João Manuel Pedrosa

Abstract
The increase in complexity of deep learning models demands explanations that can be obtained with methods like Grad-CAM. This method computes an importance map for the last convolutional layer relative to a specific class, which is then upsampled to match the size of the input. However, this final step assumes that there is a spatial correspondence between the last feature map and the input, which may not be the case. We hypothesize that, for models with large receptive fields, the feature spatial organization is not kept during the forward pass, which may render the explanations devoid of meaning. To test this hypothesis, common architectures were applied to a medical scenario on the public VinDr-CXR dataset, to a subset of ImageNet and to datasets derived from MNIST. The results show a significant dispersion of the spatial information, which goes against the assumption of Grad-CAM, and that explainability maps are affected by this dispersion. Furthermore, we discuss several other caveats regarding Grad-CAM, such as feature map rectification, empty maps and the impact of global average pooling or flatten layers. Altogether, this work addresses some key limitations of Grad-CAM which may go unnoticed for common users, taking one step further in the pursuit for more reliable explainability methods.

Abstract
Arterial stiffness (AS) is a well-established predictor of cardiovascular events, including myocardial infarction and stroke. One of the most recognized methods for assessing AS is through arterial pulse wave velocity (aPWV), which provides valuable clinical insights into vascular health. However, its measurement typically requires specialized equipment, making it inaccessible in primary healthcare centers and low-resource settings. In this study, we developed and validated different machine learning models to estimate aPWV using common clinical markers routinely collected in standard medical examinations. Thus, we trained five regression models: Linear Regression, Polynomial Regression (PR), Gradient Boosting Regression, Support Vector Regression, and Neural Networks (NNs) on the EVasCu dataset, a cohort of apparently healthy individuals. A 10-fold cross-validation demonstrated that PR and NN achieved the highest predictive performance, effectively capturing nonlinear relationships in the data. External validation on two independent datasets, VascuNET (a healthy population) and ExIC-FEp (a cohort of cardiopathic patients), confirmed the robustness of PR and NN (R- (2)> 0.90) across different vascular conditions. These results indicate that by using easily accessible clinical variables and AI-driven insights, it is possible to develop a cost-effective tool for aPWV estimation, enabling early cardiovascular risk stratification in underserved and rural areas where specialized AS measurement devices are unavailable.

Abstract
Chest radiography (CXR) is one of the most used medical imaging modalities. Nevertheless, the interpretation of CXR images is time-consuming and subject to variability. As such, automated systems for pathology detection have been proposed and promising results have been obtained, particularly using deep learning. However, these tools suffer from poor explainability, which represents a major hurdle for their adoption in clinical practice. One proposed explainability method in CXR is through contrastive examples, i.e. by showing an alternative version of the CXR except without the lesion being investigated. While image-level normal/healthy image synthesis has been explored in literature, normal patch synthesis via inpainting has received little attention. In this work, a method to synthesize contrastive examples in CXR based on local synthesis of normal CXR patches is proposed. Based on a contextual attention inpainting network (CAttNet), an anatomically-guided inpainting network (AnaCAttNet) is proposed that leverages anatomical information of the original CXR through segmentation to guide the inpainting for a more realistic reconstruction. A quantitative evaluation of the inpainting is performed, showing that AnaCAttNet outperforms CAttNet (FID of 0.0125 and 0.0132 respectively). Qualitative evaluation by three readers also showed that AnaCAttNet delivers superior reconstruction quality and anatomical realism. In conclusion, the proposed anatomical segmentation module for inpainting is shown to improve inpainting performance.

Abstract
Objectives: The shape is commonly used to describe the objects. State-of-the-art algorithms in medical imaging are predominantly diverging from computer vision, where voxel grids, meshes, point clouds, and implicit surface models are used. This is seen from the growing popularity of ShapeNet (51,300 models) and Princeton ModelNet (127,915 models). However, a large collection of anatomical shapes (e.g., bones, organs, vessels) and 3D models of surgical instruments is missing. Methods: We present MedShapeNet to translate data-driven vision algorithms to medical applications and to adapt state-of-the-art vision algorithms to medical problems. As a unique feature, we directly model the majority of shapes on the imaging data of real patients. We present use cases in classifying brain tumors, skull reconstructions, multi-class anatomy completion, education, and 3D printing. Results: By now, MedShapeNet includes 23 datasets with more than 100,000 shapes that are paired with annotations (ground truth). Our data is freely accessible via a web interface and a Python application programming interface and can be used for discriminative, reconstructive, and variational benchmarks as well as various applications in virtual, augmented, or mixed reality, and 3D printing. Conclusions: MedShapeNet contains medical shapes from anatomy and surgical instruments and will continue to collect data for benchmarks and applications. The project page is: https://medshapenet.ikim.nrw/.

Abstract
Large-scale datasets are essential for training deep learning models in medical imaging. However, many of these datasets contain poor-quality images that can compromise model performance and clinical reliability. In this study, we propose a framework to detect non-compliant images, such as corrupted scans, incomplete thorax X-rays, and images of non-thoracic body parts, by leveraging contrastive learning for feature extraction and parametric or non-parametric scoring methods for out-ofdistribution ranking. Our approach was developed and tested on the CheXpert dataset, achieving an AUC of 0.75 in a manually labeled subset of 1,000 images, and further qualitatively and visually validated on the external PadChest dataset, where it also performed effectively. Our results demonstrate the potential of contrastive learning to detect non-compliant images in largescale medical datasets, laying the foundation for future work on reducing dataset pollution and improving the robustness of deep learning models in clinical practice.