Tind Technologies (Norway)
Hes-so: ArODES Open Archive (University of Applied Sciences and Arts Western Switzerland / Haute école spécialisée de Suisse occidentale / FH Westschweiz)Not a member yet
15764 research outputs found
Sort by
Capstone ::mobility modeling on smartphones to achieve privacy by design
No full textSharing location traces with context-aware service providers has privacy implications. Location-privacy preserving mechanisms, such as obfuscation, anonymization and cryptographic primitives, have been shown to have impractical utility/privacy tradeoff. Another solution for enhancing user privacy is to minimize data sharing by executing the tasks conventionally carried out at the service providers' end on the users' smartphones. Although the data volume shared with the untrusted entities is significantly reduced, executing computationally demanding server-side tasks on resource-constrained smartphones is often impracticable. To this end, we propose a novel perspective on lowering the computational complexity by treating spatiotemporal trajectories as space-time signals. Lowering the data dimensionality facilitates offloading the computational tasks onto the digital-signal processors and the usage of the non-blocking signal-processing pipelines. While focusing on the task of user mobility modeling, we achieve the following results in comparison to the state of the art techniques: (i) mobility models with precision and recall greater than 80%, (ii) reduction in computational complexity by a factor of 2.5, and (iii) reduction in power consumption by a factor of 0.5. Using real-world mobility datasets, we demonstrate the suitability of our technique to function on smartphones
Geodabs ::trajectory indexing meets fingerprinting at scale
No full textFinding trajectories and discovering motifs that are similar in large datasets is a central problem for a wide range of applications. Solutions addressing this problem usually rely on spatial indexing and on the computation of a similarity measure in polynomial time. Although effective in the context of sparse trajectory datasets, this approach is too expensive in the context of dense datasets, where many trajectories potentially match with a given query. In this paper, we apply fingerprinting, a copy-detection mechanism used in the context of textual data, to trajectories. To this end, we fingerprint trajectories with geodabs, a construction based on geohash aimed at trajectory fingerprinting. We demonstrate that by relying on the properties of a space filling curve geodabs can be used to build sharded inverted indexes. We show how normalization affects precision and recall, two key measures in information retrieval. We then demonstrate that the probabilistic nature of fingerprinting has a marginal effect on the quality of the results. Finally, we evaluate our method in terms of performances and show that, in contrast with existing methods, it is not affected by the density of the trajectory dataset and that it can be efficiently distributed
Violence de couple chez les seniors ::Manuel d’aide à la détection et à la prise en charge destiné aux professionnel·le·s – version vaudoise
No full textCe manuel a été conçu sur la base d’une quarantaine d’entretiens avec des professionnel·le·s et de focus groups, ainsi que d’une dizaine de témoignages d’anciennes victimes de violence de couple, seniors au moment des faits. Il a été réalisé dans le cadre du projet de recherche appliquée «Prévention de la violence dans les couples âgés (VCA) : étude et développement de matériel de sensibilisation» et d’un module complémentaire vaudois financé par le Bureau de l’égalité entre les femmes et les hommes (VD)
Improving general practitioners’ approaches to functional somatic syndromes ::a pilot training program with a focus on compassion and communication
No full textBackground : Functional somatic syndromes are common in primary care and represent a challenge for general practitioners (GPs), with a risk of deterioration in the doctor-patient relationship, and of compassion fatigue on the part of the physician. Little is known about how to teach better management of these symptoms. Methods : The aim of our scientific team was to develop a training session about functional somatic syndromes for GPs, with the objective to improve the therapeutic attitude of the participants. The first session of the training was constructed as a pilot session, followed by a qualitative study to complete content validation. The educational framework of the training session is multimodal and includes theory on the pathophysiology of functional somatic syndromes, communication skills, and introspective learning including an introduction to compassion meditation. 20 physicians attended the pilot training session. 10 of them participated in the qualitative study. The qualitative study consisted of five individual semi-structured interviews and one focus group of five persons, investigating the impact of the training session on the clinical practices, as perceived by the participants. The interviews were analysed through an inductive method inspired by Malterud’s systematic text condensation strategy. Results : We identified three main themes in the responses of the participants: (1) the crucial issue of putting a name to chronic psychosomatic suffering; (2) the importance of self-compassion for physicians; (3) changes in therapeutic attitude fostering a reconciliation between “self” and “care”. Participants expressed a need for more regular meetings of this type. The opportunity to share their negative feelings about therapeutic relationship within a peer group, with compassionate supervision of the trainers, seemed to play an important role in the improvement of their self-compassion. Conclusion : A multimodal teaching session seems to help the physicians to feel more comfortable and competent when treating patients with functional somatic syndromes. Including compassion meditation in the teaching seems a promising tool to prevent compassion fatigue
