Focused cardiac ultrasound (FoCUS) is a valuable point-of-care method for evaluating cardiovascular structures and function, but its scope is limited by equipment and operator's experience, resulting in primarily qualitative 2D exams. This study presents a novel framework to automatically estimate the 3D spatial relationship between standard FoCUS views. The proposed framework uses a multi-view U-Net-like fully convolutional neural network to regress line-based heatmaps representing the most likely areas of intersection between input images. The lines that best fit the regressed heatmaps are then extracted, and a system of nonlinear equations based on the intersection between view triplets is created and solved to determine the relative 3D pose between all input images. The feasibility and accuracy of the proposed pipeline were validated using a novel realistic in silico FoCUS dataset, demonstrating promising results. Interestingly, as shown in preliminary experiments, the estimation of the 2D images' relative poses enables the application of 3D image analysis methods and paves the way for 3D quantitative assessments in FoCUS examinations.

Abstract Magnetic resonance (MR) imaging is widely used for assessing infant head and brain development and for diagnosing pathologies. The main goal of this work is the development of a segmentation framework to create patient-specific head and brain anatomical models from MR images for clinical evaluation. The proposed strategy consists of a fusion-based Deep Learning (DL) approach that combines the information of different image sequences within the MR acquisition protocol, including the axial T1w, sagittal T1w, and coronal T1w after contrast. These image sequences are used as input for different fusion encoder–decoder network architectures based on the well-established U-Net framework. Specifically, three different fusion strategies are proposed and evaluated, namely early, intermediate, and late fusion. In the early fusion approach, the images are integrated at the beginning of the encoder–decoder architecture. In the intermediate fusion strategy, each image sequence is processed by an independent encoder, and the resulting feature maps are then jointly processed by a single decoder. In the late fusion method, each image is individually processed by an encoder–decoder, and the resulting feature maps are then combined to generate the final segmentations. A clinical in-house dataset consisting of 19 MR scans was used and divided into training, validation, and testing sets, with 3 MR scans defined as a fixed validation set. For the remaining 16 MR scans, a cross-validation approach was adopted to assess the performance of the methods. The training and testing processes were carried out with a split ratio of 75% for the training set and 25% for the testing set. The results show that the early and intermediate fusion methodologies presented the better performance (Dice coefficient of 97.6 ± 1.5% and 97.3 ± 1.8% for the head and Dice of 94.5 ± 1.7% and 94.8 ± 1.8% for the brain, respectively), whereas the late fusion method generated slightly worst results (Dice of 95.5 ± 4.4% and 93.8 ± 3.1% for the head and brain, respectively). Nevertheless, the volumetric analysis showed that no statistically significant differences were found between the volumes of the models generated by all the segmentation strategies and the ground truths. Overall, the proposed frameworks demonstrate accurate segmentation results and prove to be feasible for anatomical model analysis in clinical practice.

Objective - Medical image segmentation is essential for several clinical tasks, including diagnosis, surgical and treatment planning, and image-guided interventions. Deep Learning (DL) methods have become the state-of-the-art for several image segmentation scenarios. However, a large and well-annotated dataset is required to effectively train a DL model, which is usually difficult to obtain in clinical practice, especially for 3D images. Methods - In this paper, we proposed Deep-DM, a learning-guided deformable model framework for 3D medical imaging segmentation using limited training data. In the proposed method, an energy function is learned by a Convolutional Neural Network (CNN) and integrated into an explicit deformable model to drive the evolution of an initial surface towards the object to segment. Specifically, the learning-based energy function is iteratively retrieved from localized anatomical representations of the image containing the image information around the evolving surface at each iteration. By focusing on localized regions of interest, this representation excludes irrelevant image information, facilitating the learning process. Results and conclusion - The performance of the proposed method is demonstrated for the tasks of left ventricle and fetal head segmentation in ultrasound, left atrium segmentation in Magnetic Resonance, and bladder segmentation in Computed Tomography, using different numbers of training volumes in each study. The results obtained showed the feasibility of the proposed method to segment different anatomical structures in different imaging modalities. Moreover, the results also showed that the proposed approach is less dependent on the size of the training dataset in comparison with state-of-the-art DL-based segmentation methods, outperforming them for all tasks when a low number of samples is available. Significance - Overall, by offering a more robust and less data-intensive approach to accurately segmenting anatomical structures, the proposed method has the potential to enhance clinical tasks that require image segmentation strategies.

The Digital Twin approach has increased in interest in recent years. Without well defined specifications, it is common for different researchers to use different approaches. Digital Twin concept emerged to support Industry 4.0, so it is of utmost importance to specify a best methodology, and tools, for its implementation for industrial use cases. This work proposes a Digital Twin architecture that follows manufacturing-centric standards of an industrial prototype testing station, for a new car infotainment system. Testing is performed by low-cost Software-Defined Radio equipment that aims to replace expensive metrological equipment. Moreover, an environment sensing device is also used to monitor the physical environment around prototype. We present the development of a solution that allows manage hardware processing, real-time monitoring of machine states in the virtual environment, and control of the overall system through a logical sequence supported by its Digital Twin. With the implementation of a standard like ISO23247, helps in the identified problem of having various Digital Twins and improves an interaction with test equipment. All steps made for the development of our architecture approach, as well as some results, are shown and explained for our use case.

Flexibility and speed in the development of new industrial machines are essential factors for the success of capital goods industries. When assembling a printed circuit board (PCB), since all the components are surface mounted devices (SMD), the whole process is automatic. However, in many PCBs, it is necessary to place components that are not SMDs, called pin through hole components (PTH), having to be inserted manually, which leads to delays in the production line. This work proposes and validates a prototype work cell based on a collaborative robot and vision systems whose objective is to insert these components in a completely autonomous or semi-autonomous way. Different tests were made to validate this work cell, showing the correct implementation and the possibility of replacing the human worker on this PCB assembly task.

