We value candidates who:• have a master degree in the field of data science, computer science, electronic engineering, biomedical engineering, bio-engineering, human health engineering or related field.• have knowledge in the field of data analysis, time series analysis, machine learning and algorithm development.• have knowledge on machine learning with Python or MATLAB.• are very fluent in English, both spoken and written.• possess strong analytical skills with the ability to collect, organize, analyse, and disseminate significant amounts of information with attention to detail and accuracy.• are adept at report writing and presenting findings.• are stress resistant and have passion for excellence.• have genuine passion for developing technology solutions that contribute to improving human health and well-being• are analytical, structured and result-oriented.• are highly self-motivated, flexible, logical thinking, goal-oriented, team players.• have the capability and interest to write scientific papers.• can work collaboratively with the research team leader and team to produce excellent research results.• can work with clinical partners and is willing to interact with patients.• are willing to become an expert in data analysis and machine learning with state-of-the-art and novel methodologies, and to understand how to balance performance with complexity to achieve implementable technologies. • are creative individuals and have a "persistent itch" to develop innovative solutions. Children and young people with dyskinetic cerebral palsy (CP) experience severe functional limitations in daily life due to involuntary movements and postures (i.e. dystonia and choreoathetosis). Dyskinetic CP is caused by a brain lesion around birth and has a prevalence of 0.14 in every 1000 live births in Belgium. Current clinical assessments of dystonia and choreoathetosis commonly video recorded within a hospital setting are not representative due to an increase in movement disorders in unfamiliar environments. Additionally, the visual assessment by an expert is time-consuming and lacks objectivity.Within this project, data extracted by markerless motion tracking from videos will be combined with data from inertial sensor units and heart rate parameters. These data will be simultaneously recorded in a familiar environment by a smart phone app. With this data a multimodal machine learning model will be trained to automatically classify severity of dystonia and choreoathetosis. With this approach a practically applicable and cost-effective tool will be prepared for evaluation of dystonia and choreoathetosis within a home-based environment. Existing protocols will be refined for performance by parents/caregivers and validity, reliability and responsiveness of the home-based approach will be evaluated. Steps will also be taken to facilitate the implementation within clinical care in Flanders after the project and will thereby impact medical and rehabilitation treatment for individuals with dyskinetic CP.Current treatment options for dyskinetic CP include invasive and expensive medical treatments such as deep brain stimulation and implanted medicine pumps and advanced rehabilitation products such as head-steering systems for computer access and mobility. Monitoring in a home-based setting will contribute to reducing hospital visits and to tailor treatments of children with dyskinetic CP - a small but severely disabled group with high medical and social cost.There will be two PhDs working on this project, a clinical PhD and a technical PhD (this position). In the technical PhD the main objectives are to contribute to: (i) developing a set-up for home-based measurements, (ii) collecting a high-quality multimodal dataset annotated for dystonia and choreoathetosis, (iii) re-fining existing video-based and sensor-based machine learning models and develop multimodal models for automated assessment of dystonia and choreoathetosis and (iv) preparing implementation of the automated assessment tool within clinical practice of children and young adults with dyskinetic CP. The PhD position will be shared by two KU Leuven research groups, M3-BIORES and the M-Group. The M3-BIORES (Measure, Model and Manage Bio-Responses) group in the Faculty of Bioscience Engineering (https://youtu.be/DcwLYgp1UhA), Department of Biosystems at KU Leuven is one of the largest scientific groups working in the field of integration of biological responses in monitoring and management of human and animal systems. M3-BIORES is based at Campus Arenberg in Leuven. The team currently have 30 members comprising postdocs, PhDs and visiting scholars, who are supported by dedicated technical and administrative staff (https://www.biw.kuleuven.be/biosyst/a2h/m3-biores/home). The research group M-Group (Mechatronics Group) of KU Leuven performs research on sensors and embedded devices as well as the networks between them. The research group comprises 5 professors and 20 research assistants or Ph.D. students and is based at the brand-new Bruges Campus. The PhD on this position is expected to work 4 days per week in Leuven (PhD promotor) and 1 day in Bruges (PhD copromotor).If you see yourself in the above description then we would really like to speak with you. We recognize that this is challenging position and will pay accordingly. KU Leuven offers a young and dynamic working environment with access to state-of-the-art infrastructure, a competitive salary and outstanding health care benefits. The project offers a PhD position for 4 years which should be sufficient to complete a PhD on the topic. You start with a full-time position for 1 year, which can be extended to four years upon positive evaluation. Other working conditions can be found at https://www.kuleuven.be/personeel/jobsite/en/phd/phd-information.