ARTORG Center for Biomedical Engineering Research

Seniorenuni visits NeuroTec Loft

GER

Sensor Based Monitoring

Sensor-based technologies and their capabilities have been rapidly growing over the past years. With this increase in performance and capabilities, our group has begun to focus on using this technology to improve the lives of a variety of individuals. In particular, our research focuses on in-home monitoring of both healthy individuals as well as those with neurodegenerative diseases such as Parkinson's Disease and Multiple Sclerosis. Our team also collaborates with various clinics in Bern to try to find solutions to real-world problems faced by healthcare professionals in caring for their patients. 

Privacy Statement (PDF, 103KB)

Digital Care Assistant

Digital Care Assistant

Current Project Members:
Lena Bruhin (PhD Student), Michael Single (PhD Student), & Stephan Gerber (Post-doc)
Project Start: 01.09.2021

Currently, there is a shortage of skilled workers and pressure to reduce costs in the health care system which is particularly evident in the care for older adults. As the number of geriatric psychiatric patients, especially the number of dementia patients, is steadily increasing in the context of increasing life expectancy and the baby boomer cohort retiring, the shortage of nurses will further increase. Therefore, there is a significant need for monitoring systems that simplify as well as optimize the processes and increase the efficiency of healthcare professionals 

It has been shown that remote monitoring of older people, while performing activities of daily living (e.g., washing hands, tying shoelaces, sleeping), with a multimodal sensor system over longer time periods is possible and provides a potential framework to detect health deteriorations. Such systems rely on contactless ambient sensors.

This project aims at developing a monitoring system, a digital care assistant, that supports healthcare professionals. The unobtrusive multimodal sensor system monitors behavioural and physiological parameters which are used to inform healthcare professionals about health-related changes and support them in decision-making of care needs.

 

Related Publications:

  • N. Schütz et al., “Contactless Sleep Monitoring for Early Detection of Health Deteriorations in Community-Dwelling Older Adults: Exploratory Study,” JMIR MHealth UHealth, vol. 9, no. 6, p. e24666, Jun. 2021, doi: 10.2196/24666.
  • A. Botros et al., “Long-Term Home-Monitoring Sensor Technology in Patients with Parkinson’s Disease—Acceptance and Adherence,” Sensors, vol. 19, no. 23, Art. no. 23, Jan. 2019, doi: 10.3390/s19235169.
  • S. M. Gerber et al., “An Instrumented Apartment to Monitor Human Behavior: A Pilot Case Study in the NeuroTec Loft,” Sensors, vol. 22, no. 4, p. 1657, Feb. 2022, doi: 10.3390/s22041657. 

 

Sleep Monitoring

Sleep Monitoring

Current Project Members:
Oriella Gnarra (PhD Student) & Stephan Gerber (Post-doc)
Project Start: 01.08.2021

Disturbed sleep is a common and problematic condition and often leads to cognitive impairment and increases the risks of accidents in daily life. In addition, sleep disorders are related to the development of depression, obesity, and neurological diseases. Some of these sleep disorders can occur during the prodromal phase of neurological conditions, such as Parkinson's disease, years before the onset of the cardinal symptoms that define the diagnosis. This finding indicates that identifying individuals in the prodromal phase will be of great interest when neuroprotective agents become available.

The gold standard for diagnosing sleep disorders is based on polysomnography (PSG) and records physiological signals such as brain waves, heart rate, muscle activity, eye movement, respiratory rate, and blood oxygen saturation. This device is composed of various sensors attached to the body and connected with multiple cables leading to an uncomfortable situation for the patient, affecting sleep quality and increasing the risk of misdiagnosis.

With advances in wireless technologies, wearable and contactless devices have been developed that show a high correlation with PSG. The novelty and strength of this project is the combination of multiple wearables and contactless sensors to test whether this unobtrusive sensor network can achieve the results that a single sensor does not currently achieve concerning the data quality and accuracy of the gold standard PSG.

Therefore, this project aims to determine how wearable and contactless technologies can be combined in a sensor fusion fashion to partially or entirely replace PSG to reduce intrusiveness and thus improve patient comfort and accuracy of diagnoses during in-hospital assessment and enable long-term assessment at patients' homes.

 

Related Publications:

  • Schindler, K. A. et al. NeuroTec Sitem-Insel bern: Closing the last mile in Neurology. Clinical and translational neuroscience 5, 13 (2021)

  • Schmidt, M. H. et al. Measuring Sleep, