Bart Vanrumste

This PhD position at KU Leuven, Belgium, focuses on developing deep learning methods for human movement and behaviour analysis, with a particular emphasis on healthcare applications such as gait assessment and the evaluation of freezing of gait (FOG) severity. The successful candidate will join the eMedia research lab within the Department of Electrical Engineering (ESAT) and be enrolled at the Arenberg Doctoral School. Supervised by Professor Bart Vanrumste, whose research centers on multimodal sensor integration and machine learning for monitoring older adults and patients with chronic diseases, the project aims to create, test, and validate decolonized, just, and trustworthy foundation models (FMs) and an FM-based application for clinical and everyday FOG severity assessment. Key objectives include identifying technical gaps in implementing social justice in AI healthcare systems, developing wearable IMU sensor-based data collection methods, curating large-scale multi-label gait datasets, and building FMs capable of bias detection and mitigation. The project will also develop and validate an FM-based application for enhanced FOG severity assessment, with potential transferability to other clinical gait assessment use cases. Expected outcomes include meticulously curated datasets, a dedicated FM for FOG severity, and a validated application. The position offers opportunities for international collaboration, with planned secondments at Radboud University Medical Center (Netherlands) and Emory University (USA). Applicants must have a Master’s degree in engineering (mechanical, electrical, computer science, AI, or related), excellent academic performance, and programming experience in Python and deep learning frameworks. The role requires independent research skills, comfort with data collection involving older participants, and strong communication and teamwork abilities. The employment duration is 48 months starting September 2026, with a gross monthly salary of €4,010, plus mobility and family allowances as per Marie Skłodowska-Curie rules. KU Leuven is committed to diversity, inclusion, and equal opportunity. Applications must be submitted in English via the online system, including all required documents. For further information, contact Prof. Bart Vanrumste or Dr. Benjamin Filtjens.

This PhD position at KU Leuven focuses on developing just and trustworthy foundation models (FMs) for the detection and staging of rheumatic heart disease (RHD), with a strong emphasis on social justice in AI for healthcare. The doctoral candidate will enroll in the Arenberg Doctoral School and join the eMedia research lab under the supervision of Prof. Bart Vanrumste, within the Department of Electrical Engineering (ESAT). The research aims to develop, test, validate, and valorise decolonized, fair, and reliable FMs and an FM-based application for community-based RHD screening and staging. Key objectives include analyzing technical gaps in implementing social justice in healthcare AI, identifying requirements for just FM-based solutions, developing data collection workflows using handheld echocardiography and phonocardiography (PCG), curating a large-scale database of heart sound and echocardiogram data, and creating scalable FMs to help clinicians detect and mitigate biases in RHD screening. The project also involves developing and piloting an application for RHD screening and staging, with potential adaptation for other cardiovascular diseases and contributions to technical standardization. Expected outcomes include a curated dataset, a validated FM for RHD, and an application usable by nurses or community health workers in real-world settings. The position offers planned secondments at the University of Cape Town (South Africa) and Emory University (USA), providing international research exposure. The employment duration is 48 months starting September 1, 2026, with a gross monthly salary of €4,010, plus mobility and family allowances as per Marie Skłodowska-Curie rules. Applicants must have a master's degree in a relevant field, strong academic performance, programming experience (especially in Python and deep learning), and be willing to participate in community-based data collection and international research stays. The application process requires submission of a Europass CV, academic CV, motivation letter, references, degree certificate, proof of language proficiency, identification, and a Declaration of Honour, all in English and via the specified online system. KU Leuven is committed to diversity, inclusion, and equal opportunity, and encourages applications from all backgrounds.

KU Leuven is offering a fully funded PhD position on Automated Eating Activity Tracking System for Anorexia Nervosa Using AI and Wearable Sensors . The project is embedded in the EAST-STADIUS Division, specifically within the eMedia Research Lab, and carried out in close collaboration with the Mind-Body Research Group in the Division of Psychiatry, Department of Neurosciences. This interdisciplinary PhD sits at the intersection of artificial intelligence, wearable sensing, digital health, clinical neuroscience, and psychiatry . The eMedia Research Lab contributes expertise in machine learning, multimodal data analysis, intelligent technologies, and human-centered digital health. The Mind-Body Research Group brings strong clinical and translational experience in eating disorders, longitudinal monitoring, and patient-centered assessment. Together, the groups provide a setting where advanced computational methods can be translated into practical healthcare tools. The research focus is anorexia nervosa, one of the most severe and persistent mental health disorders. Eating behaviors such as restrictive eating, meal avoidance, rigid patterns, prolonged meal duration, and excessive control over food intake are central to the condition and closely related to illness severity, treatment response, and relapse. Current clinical assessment methods rely heavily on self-report, questionnaires, food diaries, and interviews, which are limited by recall bias and cannot capture eating behavior continuously in daily life. This project aims to develop an AI-enhanced wearable system for objective, continuous monitoring of eating behavior using wrist-worn IMU sensors. The project will design and validate data acquisition protocols, build robust signal processing and machine learning pipelines, and develop interpretable digital biomarkers that quantify eating rate, temporal regularity, behavioral rigidity, variability, and change over time. The broader goal is to create clinically meaningful tools that can support early identification of deterioration, monitoring of treatment progress, and assessment of recovery trajectories in anorexia nervosa. As a PhD researcher, you will work on wearable data collection, AI model development, longitudinal behavioral analysis, and clinical interpretation of results. You will collaborate closely with clinicians, psychologists, and researchers in a multidisciplinary environment and disseminate findings through peer-reviewed publications, conferences, and outreach activities. The position is fully funded and full-time. The contract starts on 2 November 2026, or as soon as possible thereafter, and is initially for one year with the possibility of renewal up to four years. The salary follows KU Leuven standards. The project also offers access to state-of-the-art research infrastructure, advanced training through the Doctoral School, and opportunities for interdisciplinary and international collaboration. Eligibility highlights include a Master’s degree in Engineering or a related area such as Computer Science, Artificial Intelligence, Electrical Engineering, Mechanical Engineering, or Biomedical Engineering; strong Python programming skills; high academic performance; and a strong interest in psychiatry, neuroscience, and clinical research. Excellent English is required, while Dutch is highly desirable due to participant-based data collection and interaction with clinical partners. Applications must be submitted via the KU Leuven jobsite before 2026-10-15 . For further information, prospective applicants may contact Dr. Chunzhuo Wang or Prof. Bart Vanrumste by email.