Pectus excavatum (PE) is the most common abnormality of the thoracic cage, whose severity is evaluated by extracting three indices (Haller, correction and asymmetry) from computed tomography (CT) images. To date, this analysis is performed manually, which is tedious and prone to variability. In this paper, a fully automatic framework for PE severity quantification from CT images is proposed, comprising three steps: (1) identification of the sternum's greatest depression point; (2) detection of 8 anatomical keypoints relevant for severity assessment; and (3) measurements' geometric regularization and extraction. The first two steps rely on heatmap regression networks based on the Unet++ architecture, including a novel variant adapted to predict 1D confidence maps. The framework was evaluated on a database with 269 CTs. For comparative purposes, intra-observer, inter-observer and intra-patient variability of the estimated indices were analyzed in a subset of patients. The developed system showed a good agreement with the manual approach (a mean relative absolute error of 4.41%, 5.22% and 1.86% for the Haller, correction, and asymmetry indices, respectively), with limits of agreement comparable to the inter-observer variability. In the intra-patient analysis, the proposed framework outperformed the expert, showing a higher reproducibility between indices extracted from distinct CTs of the same patient. Overall, these results support the feasibility of the developed framework for the automatic, accurate and reproducible quantification of PE severity in a clinical context.

Shared autonomous vehicles (SAV) numbers are going to increase over the next years. The absence of human driver will create a new paradigm for in-car safety. This paper addresses the problem, presenting a monitoring system capable of estimating the state of the car interior, namely the presence of damage, dirt and stains. We propose the use of Semantic Segmentation methods to perform appropriate pixel-wise classification of certain textures found in the car's cabin as defect classes. Two methods, U-Net and DeepLabV3+, were trained and tested for different hiper-parameter and ablation scenarios, using RGB images. To be able to test and validate these approaches an In-car dataset was created, comprised by 1861 samples from 78 cars, and than splitted in 1303 train, 186 validation and 372 test RGB images. DeepLabV3+ showed promissing results, achieving an average accuracy for good, damage, stain and dirt of 77.17%, 58.60%, 65.81% and 68.82%, respectively.

Landmark labeling in 3D head surfaces is an important and routine task in clinical practice to evaluate head shape, namely to analyze cranial deformities or growth evolution. However, manual labeling is still applied, being a tedious and time-consuming task, highly prone to intra-/inter-observer variability, and can mislead the diagnose. Thus, automatic methods for anthropometric landmark detection in 3D models have a high interest in clinical practice. In this paper, a novel framework is proposed to accurately detect landmarks in 3D infant's head surfaces. The proposed method is divided into two stages: (i) 2D representation of the 3D head surface; and (ii) landmark detection through a deep learning strategy. Moreover, a 3D data augmentation method to create shape models based on the expected head variability is proposed. The proposed framework was evaluated in synthetic and real datasets, achieving accurate detection results. Furthermore, the data augmentation strategy proved its added value, increasing the method's performance. Overall, the obtained results demonstrated the robustness of the proposed method and its potential to be used in clinical practice for head shape analysis.

Over the years, medical image tracking has gained considerable attention from both medical and research communities due to its widespread utility in a multitude of clinical applications, from functional assessment during diagnosis and therapy planning to structure tracking or image fusion during image-guided interventions. Despite the ever-increasing number of image tracking methods available, most still consist of independent implementations with specific target applications, lacking the versatility to deal with distinct end-goals without the need for methodological tailoring and/or exhaustive tuning of numerous parameters. With this in mind, we have developed the medical image tracking toolbox (MITT)-a software package designed to ease customization of image tracking solutions in the medical field. While its workflow principles make it suitable to work with 2-D or 3-D image sequences, its modules offer versatility to set up computationally efficient tracking solutions, even for users with limited programming skills. MITT is implemented in both C/C++ and MATLAB, including several variants of an object-based image tracking algorithm and allowing to track multiple types of objects (i.e., contours, multi-contours, surfaces, and multi-surfaces) with several customization features. In this paper, the toolbox is presented, its features discussed, and illustrative examples of its usage in the cardiology field provided, demonstrating its versatility, simplicity, and time efficiency.

A novel semi-automatic algorithm for aortic valve (AV) wall segmentation is presented for 3D transesophageal echocardiography (TEE) datasets. The proposed methodology uses a 3D cylindrical formulation of the B-spline Explicit Active Surfaces (BEAS) framework in a dual-stage energy evolution process, comprising a threshold-based and a localized region-based stage. Hereto, intensity and shape-based features are combined to accurately delineate the AV wall from the ascending aorta (AA) to the left ventricular outflow tract (LVOT). Shape-prior information is included using a profile-based statistical shape model (SSM), and embedded in BEAS through two novel regularization terms: one confining the segmented AV profiles to shapes seen in the SSM (hard regularization) and another penalizing according to the profile's degree of likelihood (soft regularization). The proposed energy functional takes thus advantage of the intensity data in regions with strong image content, while complementing it with shape knowledge in regions with nearly absent image data. The proposed algorithm has been validated in 20 3D-TEE datasets with both stenotic and non-stenotic valves. It was shown to be accurate, robust and computationally efficient, taking less than 1 second to segment the AV wall from the AA to the LVOT with an average accuracy of 0.78 mm. Semi-automatically extracted measurements at four relevant anatomical levels (LVOT, aortic annulus, sinuses of Valsalva and sinotubular junction) showed an excellent agreement with experts' ones, with a higher reproducibility than manually-extracted measures.