BIPBIP: Wearable System for Parkinson’s Disease
Award Ceremony Best FSE Project DGR 11/2018
6 November from 9.30 a.m.
Where?Auditorium of the Garden of Biodiversity (entrance from via Orto Botanico), University of Padua
On Wednesday, November 6th, the best innovation, development and research projects related to ESF (European Social Fund) funding were awarded. The BIP project, conducted together with the University of Verona and Humatics, was the winner among the 51 projects that received this funding.
Title Project: BIPBIP
A wearable system to reduce falls by assessing, preventing and reducing freezing of gait and orthostatic hypotension in Parkinson’s disease.
BIPBIP: Wearable IoT for FoG Prevention of Parkinson’s Patients
Freezing of Gait (FoG) and Orthostatic Hypotension (OH) are common and disabling symptoms which occur in the advanced stages of Parkinson’s Disease (PD) and increase the risk of falls. FoG and OH are often under-recognized and do not respond or may be worsened by levodopa and dopaminergic treatment, because their pathogenesis is related to neurodegeneration of non-dopaminergic systems in the brain and the peripheral nervous system.
Practical evidence from physiotherapy sessions has shown that PD patients more easily exit and overcome FoG or reduce its duration when they are externally stimulated with a Rhythmic Auditory Stimulation (RAS) (e.g., simulating a military march) . The generation of RAS generally causes embarrassment on the patient when other people are present. In addition, it requires the patient to be assisted by a caregiver.
High-waist compression stockings producing at least 15 to 20 mm Hg of pressure, may reduce OH by increasing blood pressure, but stockings are uncomfortable, need to be worn when not necessary, and PD patients struggle to put the stockings on, a number of points that limit the usefulness of compression stockings in everyday life .
Moreover, efficacious pharmacological strategies for FoG and OH are currently lacking.
Therefore, the project aims to develop a smart, non-invasive, wearable system that can assess, predict and be applied to treat FoG and OH, regardless of the presence of a caregiver, in order to reduce the risk of falls, therefore improving patient’s quality of life and participation to everyday’s activities.
The wearable system will be composed of a couple of smart elastic stockings and a control application. The stockings will be equipped with an embedded processor, where the control application will run, and a set of sensors including accelerometers, gyroscopes, magnetometers and blood pressure sensors.
In addition, the stockings will be able to provide the patient with tactile stimuli (i.e., rhythmic vibration), and they will include an automatic inflation/shrinkage device to augment venous return and increase blood pressure. The control application will exploit machine learning-based algorithms to assess and predict FoG and OH by analyzing the data gathered by the stockings’ sensors. In particular, sensors will allow to recognize patient’s activities (e.g., sitting, walking, lying, turning), motor status (e.g., ON, OFF, dyskinesia) and vital parameters (e.g., heartrate, blood pressure).
Once FoG or OH are predicted, the application will notify the stockings such that they generate tactile rhythmic stimuli or activate the inflation/shrinkage mechanism to, respectively, help the patient overcome the FoG or reduce its duration, and increase the blood pressure during sitting or standing. In this way, rhythmic stimuli generation and leg compression will be activated only when necessary.
Gathered data will be also remotely accessible by the physician such that gait, general motor performance and blood pressure, at home, can be easily monitored during routine clinical follow-up or on-demand, to optimize the treatment of the patient.
Stage of Development
At the current stage:
We have developed a machine learning-based algorithm to detect pre-FoG situations in order to generate the tactile rhythmic stimulation before FoG happens. Preliminary results on an 8-patient dataset have shown average precision and recall in predicting FoG by recognizing the pre-FoG episodes with specificity and sensitiviy of, respectively, 97% and 94%;
We have proved the algorithm can be executed on a resourced-constraint device in terms of computing power, memory size and battery consumption. This prove the algorithm is suitable to be executed on a computing unit embedded in the smart stockings.
Next steps to be accomplished will be:
Assessment of the effectiveness of tactile instead of auditory stimuli to help the patient prevents or exits FoG episodes;
Identification of a partner company for the realization of a smart elastic fabric equipped with accelerometers, gyroscope, magnetometer, blood pressure sensors, and an integrated automatic inflation/shrinkage mechanism;
Development of the control application to assess and prevent OH and notify the stocking the need of compressing legs;
Testing the overall system in a pilot study involving a small group of patients.
Because of the exploratory nature of the study and the need to improve the proposed technology, no a-priori hypotheses will be drawn, and the protocol will be applied to a group of 20 patients with advanced PD and FoG and/or OH. Clinical outcomes will include:
– Reduction of falls (primary outcome);
– Reduction of FoG and OH episodes;
– Improvement of patient’s quality of life and ability to participate to daily life activities;
– Feasibility and acceptability of the proposed device.
All existing devices are large, distracting and cannot be easily worn during everyday life. Our approach will significantly contribute to reduce the risk of falls, to improve the patient’s quality of life and active participation to everyday life activities, improving the wellbeing of PD patients. To the best of our knowledge, no long-time data from a consistent number of PD patients is available, to date, hampering the application of such devices in a personalized approach to improving FoG and OH in this condition.
In case of successful outcome of the proposed system, the partners are intended to found a start-up for the engineering of the proof-of-concept and the commercialization of the product. A crowdfunding campaign will be launched to initially support the activities of the start-up.
Future clinical aims will include the exploration of how the proposed technology may be combined to pharmacological and rehabilitation strategies to better tailor the most appropriate treatment to each single patient with advanced PD. Given the lack of efficacious therapies for FoG, OH and to reduce the risk of falls, the present proposal may pave the way to improve these outcomes in PD.
Partner: University of Verona, EDALab s.r.l., Humatics s.r.l.
Amount financed: 75.000 euro
Currently the project has passed the first step of selection and is being reviewed by the MJFOX Foundation for further funding and development.