Abstract
Deep Learning (DL) algorithms allow fast results with high accuracy in medical imaging analysis solutions. However, to achieve a desirable performance, they require large amounts of high quality data. Active Learning (AL) is a subfield of DL that aims for more efficient models requiring ideally fewer data, by selecting the most relevant information for training. CheXpert is a Chest X-Ray (CXR) dataset, containing labels for different pathologic findings, alongside a Support Devices (SD) label. The latter contains several misannotations, which may impact the performance of a pathology detection model. The aim of this work is the detection of SDs in CheXpert CXR images and the comparison of the resulting predictions with the original CheXpert SD annotations, using AL approaches. A subset of 10,220 images was selected, manually annotated for SDs and used in the experimentations. In the first experiment, an initial model was trained on the seed dataset (6,200 images from this subset). The second and third approaches consisted in AL random sampling and least confidence techniques. In both of these, the seed dataset was used initially, and more images were iteratively employed. Finally, in the fourth experiment, a model was trained on the full annotated set. The AL least confidence experiment outperformed the remaining approaches, presenting an AUC of 71.10% and showing that training a model with representative information is favorable over training with all labeled data. This model was used to obtain predictions, which can be useful to limit the use of SD mislabelled images in future models.

Abstract
Organizations are more frequently turning towards robotic process automation (RPA) as a solu-tion for employees to focus on higher complexity and more valuable tasks while delegating routine, monotonous and rule-based tasks to their digital colleagues. These software robots can handle various rule-based, digital, repetitive tasks. However, currently available process identi-fication methods must be qualified to select suitable automation processes accurately. Wrong process selection and failed attempts are often the origin of process automation's bad reputation within organizations and often result in the avoidance of this technology. As a result, in this research, a method for selecting processes for automation combining two multi-criteria decision -making techniques, 'Analytic Hierarchy Process (AHP) and Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), will be proposed, demonstrated, and evaluated. This study follows the Design Science Research Methodology (DSRM) and applies the proposed method for selecting processes for automation to a real-life scenario. The result will be a method to support the proper selection of business processes for automation, increasing the success of implementing RPA tools in an organization.

Abstract
Peripheral arterial disease (PAD) causes blockage of the arteries, altering the blood flow to the lower limbs. This blockage can cause the individual with PAD to feel severe pain in the lower limbs. The main contribution of this research is the discovery of a solution that allows the automatic detection of the onset of claudication based on data analysis from patients' smartphones. For the data-collection procedure, 40 patients were asked to walk with a smartphone on a thirty-meter path, back and forth, for six minutes. Each patient conducted the test twice on two different days. Several machine learning models were compared to detect the onset of claudication on two different datasets. The results suggest that we can identify the onset of claudication using inertial sensors with a best case accuracy of 92.25% for the Extreme Gradient Boosting model.

Abstract
O Grupo Internacional de Pesquisa Educação Digital (GPe-dU – UNISINOS/CAPES-CNPq)1 iniciou, em 2016, a pesquisa intitulada “A Cidade como Espaço de Aprendizagem: games e gamificação na constituição de Espaços de Convivência Híbridos, Multimodais, Pervasivos e Ubíquos para o desenvolvimento da Cidadania”, financiada pelo Edital Universal, na qual teve origem o We, Learning with the Cibricity – WLC. O evento foi construído a partir da necessidade da criação de um espaço-tempo comum de convivência e compartilhamento dos games e projetos gamificados desenvolvidos pelas crianças e adolescentes na cidade, bem como das práticas pedagógicas desenvolvidas pelos professores, realizadas nas escolas participantes do projeto, ao longo dos anos letivos em que o projeto se desenvolveu. Assim, o WLC foi realizado em 2016 seguido por quatro edições, até o ano de 2019, na modalidade presencial física. A pesquisa teve continuidade no projeto “A Cibricidade como Espaço de Aprendizagem: Pra´ticas pedago´gicas inovadoras para a promoc¸a~o da cidadania e do desenvolvimento social sustentável”, financiado pelo Edital “Anos finais do Ensino Fundamental: adolescências, qualidade e equidade na escola pública” da Fundação Carlos Chagas e Itaú Social, desenvolvido nos anos de 2019 a 2022. A articulação entre pesquisa, ensino (superior, pós-graduação e educação básica) e extensão (formação continuada de professores e a organização e realização do evento WLC) tomou tal dimensão, que fez surgir oportunidades de compartilhamento de projetos realizados também pelos professores, em um profícuo diálogo com pesquisadores, bem como a ampliação geográfica dos participantes, extrapolando as fronteiras nacionais. Essa história está sendo construída a partir das diferentes temáticas escolhidas e propostas em conexão com o que vivemos com o outro, sendo esse outro e entendido como tudo que faz parte da nossa ecologia.

