Professor of Physical and Rehabilitation Medicine
Department of Experimental and Clinical Medicine
Politecnica delle Marche University, Ancona, ITALY
A smartphone-based architecture to detect and quantify gait impairment in neurodegenerative disorders
Parkinson’s Disease (PD) is a common pathology characterized by a complex neurological degeneration of basal
ganglia that provokes difficulties in motor planning, action and execution of non-attention demanding tasks. Although gait disorders are quite a distinctive feature across all disease stages, freezing of gait (FoG) may be regarded as the most frequent, disabling, and least understood condition, involving up to 70% PD people after 10 years of clinical onset. FoG is defined as an episodic and transient inability to generate effective stepping in the absence of any known cause other than parkinsonism or high-level gait disorders. FoG impairs mobility, causes falls, reduces quality of life and social participation. The pathogenesis of FoG is still considered mysterious, given its episodic nature and heterogeneous manifestation. Furthermore, pharmacological treatments are of poor efficacy. A wearable system for automatic FoG detection can be helpful for several reasons: providing an objective assessment represents an add value to clinician judgement, that is currently regarded as the gold standard in FOG evaluation; moreover, quantitative measurements allow more reliable inter-individual and intra-individual comparisons over time or across different treatments; finally, and most importantly, the monitoring and assessment of FoG can be conducted during daily living, where almost all the episodes manifest.
Despite the effort of researchers in the last years to investigate wearable systems for FoG detection and gait monitoring, the use of these systems in the clinical practice is still missing. An important obstacle to this goal is the aggregation of big data into meaningful information for clinician. In fact, four levels can be distinguished in ubiquitous healthcare systems for remote monitoring: data creation, information generation, meaning making, and action taking. Until now, the most part of research focused on the first two topics, while less attention has been paid to the last two levels. Meaning making means to present the information extracted from data elaboration in a way that is easily accessible by clinicians, thus helping or suggesting the choice of the best action (action taking). For what concerns FoG, the number of episodes and the freezing time are two important metrics to evaluate FoG severity, hence they are apt to be used as aggregate indicators to interpret data collected during daily living. I will present an architecture for FoG monitoring based on a smartphone application. The FoG detection algorithm has gone through progressive improvements from a crisp logic to a fuzzy logic algorithm. However, the smartphone app would collect a great amount of useless and nonsense data during daily living, i.e. non-walking periods. Hence, the mentioned architecture has been improved by the addiction of a gait detection algorithm and its reliability in the estimation of FoG episodes count and FoG time has been evaluated against clinical observation. The gait detection algorithm works in synergy with the FoG detection algorithm in order to obtain a system that can provide aggregate, significant, and useful information to clinicians.
- Capecci M, Ceravolo MG, Ferracuti F, Grugnetti M, Iarlori S, Longhi S, Romeo L, Verdini F. An instrumental approach for monitoring physical exercises in a visual markerless scenario: A proof of concept. J Biomech. 2018 Mar 1;69:70-80.
- Capecci M, Ceravolo MG, Ferracuti F, Iarlori S, Kyrki V, Monteriù A, Romeo L, Verdini F. A Hidden Semi-Markov Model based approach for rehabilitation exercise J Biomed Inform. 2018 Feb;78:1-11.
- Pepa L, Verdini F, Capecci M, Ceravolo MG An unobtrusive expert system to detect freezing of gait during daily living in people with Parkinson’s disease. Splitech Conference 2017
- Capecci M, Ceravolo MG, Ferracuti F, Iarlori S, Longhi S, Romeo L, Russi SN, Verdini F. Accuracy evaluation of the Kinect v2 sensor during dynamic movements in a rehabilitation scenario. Conf Proc IEEE Eng Med Biol Soc. 2016 Aug;2016:5409-5412.
- Capecci M, Pepa L, Verdini F, Ceravolo MG. A smartphone-based architecture to detect and quantify freezing of gait in Parkinson’s disease. Gait Posture. 2016 Oct;50:28-33.
