alters tissue geometry. Although finite element analysis (FEA) provides high-fidelity modeling, it is computationally intensive and not well suited for rapid simulations. To address this, the physics-based graph neural network (PhysGNN) has been introduced as a computationally efficient approximation model trained on FEA-generated deformations. We extend prior work by evaluating the performance of PhysGNN on new digital breast phantoms and assessing the impact of training on multiple phantoms. Approach:PhysGNN was trained on both single-phantom (per-geometry) and multiphantom (multigeometry) datasets generated from incremental FEA simulations. The digital breast phantoms represent the uncompressed state, serving as input geometries for predicting compressed configurations. A leave-one-deformation-out evaluation strategy was used to assess predictive performance under compression. Results:Training on new digital phantoms confirmed the model's robust performance, though with some variability in prediction accuracy reflecting the diverse anatomical structures. Multiphantom training further enhanced this robustness and reduced prediction errors. Conclusions:PhysGNN offers a computationally efficient alternative to FEA for simulating breast compression. The results showed that model performance remains robust when trained per-geometry, and further demonstrated that multigeometry training enhances predictive accuracy and robustness for the geometries included in the training set. This suggests a strong potential path toward developing reliable models for generating compressed breast volumes, which could facilitate image registration and algorithm development.

Breast cancer is a heterogeneous disease requiring accurate imaging interpretation for diagnosis and treatment planning. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) plays a critical role in this setting by capturing tumour vascular dynamics. In this work, we present our contribution to the MAMA-MIA Challenge, which targets primary tumour segmentation (Task 1) and treatment response prediction from pre-treatment DCE-MRI (Task 2). For Task 1, we adopt the nnU-Net framework, combined with a tailored preprocessing pipeline designed to enhance generalisation and fairness across heterogeneous imaging centres and scanner vendors. Our pipeline focuses on extracting meaningful dynamic information from three representative DCE-MRI phases, used as multi-channel input to guide the model in learning temporal enhancement behaviour around the tumour. For Task 2, we extract temporal dynamics features from all phases and combine them with radiomics-based shape descriptors, derived from the predicted tumour mask, to train an XGBoost classifier for (pathological complete response (pCR)) prediction. Our work aligns with the broader goals of developing robust, generalisable, and equitable AI tools for breast cancer imaging using real-world, multi-centre MRI data.

The application of generative models for the synthetic expansion of medical image datasets has been explored in recent research, offering new opportunities to combat data scarcity and improve computer-aided diagnosis (CAD) systems. In this study, we explore the use of a generative framework based on neural cellular automata (NCA) for the synthesis of mammographic images. While UNet and similar models focus on global coherence, NCAs inherently prioritize local dependencies through repeated, neighborhood-based updates. This feature facilitates capturing local details such as microcalcifications, subtle spiculations, and fine tissue textures, crucial for accurate diagnosis with mammographic images. To our knowledge, this is the first implementation of this type of framework for generating synthetic mammograms. The framework was adapted to process images from the VinDr-Mammo dataset, supporting both full mammograms and extracted findings in a highly configurable workflow. We integrated a custom breast density classifier to evaluate the impact of the addition of synthetic images on downstream classification performance. Qualitative and quantitative evaluations demonstrate that the adapted model is capable of generating perceptually realistic mammograms, with Kernel Inception Distance scores between 0.015 and 0.032. The addition of synthetic data to the density classifier's training dataset improved macro F1-scores by up to 12% in underrepresented density classes across multiple dataset configurations, although the effects on binary classification metrics remained mixed. These findings suggest that neural cellular diffusion models hold promise for mammographic image synthesis and further work is warranted to validate these results in clinical settings and with additional datasets.

Breast compression simulation is essential for accurate image registration from 3D modalities to X-ray procedures like mammography. It accounts for tissue shape and position changes due to compression, ensuring precise alignment and improved analysis. Although Finite Element Analysis (FEA) is reliable for approximating soft tissue deformation, it struggles with balancing accuracy and computational efficiency. Recent studies have used data-driven models trained on FEA results to speed up tissue deformation predictions. We propose to explore Physics-based Graph Neural Networks (PhysGNN) for breast compression simulation. PhysGNN has been used for data-driven modelling in other domains, and this work presents the first investigation of their potential in predicting breast deformation during mammographic compression. Unlike conventional data-driven models, PhysGNN, which incorporates mesh structural information and enables inductive learning on unstructured grids, is well-suited for capturing complex breast tissue geometries. Trained on deformations from incremental FEA simulations, PhysGNN's performance is evaluated by comparing predicted nodal displacements with those from finite element (FE) simulations. This deep learning (DL) framework shows promise for accurate, rapid breast deformation approximations, offering enhanced computational efficiency for real-world scenarios.

