In the biomedical field we work on the detection of neurological diseases using wearable sensors and video cameras, in particular: Parkinson's disease and cervical dystonia. A special feature of the project is that we are engaged in multimodal analysis of data collected from patients. These are time series from sensors and patient videos. By processing them simultaneously we can classify Parkinson's disease, essential tremor, dystonia, etc. Now we have made an attempt to classify the first and second stages of Parkinson's disease. Because that's the biggest headache for neurologists. Because that's where most of the misdiagnosis is. That is, up to 30% of misdiagnoses occur worldwide.