A new index to assess turning quality and postural stability in patients with Parkinson's disease. (September 2020)

• Record Type: Journal Article
• Authors: Borzì, Luigi; Olmo, Gabriella; Artusi, Carlo Alberto; Fabbri, Margherita; Rizzone, Mario Giorgio; Romagnolo, Alberto; Zibetti, Maurizio; Lopiano, Leonardo
• Is Part Of: Biomedical signal processing and control. Volume 62(2020)
• Publication Date: 2020-09
• Subjects: Parkinson's disease (PD) -- Turns -- UPDRS scores -- Smartphone -- Wearable inertial sensors -- Machine learning; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2020.102059
• Languages: English
• ISSNs: 1746-8094

Abstract

Abstract: Parkinson's disease is a neuro-degenerative disorder characterized by the progressive death of dopamine neurons. This leads to delayed and uncoordinated movements, and impacts on the patients' motor performance with reduced movement intensity, increased axial rigidity and impaired cadence regulation. Turning provides privileged insights in postural instability and fall prediction, as it is regularly performed during daily activities, requires multi-limb coordination. The objective of this work was to define a Quality of Movement (QoM) index, inferred from inertial data related to turns, and strictly correlated with the patient's motor conditions, postural stability, and stage of the disease. Such a concise representation finds its main application in the remote monitoring of patients during daily activities at home. We have recorded and analyzed 180° turns in 72 patients, using inertial sensors embedded in the smartphone. We have set up an algorithm for binary classification of patients: mild vs. moderate/severe conditions, according to the Hoehn and Yahr scale of disease progression and disability degree. Our QoM index is defined as the a posteriori probability output by this binary classifier. It exhibits high correlation ( r = 0.73) with the clinical score of postural stability, as well as with the average of four clinical scores related to movement impairment ( r = 0.75). These results, together with the widespread smartphone use, provide a step in the direction of a practical, objective and reliable tool for PD patients remote monitoring in domestic environment.