IntroductionHeart disease is a major cause of mortality in aging dogs and cats, with cardiomegaly being the most frequent radiographic finding. While deep learning methods have shown potential in detecting and quantifying cardiomegaly, their integration into clinical veterinary practice remains limited due to challenges in interpretability and workflow alignment.MethodsWe developed a deep learning framework for the automatic estimation of Vertebral Heart Size (VHS) and Cardiothoracic Ratio (CTR) from thoracic radiographs of dogs and cats. A diverse dataset collected from two veterinary institutions was used. Segmentation of cardiac and thoracic anatomical regions was performed using Mask R-CNN, followed by automatic measurement of VHS and CTR. Model performance was evaluated against expert radiologist annotations.ResultsThe proposed framework demonstrated strong agreement with manual evaluations. Pearson correlation coefficients reached 0.922 for VHS and 0.933 for CTR, with regression slopes close to unity and minimal intercepts. The method was validated on both lateral and ventrodorsal projections, confirming its versatility across common clinical views.Discussion/conclusionThis work introduces an automated, robust approach for cardiac size assessment in dogs and cats. By supporting objective and reproducible measurements of VHS and CTR, the framework has potential to aid in the early detection and monitoring of heart disease, particularly in veterinary settings with limited access to specialized radiology expertise.

Integrating 2D mammography with 3D magnetic resonance imaging (MRI) is crucial for improving breast cancer diagnosis and treatment planning. However, this integration is challenging due to differences in imaging modalities and the need for precise tissue segmentation and alignment. This paper addresses these challenges by enhancing biomechanical breast models in two main aspects: improving tissue identification using nnU-Net segmentation models and evaluating finite element (FE) biomechanical solvers, specifically comparing NiftySim and FEBio. We performed a detailed six-class segmentation of breast MRI data using the nnU-Net architecture, achieving Dice Coefficients of 0.94 for fat, 0.88 for glandular tissue, and 0.87 for pectoral muscle. The overall foreground segmentation reached a mean Dice Coefficient of 0.83 through an ensemble of 2D and 3D U-Net configurations, providing a solid foundation for 3D reconstruction and biomechanical modeling. The segmented data was then used to generate detailed 3D meshes and develop biomechanical models using NiftySim and FEBio, which simulate breast tissue's physical behaviors under compression. Our results include a comparison between NiftySim and FEBio, providing insights into the accuracy and reliability of these simulations in studying breast tissue responses under compression. The findings of this study have the potential to improve the integration of 2D and 3D imaging modalities, thereby enhancing diagnostic accuracy and treatment planning for breast cancer.

Deep learning models are the state-of-the-art for most medical imaging application, including mammography. However, large amounts of data are generally required for their usage. Radiomic analysis has shown the potential to improve clinical decision support systems for small datasets. One of the challenges facing the clinical implementation of radiomics is reproducibility. Our goal is to show that assessing radiomic features uncertainty can improve the robustness and performance of radiomic-based prediction models for contrast-enhanced digital mammographic images. Additionally, we propose the use of a pretrained tomosynthesis (DBT) lesion detection model as feature extractor for the boosting of the prediction framework. The prediction goal was the immunohistochemical status of breast cancer in 33 patients. We assessed two sources of uncertainty: misalignment between the subtracted images and region-of-interest delineation variability. Including uncertainties in the training step improved the performance of the prediction models, and the use of the DBT lesion detection model to boost the prediction improved the overall radiomic model performance for PR, ER and Ki67 receptors.

