Currently, patients with hyperkinetic movement disorders are classified on the basis of expert opinion. Electromyography (EMG) is used in some cases. The classification is therefore mainly based on clinical assessment. Assessing the type of hyperkinetic movement disorder is complex because there are small nuances between diseases and patients may have multiple disorders. In addition, man does its observation from a holite manner and therefore always looks at the coherence of what is being observed. For a proper classification and diagnosis of hyperkinetic movement disorders, an objective observation of (parts of) the body is essential. This concerns the frequency of movements of, for example, the upper arm, the angles under which this occurs and (in)random. The result of all this is that the correct classification and diagnosis of movement disorders currently has a Kappa value, a measure used to reflect the agreement between the specialists, from an average of 0.5 to 0.6. This means that the chances are relatively high that a wrong diagnosis is done, a wrong treatment is started and therefore the effectiveness of Dutch care is not optimal. ZiuZ and UMCG want to research in this project how artificial intelligence can contribute to improving the classification and diagnosis with the aim of increasing it to at least 0.8 and thus reducing the number of ‘wrong’ treatment. The aim of this project is therefore to develop a first proof of principle of a Computer Aided Diagnostic Tool (CAD tool) that aims to improve the diagnosis, treatment and evaluation of natural course of hyperkinetic movement disorders and using multiple data sources (video/sensors/medical information). The project results could then also be applied to e.g. Parkinson’s research, which has interfaces with tremors.