In clinical practice, it is sometimes difficult to establish whether a patient's tremor is due to Parkinson's disease or essential tremor. The distinction is crucial as the health implications differ significantly between the two conditions. Therefore, the present study aims to develop a diagnostic method based on machine learning techniques to help differentiate whether a patient's tremor is due to one condition or the other. To achieve this, 110 patients with tremor, correctly diagnosed with either Parkinson's disease or essential tremor, will participate. They will undergo two diagnostic tests (tapping test and Archimedean spiral) to capture data that can be processed using machine learning techniques.
Possibility to collaborate in the necessary evaluations.
Follow-up in the specialized consultation of Movement Disorders at the Neurology Service of Hospital Sant Camil-Consorci Sanitari Alt Penedes i Garraf.
Legal capacity to provide informed consent.
Signature of informed consent for study inclusion, either by the participant themselves or by their legal representative.
Patients undergoing treatment with antipsychotics or antidepressants.
Patients with Parkinson's disease and dyskinesias.
Patients undergoing treatment with dopaminergic agonists or primidone.
Tremor of such severity that it prevents continuous tracing, at the investigator's discretion.
The participants will be contacted based on the patient registry with follow-up in the neurology clinic of the study center.
Case-crossover
Prospective
Tapping Test
Diagnostic testFor the administration of the tapping test, a computer application installed on a Tablet will be used. The test will be administered up to 3 times, with a 15-minute interval between each administration. Each tapping test trial will last 15 seconds. At the beginning of the session, the patient will perform 2 practice attempts.
Archimedes Spiral
Diagnostic testFor the administration of the Archimedean spiral, a computer application installed on a Tablet will be used. The Tablet screen will display a drawing of the spiral, serving as a reference for the participant. The test will be administered up to 3 times, with a 15-minute interval between each administration. At the beginning of the session, the patient will perform 2 practice attempts.
Sensitivity
Proportion of participants with confirmed Parkinson's disease for whom the machine learning-based diagnostic algorithm yields a 'positive' result.
Specificity
Proportion of participants with confirmed essential tremor for whom the machine learning-based diagnostic algorithm yields a 'negative' result.
Reliability
The reliability of the diagnostic algorithm will be evaluated based on the repeatability of the classification result (positive or negative) among the different tests conducted on the same patient
Click on the lead sponsor to view all of their trials.
