Abstract
This paper studies the use of deep convolutional neural networks to segment heart sounds into their main components. The proposed methods are based on the adoption of a deep convolutional neural network architecture, which is inspired by similar approaches used for image segmentation. Different temporal modeling schemes are applied to the output of the proposed neural network, which induce the output state sequence to be consistent with the natural sequence of states within a heart sound signal (S1, systole, S2, diastole). In particular, convolutional neural networks are used in conjunction with underlying hidden Markov models and hidden semi-Markov models to infer emission distributions. The proposed approaches are tested on heart sound signals from the publicly available PhysioNet dataset, and they are shown to outperform current state-of-the-art segmentation methods by achieving an average sensitivity of 93.9 and an average positive predictive value of 94 in detecting S1 and S2 sounds.

Abstract
This work addresses the automatic segmentation of the joint capsule in ultrasound images of the metacarpophalangeal joint using an adapted version of the well known UNet model. These images are used in the diagnosis of rheumatic diseases, one of the main causes of impairment and pain in developed countries. The identification of the joint capsule gives important clues about the presence or Rheumatoid Arthritis. This structure can be used to extract metrics to help quantify the disease stage and progression. The solution proposed here has the potential to reduce the burden on the radiologists as well as the subjectivity of the diagnosis by providing quantitative measurements, such as the synovitis area. The proposed approach was compared with two other works present in the literature. Results show that our solution outperforms the two reference methods with 90% of the joint capsules identified with a DICE higher than 0.67.

Abstract
Rheumatic heart disease remains a major burden in the developing countries. The World Heart Federation proposed guidelines for the echocardiographic detection of the disease, in which the mitral leaflets' morphology assessment is a key indicator. The drawback is that these guidelines are dependent on the clinician experience. To overcome this limitation, we propose an automatic segmentation of the mitral leaflets using a new method based on convolutional neural network, specifically the UNet architecture. The results indicate a median DICE coefficient of 0.74 in PLAX and 0.79 in A4C for the anterior mitral leaflet segmentation, while median DICE of 0.60 in PLAX and 0.69 A4C are met for the posterior leaflet. A visual evaluation of this segmentation approach by two cardiologists is in line with the numerical results. The false detection due to overestimation and artifacts remains an issue to be addressed in the future.

Abstract
Combining Phonocardiography (PCG) and Electrocardiography (ECG) data has been recognized within the state-of-the-art as of added value for enhanced cardiovascular assessment. However, multiple aspects of ECG data acquisition in a stethoscope form factor remain unstudied, and existing devices typically enforce a substantial change into routine clinical auscultation procedures, with predictably low technology acceptance. As such, in this paper, we present a novel approach to ECG data acquisition throughout the five main cardiac auscultation points, and that intends to be incorporated in a commonly used electronic stethoscope. Therefore, it enables analysis and acquisition of both PCG and ECG signals in a single pass. We describe the development, experimental evaluation, and comparison of the ECG signals obtained using our proposed approach and a gold standard medical device, through metrics that allow the evaluation of morphological similarities. Results point to a high correlation between the two evaluated setups, thus supporting the idea of meaningfully collecting ECG data along medical auscultation points with the proposed form factor. Moreover, this work has led us to conclude that for the studied population, signals acquired on focuses F1, F2, and F3 are usually highly correlated with leads V1 and V2 of the standard ECG medical recording procedure.

Abstract
The application of pattern recognition techniques to data collected from accelerometers available in off-the-shelf devices, such as smartphones, allows for the automatic recognition of activities of daily living (ADLs). This data can be used later to create systems that monitor the behaviors of their users. The main contribution of this paper is to use artificial neural networks (ANN) for the recognition of ADLs with the data acquired from the sensors available in mobile devices. Firstly, before ANN training, the mobile device is used for data collection. After training, mobile devices are used to apply an ANN previously trained for the ADLs' identification on a less restrictive computational platform. The motivation is to verify whether the overfitting problem can be solved using only the accelerometer data, which also requires less computational resources and reduces the energy expenditure of the mobile device when compared with the use of multiple sensors. This paper presents a method based on ANN for the recognition of a defined set of ADLs. It provides a comparative study of different implementations of ANN to choose the most appropriate method for ADLs identification. The results show the accuracy of 85.89% using deep neural networks (DNN).

Abstract
In this work, we present a method to separate aortic (A2) and pulmonary (P2) components from second heart sounds (S2). The proposed approach captures the different dynamical behavior of A2 and P2 components via a joint Gaussian mixture model, which is then used to perform separation via a closed-form conditional mean estimator.The proposed approach is tested over synthetic heart sounds and it is shown guarantee a reduction of approximately 25% of the normalized root mean-squared error incurred in signal separation, with respect to a previously presented approach in the literature.