- Capecci M, Ceravolo MG, D’Orazio F, Ferracuti F, Iarlori S, Lazzaro G, Longhi S, Romeo L, Verdini F. A tool for home-based rehabilitation allowing for clinical evaluation in a visual markerless scenario. Conf Proc IEEE Eng Med Biol Soc. 2015 Aug;2015:8034-7.
- Capecci, M, Ceravolo, MG, Ferracuti F, Iarlori S, Kyrki V, Longhi S, Romeo L, Verdini F. Physical rehabilitation exercises assessment based on Hidden Semi-Markov Model by Kinect v2. 3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016: 256-259
- Maranesi E, Capitanelli L, Capecci M, Ghetti GG, Mercante O, Di Nardo F, Burattini L, Ceravolo MG, Fioretti S. A stereophotogrammetric-based method to assess spatio-temporal gait parameters on healthy and Parkinsonian subjects. Conf Proc IEEE Eng Med Biol Soc. 2015 Aug ;2015:5501-4.
- Pepa L, Capecci M, Verdini F, Ceravolo MG, Spalazzi L. An architecture to manage motor disorders in Parkinson’s disease. IEEE World Forum on Internet of Things, WF-IoT 2015 – Proceedings 2015: 615-620
- Pepa, L., Verdini, F., Capecci, M., Ceravolo, MG. Smartphone based freezing of gait detection for Parkinsonian patients. In: IEEE International Conference on Consumer Electronics (ICCE), 2015
- Benettazzo F, Iarlori S, Ferracuti F, Giantomassi A, Ortenzi D, Freddi A Monteriu`, Capecci M, Ceravolo MG, Innocenzi S, Longhi Low cost RGB-D vision based system for on-line performance evaluation of motor disabilities rehabilitation at home. In: Andò B, Siciliano P , Marletta V, Monteriù A. (Ed). Springer Verlag Ed 2014: 449-464
- Giantomassi A, Capecci M, Benettazzo F, Iarlori S, Ferracuti F, Freddi A, Monteriù A, Innocenzi S, Casoli P, Ceravolo MG, Longhi S, Leo T. Training and Retraining Motor Functions at Home with Help of Current
- Technology for Video Games: Basis for the Project. In: Andò B, Siciliano P , Marletta V, Monteriù A. (Ed). Springer Verlag 2014 : 439-448
- Pepa L, Verdini F, Capecci M, Maracci F, Ceravolo MG, Leo T. Predicting Freezing of Gait in Parkinson’s Disease with a smartphone: comparison between two algorithms. In: Andò B, Siciliano P , Marletta V, Monteriù A. (Ed). Springer Verlag 2014; 61-70
- Pepa L, Ciabattoni L, Verdini F, Capecci M, Ceravolo MG: Smartphone based fuzzy logic freezing of gait detection in Parkinson’s disease. In: 10th IEEE/ASME International Conference on Mechatronics and Embedded Systems and Applications (MESA) 2014: 1-6
- Pepa, L., Verdini, F., Capecci, M., Ceravolo, MG, Leo, T.: An architecture for reducing the freezing of gait during the daily life of patients with Parkinson’s disease. Gait Posture 2014; 40, S2
- Pepa L, Verdini F, Capecci M, Ceravolo MG, Leo T Can the current mobile technology help for medical assistance? The case of Freezing of Gait in Parkinson Disease In: Longhi S, Siciliano P (Eds) Springer Verlag 2013; 177-186
Maria Gabriella Ceravolo, MD, PhD in Neurosciences, Neurologist and Psychiatrist, is Professor of Physical and Rehabilitation Medicine at Università Politecnica delle Marche. She runs the Neurorehabilitation Clinic at the University Hospital of Ancona, Italy. Her scientific research points to developing innovative approaches to the assessment and rehabilitation of motor and cognitive impairment in subjects with acute or chronic-progressive brain diseases, and especially focuses on unveiling the different factors of disability progression and treatment efficacy in Parkinson’s disease. Thanks to a strong collaboration with the Dept. of Information Engineering, UNIVPM, her research group has developed not intrusive approaches to motor activity monitoring at home, by means of either wearable or sensor-less systems.