Abstract
With the massive adoption of mobile devices, it became more mandatory for developers to provide high-quality applications. Nowadays, mobile devices are used for different purposes: entertainment, shopping, banking, and communication. Moreover, mobile devices can communicate and exchange information with various IoT devices distributed across the city. However, mobile application testing has different challenges when compared to other types of applications (i.e., desktop and client-server applications). First, we must consider mobile devices' different characteristics and limitations, such as connectivity, screen size, density, sensors, and limited battery. Second, there is a wide range of mobile devices from diverse vendors and models. Hence, there is a need to consider different device configurations to reduce compatibility issues that may occur in a high-fragmented ecosystem. In this case, several tools and services with various features and business models aim to run tests on multiple devices. In this practical experience report, we present the initial results of implementing a testing tool/service at Von Braun Labs to support the execution of tests across multiple Android devices. The stakeholders stated the need to (i) execute the tests on physical devices; and (ii) the tool/service must support tests that interact with a specialized IoT device. We start the study by comparing different tools/services to select the most suitable one for Von Braun Labs. We propose a comparison framework to help evaluate six tools/services based on their technical, usability, and customization features. Then, we present a case study with an app from Von Braun Labs to validate the selected testing environment. Finally, we discuss the lessons learned, contributions, and future directions, pinpointing the need for a testing process since the beginning of the development project and the importance of lessening the gap between academia and industry.

Abstract
To diagnose mobility impairments and select appropriate physiotherapy, gait assessment studies are often recommended. These studies are usually conducted in confined clinical settings, which may feel foreign to a subject and affect their motivation, coordination, and overall mobility. Conducting gait studies in unconstrained natural settings instead, such as the subject’s Activities of Daily Life (ADL), could provide a more accurate assessment. To appropriately diagnose gait deficiencies, muscle activity should be recorded in parallel with typical kinematic studies. To achieve this, Electromyography (EMG) and kinematic are collected synchronously. Our protocol sMaSDP introduces a simplified markerless gait event detection pipeline for the segmentation of EMG signals via Inertial Measurement Unit (IMU) data, based on a publicly available dataset. This methodology intends to provide a simple, detailed sequence of processing steps for gait event detection via IMU and EMG, and serves as tutorial for beginners in unconstrained gait assessment studies. In an unconstrained gait experiment, 10 healthy subjects walk through a course designed to mimic everyday walking, with their kinematic and EMG data recorded, for a total of 20 trials. Five different walking modalities, such as level walking, ramp up/down, and staircase up/down are included. By segmenting and filtering the data, we generate an algorithm that detects heel-strike events, using a single IMU, and isolates EMG activity of gait cycles. Applicable to different datasets, sMaSDP was tested in healthy gait and gait data of Parkinson’s Disease (PD) patients. Using sMaSDP, we extracted muscle activity in healthy walking and identified heel-strike events in PD patient data. The algorithm parameters, such as expected velocity and cadence, are adjustable and can further improve the detection accuracy, and our emphasis on the wearable technologies makes this solution ideal for ADL gait studies.