Continuous monitoring, continuous auditing and continuous assurance are three methods that utilize a high degree of business intelligence and analytics. The increased interest in the three methods has led to multiple studies that analyze each method or a combination of methods from a micro-level. However, limited studies have focused on the perceived usage scenarios of the three methods from a macro level through the eyes of the end-user. In this study, we bridge the gap by identifying the different usage scenarios for each of the methods according to the end-users, the accountants. Data has been collected through a survey, which is analyzed by applying a nominal analysis and a process mining algorithm. Results show that respondents indicated 13 unique usage scenarios, while not one of the three methods is included in all of the 13 scenarios, which illustrates the diversity of opinions in accountancy practice in the Netherlands.
DOCUMENT
Background: In 2009, the Steering Committee for Pregnancy and Childbirth in the Netherlands recommended the implementation of continuous care during labor in order to improve perinatal outcomes. However, in current care, routine maternity caregivers are unable to provide this type of care, resulting in an implementation rate of less than 30%. Maternity care assistants (MCAs), who already play a nursing role in low risk births in the second stage of labor and in homecare during the postnatal period, might be able to fill this gap. In this study, we aim to explore the (cost) effectiveness of adding MCAs to routine first- and second-line maternity care, with the idea that these MCAs would offer continuous care to women during labor. Methods: A randomized controlled trial (RCT) will be performed comparing continuous care (CC) with care-as-usual (CAU). All women intending to have a vaginal birth, who have an understanding of the Dutch language and are > 18 years of age, will be eligible for inclusion. The intervention consists of the provision of continuous care by a trained MCA from the moment the supervising maternity caregiver establishes that labor has started. The primary outcome will be use of epidural analgesia (EA). Our secondary outcomes will be referrals from primary care to secondary care, caesarean delivery, instrumental delivery, adverse outcomes associated with epidural (fever, augmentation of labor, prolonged labor, postpartum hemorrhage, duration of postpartum stay in hospital for mother and/or newborn), women’s satisfaction with the birth experience, cost-effectiveness, and a budget impact analysis. Cost effectiveness will be calculated by QALY per prevented EA based on the utility index from the EQ-5D and the usage of healthcare services. A standardized sensitivity analysis will be carried out to quantify the outcome in addition to a budget impact analysis. In order to show a reduction from 25 to 17% in the primary outcome (alpha 0.05 and bèta 0.20), taking into account an extra 10% sample size for multi-level analysis and an attrition rate of 10%, 2 × 496 women will be needed (n = 992). Discussion: We expect that adding MCAs to the routine maternity care team will result in a decrease in the use of epidural analgesia and subsequent costs without a reduction in patient satisfaction. It will therefore be a costeffective intervention. Trial registration: Trial Registration: Netherlands Trial Register, NL8065. Registered 3 October 2019 - Retrospectively registered.
DOCUMENT
In this post I give an overview of the theory, tools, frameworks and best practices I have found until now around the testing (and debugging) of machine learning applications. I will start by giving an overview of the specificities of testing machine learning applications.
LINK
Automated driving nowadays has become reality with the help of in-vehicle (ADAS) systems. More and more of such systems are being developed by OEMs and service providers. These (partly) automated systems are intended to enhance road and traffic safety (among other benefits) by addressing human limitations such as fatigue, low vigilance/distraction, reaction time, low behavioral adaptation, etc. In other words, (partly) automated driving should relieve the driver from his/her one or more preliminary driving tasks, making the ride enjoyable, safer and more relaxing. The present in-vehicle systems, on the contrary, requires continuous vigilance/alertness and behavioral adaptation from human drivers, and may also subject them to frequent in-and-out-of-the-loop situations and warnings. The tip of the iceberg is the robotic behavior of these in-vehicle systems, contrary to human driving behavior, viz. adaptive according to road, traffic, users, laws, weather, etc. Furthermore, no two human drivers are the same, and thus, do not possess the same driving styles and preferences. So how can one design of robotic behavior of an in-vehicle system be suitable for all human drivers? To emphasize the need for HUBRIS, this project proposes quantifying the behavioral difference between human driver and two in-vehicle systems through naturalistic driving in highway conditions, and subsequently, formulating preliminary design guidelines using the quantified behavioral difference matrix. Partners are V-tron, a service provider and potential developer of in-vehicle systems, Smits Opleidingen, a driving school keen on providing state-of-the-art education and training, Dutch Autonomous Mobility (DAM) B.V., a company active in operations, testing and assessment of self-driving vehicles in the Groningen province, Goudappel Coffeng, consultants in mobility and experts in traffic psychology, and Siemens Industry Software and Services B.V. (Siemens), developers of traffic simulation environments for testing in-vehicle systems.
