Abstract
Interpretability is a fundamental property for the acceptance of machine learning models in highly regulated areas. Recently, deep neural networks gained the attention of the scientific community due to their high accuracy in vast classification problems. However, they are still seen as black-box models where it is hard to understand the reasons for the labels that they generate. This paper proposes a deep model with monotonic constraints that generates complementary explanations for its decisions both in terms of style and depth. Furthermore, an objective framework for the evaluation of the explanations is presented. Our method is tested on two biomedical datasets and demonstrates an improvement in relation to traditional models in terms of quality of the explanations generated. © Springer Nature Switzerland AG 2018.

Abstract
Two limitations hamper performance of deep architectures for classification and/or detection in medical imaging: (i) the small amount of available data, and (ii) the class imbalance scenario. While millions of labeled images are available today to build classification tools for natural scenes, the amount of available annotated data for automatic breast cancer screening is limited to a few thousand images, at best. We address these limitations with a method for data augmentation, based on the introduction of random elastic deformations on images of mammograms. We validate this method on three publicly available datasets. Our proposed Convolutional Neural Network (CNN) archi-tecture is trained for mass classification - in a conventional way - , and then used in the more interesting problem of mass detection in full mammograms by transforming the CNN into a Fully Convolutional Network (FCN). © 2018 IEEE.

Abstract
Facial recognition under uncontrolled acquisition environments faces major challenges that limit the deployment of real-life systems. The use of 2.5D information can be used to improve discriminative power of such systems in conditions where RGB information alone would fail. In this paper we propose a multimodal extension of a previous work, based on SIFT descriptors of RGB images, integrated with LBP information obtained from depth scans, modeled by an hierarchical framework motivated by principles of human cognition. The framework was tested on EURECOM dataset and proved that the inclusion of depth information improved significantly the results in all the tested conditions, compared to independent unimodal approaches.

Abstract
Infantile Hemangiomas (IH) make up the most common type of benign vascular tumors affecting children. They can grow for several months until beginning to involute. In present-day clinical practice there's no objective monitoring protocol. For more objective measures, an automatic evaluation system (CAD system) is needed to aid clinicians in assessing the effectiveness of a given patient's response to a treatment. One of the stages of these systems is the lesion segmentation. This work addresses the automatic segmentation of lesions in IH. Acknowledging that the methods in the literature for IH lesion segmentation lag behind the state-of-the-art in the image segmentation community, we conduct a comparison of various methodologies for the segmentation of the IH, including both shallow and deep methodologies. Acknowledging the lack of data in the field for a robust learning of deep models, we also evaluate transfer learning techniques to benefit from knowledge extracted in other skin lesions. The best results were obtained with the shortest path method and a multiscale convolutional neural network that merges two pipelines working at different scales. Although promising, the results put in evidence the need for better databases, collected under suitable acquisition protocols.

Abstract
Driver drowsiness is a major cause of road accidents, many of which result in fatalities. A solution to this problem is the inclusion of a drowsiness detector in vehicles to alert the driver if sleepiness is detected. To detect drowsiness, physiologic, behavioral (visual) and vehicle-based methods can be used, however, only measures that can be acquired non-intrusively are viable in a real life application. This work uses data from a real-road experiment with sleep deprived drivers to compare the performance of driver drowsiness detection using intrusive acquisition methods, namely electrooculogram (EOG), with camera based, non-intrusive, methods. A hybrid strategy, combining the described methods with electrocardiogram (ECG) measures, is also evaluated. Overall, the obtained results show that drowsiness detection performance is similar using non-intrusive camera based measures or intrusive EOG measures. The detection performance increases when combining two methods (ECG + visual) or (ECG + EOG).

Abstract
Currently the breakthroughs in most computer vision problems have been achieved by applying deep learning methods. The traditional methodologies that used to successfully discriminate the data features appear to be overwhelmed by the capabilities of learning of the deep network architectures. Nevertheless, many recent works choose to integrate the old handcrafted features into the deep convolutional networks to increase even more their impressive performance. In fingerprint recognition, the minutiae are specific points used to identify individuals and their extraction is a crucial module in a fingerprint recognition system. This can only be emphasized by the fact that the US Federal Bureau of Investigation (FBI) sets as a threshold for a positive identification a number of 8 common minutiae. Deep neural networks have been used to learn possible representations of fingerprint minutiae but, however surprisingly, in this paper it is shown that for now the best choice for an automatic minutiae extraction system is still the traditional road map. A comparison study was conducted with state-of-the-art methods and the best results were achieved by handcraft features. © Springer Nature Switzerland AG 2019.