Sensor-supported Classification of Gait Patterns in Everyday Movement of Patients With Parkinson's Disease (PD-GPC)

Sensor-supported Classification of Gait Patterns in Everyday Movement of Patients With Parkinson's Disease

Sensor-supported Classification of Gait Patterns in Everyday Movement of Patients With Parkinson's Disease

Within this study we conduct long-term measurements on Parkinson patients using a three sensor IMU setup: one sensor is located at the wrist, one sensor at the belt/ in the pocket and one sensor is worn as a pendant. Before and after each long-term measurement phase (5 days within 6-8 weeks) we perform an UPDRS to identify changes in everyday life behaviour that correlate with UPDRS trends (especially part 3).

The aim of the study is to collect a database for the development of algorithms for the classification of everyday movement patterns of PD patients with the underlying sensor concepts. From the recorded movement data of PD patients and healthy volunteers (reference basis), characteristic features will be derived which could allow a distinction between pathological and unobtrusive non-disturbed movement patterns (e.g. on/off-state classification) and an assessment of the fall potential. Study Timeline The present study will be carried out in three, partly successive stages. The first two stages are measurements with the presented system during a medical examination (or afterwards) under annotation of certain motor exercises to determine the state of health of the patient or test person. In order to generate a sufficient reference database of comparable movement patterns, a series of measurements with subjects of a similar average age is carried out in parallel with the same sensor configuration and the same motor exercises. The third part of the study includes measurements in the home environment of Parkinson's patients. These measurements are used to identify the previously annotated movement patterns in the everyday life of the patient of the same previously included patient group. The individual stages of the study comprise in detail: Acquisition of patients/test persons The voluntary participants of this study will be recruited during the doctor's visit. The group of study participants with PD may consist of both outpatients and inpatients with Parkinson's disease (see also sub-section 4). The group of healthy volunteers is recruited from relatives or acquaintances of the patients or by request of uninvolved persons (in each case without motor impairments, see inclusion/exclusion criteria). If a patient or subject is suitable to participate in the study based on the inclusion and exclusion criteria, he or she will be informed and sign the consent form so that he or she can be included in the study. Each study participant receives an individual identification number, which is used to identify all other patient-specific documents. Annotation Measurements Parkinson Patients As soon as the patients have agreed to voluntarily participate in the study, the current disease status is first determined during a medical examination on the basis of a UPDRS and documented for the study evaluation. Following this, typical exercises for recording motor deficits are carried out. For this purpose, the patient is equipped with the intended sensor system and instructed in its use. The sensor system comprises three different sensor nodes, which are placed around the wrist, as a collar and in the trouser pocket. The wrist sensor is fixed to the patient's wrist by a flexible strap with a Velcro fastener. The neck sensor is positioned under the upper layer of clothing and hung around the neck by means of a special textile collar (approx. 35g). The third sensor knot has no special fastening device and is carried loosely in the trouser pocket. The attachment of the sensors is illustrated in Fig. 1. The movement data are recorded parallel to the exercises and extensively annotated according to the exercise characteristics. The exercises contain typical everyday movement patterns, which are to be identified in the further course of the study in the measurements in the domestic environment (subitem 3) and used for the evaluation of the state of health. These include among others: sit-to-stand test (STS), five-times-sit-to-stand test (5STS), sit-to-walk test (STW) or timed-up-and-go test (TUG), normal walking, climbing stairs, and handwriting. Annotation measurements reference group The volunteers participating in the study will be equipped with the same measuring equipment as the group of Parkinson's patients. The measurements on subjects of comparable average age serve as a reference to physiological movement sequences during the same motor exercises. On the one hand, the aim is to cover as broad a spectrum as possible for the classification of the various movement patterns. On the other hand, the data are to be used to identify pathological features that distinguish them from healthy movement patterns. Movement monitoring in the domestic/clinical environment of patients The group of outpatient/inpatient Parkinson's patients also take part in a long-term measurement in the domestic/clinical environment. Following the first motor examination, 5 consecutive measurements over a longer period of one day each are carried out in this section of the study. The equipment with the required measuring electronics takes place at the patient's place of residence/ in the clinic. In addition to the three previously used sensor nodes, the patient is equipped with a further node for recording on an SD card, which is also carried in the trouser pocket (approx. 35g). The SD card is written using low-level access and an internally developed protocol. This prevents third parties from accessing the transaction data. The measuring equipment is handed over on the morning of the scheduled day by a co-investigator. The patient is equipped with the system and the necessary handling of the sensor nodes is explained over the course of the day. As a daily form dependent annotated reference, the motor exercises mentioned under 2. are repeated - as far as possible in the patient's home environment / in the clinic. The patient is asked to keep an event log of the day at his own discretion and ability. On the evening of the same day, the measuring system is received again on site by a co-investigator. Any complications or questions regarding the further study protocol will be received, documented and, if necessary, reported to the study director. Final motor examination After completion of the third part of the study, each participant undergoes a final medical examination. This will take place following the nearest routine visit. In this examination, the motor exercises are carried out again in order to establish a basis for comparison with the initial situation (UPDRS Trend).

