Abstract
System identification approaches have been used to design an experiment, generate data, and estimate dynamical system models for Just Walk, a behavioral intervention intended to increase physical activity in sedentary adults. The estimated models serve a number of important purposes, such as understanding the factors that influence behavior and as the basis for using control systems as decision algorithms in optimized interventions. A class of identification algorithms known as matchable-observable linear identification has been reformulated and adapted to estimate linear time-invariant models from data obtained from this intervention. The experimental design, estimation algorithms, and validation procedures are described, with the best models estimated from data corresponding to an individual intervention participant. The results provide insights into the individual and the intervention, which can be used to improve the design of future studies.

Abstract
This study presents an optimal power management for a microgrid (MG) in distribution market environment. The MG operator is able to have interactions with the distribution market operator (DMO) and adjacent MG (AMG) operator to supply its local loads. The DMO regulates the electricity market, and assigns the electricity price and power profile for the MG. The motivation behind the use of The DMO is that it works as an entity between MG and independent system operator (ISO) in order to guarantee a flexible operation by reducing power fluctuation and reduce the unintentional peak loads in low market price hours. The unused power capacity in AMG is used during the islanded hours through an additional interconnection point connected to the MG. Using this method reduces the need for installing new generation resources, which will be a practical and economical solution for MG developer. This MG which is able to provide power from distribution market and the AMG with two interconnection points is named dependent MG (DMG). A market-based stochastic model containing distribution market constraints and AC power flow formulations employs conditional value at risk (CVaR) methodology to capture the loads and wind uncertainties. The effectiveness of the presented model is evaluated on a 20 kV test system using different case studies. In this test system, the operator is the owner of all generation units. The numerical analysis explicate that presented model reduces the operation cost of DMG with the aim of responsive loads, unused power capacity, energy storage system (ESS) and power generation units. The market-based scheduling also provides operational flexibility for distribution system by adjusting flexibility limit of market constraints. It is also shown that changing risk preferences level changes the power generation pattern in ESS and causes a costly operation of resources in risk-averse strategy. The competence of this model is significant when the preventive maintenance (PM) program is carried out, and the DMG should rely on its flexible resources and AMG's available power capacity, which could reduce the load shedding.

Abstract
The ever-increasing number of interconnected devices in smart environments, i.e., homes and cities, is bolstering the amount of data generated and exchanged. These devices can range from small embedded platforms, such as those included in home appliances, to critical operational systems, such as traffic lights. However, this increasing adoption is raising significant security and privacy concerns. Although some researchers have already solved some of these issues, data privacy still lacks a viable solution, especially when considering a flexible, decentralized approach to avoid a central overseer. One of the biggest challenges regarding privacy is the lack of transparency about how data flows are mediated and regulated as, often, these resources share data with external entities without the users' knowledge. We argue that a novel data-sharing control mechanism is required to properly control users' privacy and their respective Internet of Things (IoT) devices. This work focuses on a middleware layer solution for the IoT devices, which allows the control of the data generated by the device by its owner. The platform places the user as an active participant in the data market, behaving as its own data intermediary for potential consumers by monitoring, controlling, and negotiating the usage of their data.