This work presents a framework for lesion segmentation on 3D Automated Breast Ultrasound. The method consists on the implementation of a state-of-the-art foundation model for 2D segmentation pipeline called Segment anything model (SAM), adapted for 3D segmentation through a probabilistic refinement technique. The presented method obtained second place in the segmentation task of the 2023 MICCAI Challenge on Tumor Detection, Segmentation and Classification Challenge on Automated 3D Breast Ultrasound (TDSC-ABUS 2023), being the most robust approach in terms of the Hausdorff distance. The paper describes the approaches developed for the challenge submission as well as suggestions for future improvement.

Breast density has demonstrated to be an important risk factor for the development of breast cancer and, therefore, different fully automated density assessment tools have been introduced to obtain quantitative glandular tissue measures. Density maps (DMs) provide local tissue information, representing the amount of glandular tissue between the image receptor and the x-ray source at every pixel in the image. Usually, DMs are obtained from for processing, i.e. raw, mammograms. This fact could become a tricky problem because this type of images are not preserved in the clinical setting. The aim of this work is to introduce a deep learning based framework to synthesize glandular tissue DMs from for presentation mammograms. First, the breast region is located using a dedicated object detector network. Next, a generative adversarial network is used to obtain synthetic density maps, that are useful to evaluate not only the glandular tissue distribution but also the total glandular tissue volume within the breast. Results show that synthetic DMs obtain a structural similarity index of SSIM = 0.93 +/- 0.06 with respect to real images. Similarly, shared information between the real and synthetic images, computed using the histogram intersection, corresponds to HI = 0.84 +/- 0.10, while the average pixel difference represents only 3.85 +/- 2.78 % of breast thickness. Furthermore, glandular tissue volume (GTV) obtained from synthetic density map show a strong correlation with the value provided by the real one (rho= 0.89 [C.I 0.87 - 0.91]). In conclusion, generative deep learning models can be useful to evaluate breast composition, from local to global tissue distribution.

Segmentation of the fetal and maternal structures, particularly intrapartum ultrasound imaging as advocated by the International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) for monitoring labor progression, is a crucial first step for quantitative diagnosis and clinical decision-making. This requires specialized analysis by obstetrics professionals, in a task that i) is highly time- and cost-consuming and ii) often yields inconsistent results. The utility of automatic segmentation algorithms for biometry has been proven, though existing results remain suboptimal. To push forward advancements in this area, the Grand Challenge on Pubic Symphysis-Fetal Head Segmentation (PSFHS) was held alongside the 26th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2023). This challenge aimed to enhance the development of automatic segmentation algorithms at an international scale, providing the largest dataset to date with 5,101 intrapartum ultrasound images collected from two ultrasound machines across three hospitals from two institutions. The scientific community's enthusiastic participation led to the selection of the top 8 out of 179 entries from 193 registrants in the initial phase to proceed to the competition's second stage. These algorithms have elevated the state-of-the-art in automatic PSFHS from intrapartum ultrasound images. A thorough analysis of the results pinpointed ongoing challenges in the field and outlined recommendations for future work. The top solutions and the complete dataset remain publicly available, fostering further advancements in automatic segmentation and biometry for intrapartum ultrasound imaging.

Representational transfer from publicly available models is a promising technique for improving medical image classification, especially in long-tailed datasets with rare diseases. However, existing methods often overlook the frequency-dependent behavior of these models, thereby limiting their effectiveness in transferring representations and generalizations to rare diseases. In this paper, we propose FoPro-KD, a novel framework that leverages the power of frequency patterns learned from frozen pre-trained models to enhance their transferability and compression, presenting a few unique insights: 1) We demonstrate that leveraging representations from publicly available pre-trained models can substantially improve performance, specifically for rare classes, even when utilizing representations from a smaller pre-trained model. 2) We observe that pre-trained models exhibit frequency preferences, which we explore using our proposed Fourier Prompt Generator (FPG), allowing us to manipulate specific frequencies in the input image, enhancing the discriminative representational transfer. 3) By amplifying or diminishing these frequencies in the input image, we enable Effective Knowledge Distillation (EKD). EKD facilitates the transfer of knowledge from pre-trained models to smaller models. Through extensive experiments in long-tailed gastrointestinal image recognition and skin lesion classification, where rare diseases are prevalent, our FoPro-KD framework outperforms existing methods, enabling more accessible medical models for rare disease classification. Code is available at https://github.com/xmed-lab/FoPro-KD.