Factors associated with intent to stay in the profession ::an exploratory cluster analysis across healthcare professions in Switzerland
No full text
Ateliers - travail sur les sources imprimées des œuvres avec alto de Rebecca Clarke, 17 octobre 2022
No full textUnderstanding artificial intelligence ::barriers and potential in wound care
No full textThe global wound burden is rising at an alarming pace due to increases in the ageing population and comorbidities and complications, such as obesity, diabetes and complex surgeries (Sen, 2021; Chen et al, 2024; Reifs et al, 2025). The World Health Organization (WHO) estimates there will be a global shortage of 18 million healthcare professionals (HCPs) by 2030 (WHO, 2016) to deliver care. To address these increasingly complex challenges, it is crucial to improve efficiency of healthcare systems, clinician education and consistency of wound care standards (The King’s Fund, 2018; Sen, 2021; Gould and Herman, 2025).
Within healthcare, artificial intelligence (AI) has emerged as a promising solution to several of these challenges with demonstrated improvements in diagnosis and treatment efficiency and clinician education, and productivity (Chen et al, 2024; Rippon et al, 2024). AI promises to replicate aspects of clinician experience and intelligence and can prove to be a useful tool in increasing the scale and speed of appropriate care provision (Bajwa et al, 2021; Rippon et al, 2024). AI has the potential to encompass all aspects of wound care and clinician education and training, including wound and risk assessment, healing prediction (e.g. by assessing patient comorbidities and social and psychological factors) and delivery of evidence-based, tailored treatment (Rippon et al, 2024; Reifs et al, 2025).
The aim of this consensus is to highlight for wound care clinicians and allied healthcare associates the multidimensional potential of AI, especially for chronic and/or complex wounds. A central theme of this consensus is to highlight the crucial role that wound care clinicians will need to play in implementing AI. It is only natural that some clinicians may be wary of the impact of AI on their job security. In this publication, we strive to dispel this myth and highlight that clinicians’ satisfaction with AI can only improve with a better understanding of what AI is and how it can be an addition to their toolbox. The expert panel also provide examples of implementing AI in their own wound care practices and share their learnings of improved outcomes, current barriers and areas of future need.
This consensus is not intended as a reference for highly technical AI terminology. Instead, the goal is to simplify the overwhelming amount of AI information for wound care clinicians, presenting key concepts in accessible language. We aim to help clinicians of all experience levels understand the implications and unmet needs in AI-driven wound care, empowering them to navigate their role in this rapidly evolving field.
Educating and preparing clinicians for the disruptive potential of AI is the first step towards creating effective, replicable, equitable and safe wound care systems that are increasingly needed for addressing the rising global wound care burden
Exploring the portability of ML-based lightning nowcasting models
No full textSignificant damages caused by lightning can be avoided by predicting lightning and taking precautionary measures. Here, a machine learning model is developed to nowcast lightning flashes using dew point temperature, precipitation, wind speed, wind direction, and previous lightning flashes. The model is trained using data from seven weather stations in Switzerland. The model demonstrates promising performance, achieving an F1 score ranging from 0.70 to 0.76. The main objective is to assess the portability of nowcasting models, by training them in one location and evaluating them in another. The goal is to develop a Machine Learning model that can be applied in regions lacking historical atmospheric measurements. The study reveals that the models generally perform well, with only a minor drop in performance in most cases (6% drop in the F1 score). However, in two cases involving mountainous terrain and tall structures, a significant drop is observed when a dataset was tested with models trained on other regions. Based on the findings, we recommend that for regions with complex topography (e.g., mountainous terrain) and/or tall structures (e.g, wind turbine parks), lightning nowcasting models should be trained on region-specific data, rather than relying on general-purpose forecasters