Abstract
Importance Mammography screening currently relies on subjective human interpretation. Artificial intelligence (AI) advances could be used to increase mammography screening accuracy by reducing missed cancers and false positives. Objective To evaluate whether AI can overcome human mammography interpretation limitations with a rigorous, unbiased evaluation of machine learning algorithms. Design, Setting, and Participants In this diagnostic accuracy study conducted between September 2016 and November 2017, an international, crowdsourced challenge was hosted to foster AI algorithm development focused on interpreting screening mammography. More than 1100 participants comprising 126 teams from 44 countries participated. Analysis began November 18, 2016. Main Outcomes and Measurements Algorithms used images alone (challenge 1) or combined images, previous examinations (if available), and clinical and demographic risk factor data (challenge 2) and output a score that translated to cancer yes/no within 12 months. Algorithm accuracy for breast cancer detection was evaluated using area under the curve and algorithm specificity compared with radiologists' specificity with radiologists' sensitivity set at 85.9% (United States) and 83.9% (Sweden). An ensemble method aggregating top-performing AI algorithms and radiologists' recall assessment was developed and evaluated. Results Overall, 144 & x202f;231 screening mammograms from 85 & x202f;580 US women (952 cancer positive <= 12 months from screening) were used for algorithm training and validation. A second independent validation cohort included 166 & x202f;578 examinations from 68 & x202f;008 Swedish women (780 cancer positive). The top-performing algorithm achieved an area under the curve of 0.858 (United States) and 0.903 (Sweden) and 66.2% (United States) and 81.2% (Sweden) specificity at the radiologists' sensitivity, lower than community-practice radiologists' specificity of 90.5% (United States) and 98.5% (Sweden). Combining top-performing algorithms and US radiologist assessments resulted in a higher area under the curve of 0.942 and achieved a significantly improved specificity (92.0%) at the same sensitivity. Conclusions and Relevance While no single AI algorithm outperformed radiologists, an ensemble of AI algorithms combined with radiologist assessment in a single-reader screening environment improved overall accuracy. This study underscores the potential of using machine learning methods for enhancing mammography screening interpretation. Question How do deep learning algorithms perform compared with radiologists in screening mammography interpretation? Findings In this diagnostic accuracy study using 144 & x202f;231 screening mammograms from 85 & x202f;580 women from the United States and 166 & x202f;578 screening mammograms from 68 & x202f;008 women from Sweden, no single artificial intelligence algorithm outperformed US community radiologist benchmarks; including clinical data and prior mammograms did not improve artificial intelligence performance. However, combining best-performing artificial intelligence algorithms with single-radiologist assessment demonstrated increased specificity. Meaning Integrating artificial intelligence to mammography interpretation in single-radiologist settings could yield significant performance improvements, with the potential to reduce health care system expenditures and address resource scarcity experienced in population-based screening programs. This diagnostic accuracy study evaluates whether artificial intelligence can overcome human mammography interpretation limits with a rigorous, unbiased evaluation of machine learning algorithms.

Abstract
Purpose This study aims to address the topic of ethics, responsibility and sustainability (ERS) orientation of students enrolled in schools of economics and management master's degrees. It examines the effect of educational background and gender on Portuguese students' orientation towards ERS, as well as the extent to which there is a relation between the scientific area of the master degree in which the student is enrolled and his/her ERS orientation. Design/methodology/approach The authors used a sample of 201 students from several master degrees offered by the School of Economics and Management of a large public Portuguese university and analysed their ERS orientation using a survey by questionnaire. Findings Findings suggest that there are differences in orientation across gender, with female students valuing ERS more than their male counterparts. Educational background has minimal effects on the responses. It was also found some sort of selection effect in terms of the scientific area of the master degree and ERS orientation. Originality/value This study contributes to the literature by analysing the issue of whether students with an educational background in economics and management present different ERS orientation than their counterparts, as well as by examining whether there is some sort of self-selection into the study of disciplines in which ERS orientation is likely to be a week. As far as the authors are aware, this is the first study analysing this type of issue regarding ERS.

Abstract
Wearable devices have been showing promising results in a large range of applications: since industry, to entertainment and, in particular, healthcare. In the scope of movement disorders, wearable devices are being widely implemented for motor symptoms objective assessment. Currently, clinicians evaluate patients' motor symptoms resorting to subjective scales and visual perception, such as in Parkinson's Disease. The possibility to make use of wearable devices to quantify this disorder motor symptoms would bring an accurate follow-up on the disease progression, leading to more efficient treatments. Here we present a novel textile embedded low-power wearable device capable to be used in any scenario of movement disorders assessment due to its seamless, comfort and versatility. Regarding our research, it has already improved the setup of a wrist rigidity quantification system for Parkinson's Disease patients: the iHandU system. The wearable comprises a hardware sensing unit integrated in a textile band with an innovative design assuring higher comfort and easiness-to-use in movement disorders assessment. It enables to collect inertial data (9-axis) and has the possibility to integrate two analog sensors. A web platform was developed for data reading, visualization and recording. To ensure inertial data reliability, validation tests for the accelerometer and gyroscope sensors were conducted by comparison with its theoretical behavior, obtaining very good results.