The "Integrated Posture and Activity NEtwork by Medit Aachen (IPANEMA)" Body Sensor Network (BSN) is a wireless sensor network with several sensor nodes. A sensor node consists of a base unit equipped with different sensor modalities: Acceleration sensor, Rotation rate sensor, Air pressure sensor, Photoplethysmography (PPG, optional), Galvanic Skin Response (GSR, optional) measurement. The sensor data is transmitted via a wireless 433/868 MHz ISM band radio link to the central radio node. From there, the data is transferred via a Bluetooth connection to a laptop or PC for processing. Three sensors are attached 1) to the wrist with a flexible Velcro strap, 2) hung around the neck on a specially designed collar and 3) worn loosely in the trouser pocket without further fixation or at the belt. The respective sensors are located in a closed, ergonomically shaped housing (size 70x40x20mm^3, approx. 35g) and have no electrically conductive contact to the skin surface of the test person.

The Unified Parkinson's Disease Rating Scale (UPDRS) will be conducted in two consecutive clinical visits (6-12 weeks) or during an inpatient stay (5 days+). In between we conduct 5 days of long-term monitoring for each patient using a wearable movement sensor system. Outcome measures of the system (overall activity, stride time, stride variability, gait symmetry, symptom appearance, etc.) will be correlated with the UPDRS trend.

Inclusion Criteria Patients: patients with Parkinson's disease patients undergoing outpatient or inpatient treatment at the UKA age of majority (age > 60 years) written declaration of consent persons who are legally competent and mentally able to follow the instructions of the study personnel Inclusion Criteria Healthy Volunteers: subjects of advanced age without neurological, psychiatric or other relevant internal diseases (age > 60 years) age of majority written declaration of consent persons who are legally competent and mentally able to follow the instructions of the study staff Exclusion Criteria (general): persons with currently known instabilities of the musculoskeletal system (high fall potential) persons with electrophobia (fear of electrical devices) persons wearing electrically active implants (e.g. cardiac or brain pacemakers) persons with neurological (except PD) or psychiatric disorders persons with relevant internal diseases (e.g. severe heart failure) persons who are accommodated in an establishment on official or judicial instruction persons who are dependent on or have an employment relationship with the sponsor or investigator simultaneous participation in another interventional clinical trial alcohol or drug abuse lack of compliance

Dennison AC, Noorigian JV, Robinson KM, Fisman DN, Cianci HJ, Moberg P, Bunting-Perry L, Martine R, Duda J, Stern MB. Falling in Parkinson disease: identifying and prioritizing risk factors in recurrent fallers. Am J Phys Med Rehabil. 2007 Aug;86(8):621-32. doi: 10.1097/PHM.0b013e311611583.

Pickering RM, Grimbergen YA, Rigney U, Ashburn A, Mazibrada G, Wood B, Gray P, Kerr G, Bloem BR. A meta-analysis of six prospective studies of falling in Parkinson's disease. Mov Disord. 2007 Oct 15;22(13):1892-900. doi: 10.1002/mds.21598.

Wood BH, Bilclough JA, Bowron A, Walker RW. Incidence and prediction of falls in Parkinson's disease: a prospective multidisciplinary study. J Neurol Neurosurg Psychiatry. 2002 Jun;72(6):721-5. doi: 10.1136/jnnp.72.6.721.

Stolze H, Klebe S, Zechlin C, Baecker C, Friege L, Deuschl G. Falls in frequent neurological diseases--prevalence, risk factors and aetiology. J Neurol. 2004 Jan;251(1):79-84. doi: 10.1007/s00415-004-0276-8.

Koller WC, Glatt S, Vetere-Overfield B, Hassanein R. Falls and Parkinson's disease. Clin Neuropharmacol. 1989 Apr;12(2):98-105. doi: 10.1097/00002826-198904000-00003.

Gray P, Hildebrand K. Fall risk factors in Parkinson's disease. J Neurosci Nurs. 2000 Aug;32(4):222-8. doi: 10.1097/01376517-200008000-00006.

Bloem BR, Grimbergen YA, Cramer M, Willemsen M, Zwinderman AH. Prospective assessment of falls in Parkinson's disease. J Neurol. 2001 Nov;248(11):950-8. doi: 10.1007/s004150170047.

Paul SS, Canning CG, Sherrington C, Lord SR, Close JC, Fung VS. Three simple clinical tests to accurately predict falls in people with Parkinson's disease. Mov Disord. 2013 May;28(5):655-62. doi: 10.1002/mds.25404. Epub 2013 Feb 28.