Movebite aims to combat the issue of sedentary behavior prevalent among office workers. A recent report of the Nederlandse Sportraad reveal a concerning trend of increased sitting time among Dutch employees, leading to a myriad of musculoskeletal discomforts and significant health costs for employers due to increased sick leave. Recognizing the critical importance of addressing prolonged sitting in the workplace, Movebite has developed an innovative concept leveraging cutting-edge technology to provide a solution. The Movebite app seamlessly integrates into workplace platforms such as Microsoft Teams and Slack, offering a user-friendly interface to incorporate movement into their daily routines. Through scalable AI coaching and real-time movement feedback, Movebite assists individuals in scheduling and implementing active micro-breaks throughout the workday, thereby mitigating the adverse effects of sedentary behavior. In collaboration with the Avans research group Equal Chance on Healthy Choices, Movebite conducts user-centered testing to refine its offerings and ensure maximum efficacy. This includes testing initiatives at sports events, where the diverse crowd provides invaluable feedback to fine-tune the app's features and user experience. The testing process encompasses both quantitative and qualitative approaches based on the Health Belief Model. Through digital questionnaires, Movebite aims to gauge users' perceptions of sitting as a health threat and the potential benefits of using the app to alleviate associated risks. Additionally, semi-structured interviews delve deeper into user experiences, providing qualitative insights into the app's usability, look, and feel. By this, Movebite aims to not only understand the factors influencing adoption but also to tailor its interventions effectively. Ultimately, the goal is to create an environment encouraging individuals to embrace physical activity in small, manageable increments, thereby fostering long-term engagement promoting overall well-being.Through continuous innovation and collaboration with research partners, Movebite remains committed to empowering individuals to lead healthier, more active lifestyles, one micro-break at a time.
Vacation travel is an essential ingredient in quality of life. However, the contriubtion of vacations to quality of life could be improved in two ways: by optimizing the decisions people make when planning and undertaking their vacations, and by travel industry testing and implementing––based on evidence––innovative experience products which touch customers' emotions. Secondary analysis of two longitudinal panel datasets will address the impact of people's decisions in planning and undertaking their vacations, on their quality of life. Field experiments in cooperation with travel industry partners will address the effects of innovative experience products, such as apps designed to help vacationers meet fellow travelers, or personalized memory books designed to help people relive their vacations after return home. Experience data in these field experiments will be collected using technology of the Breda University of Applied Sciences' Experience Measurement Lab, a unique facility for measuring emotions continuously from research participants' body and mind. Thus, the project will contribute to general understanding of quality of life, will feed valuable knowledge about experience design, measurement, and implementation to the Dutch travel industry, and will support the Breda University of Applied Sciences' key research theme of Designing, Measuring, and Managing Experiences. Inspiring examples from the project will reinforce research methods courses in the academic Bachelor of Science in Tourism, the HBO Master in Tourism Destination Management, and the academic Master of Science in Leisure Studies. Wearable emotion measurement from the field experiment will be a cornerstone of the fourth-year HBO-bachelor module Business Intelligence, where students will conduct their own research projects on experience measurement using consumer wearables, based on knowledge from this postdoc project. Finally, a number of methodological and content questions within the project will serve as suitable thesis assignments for graduation students in the above educational tracks.
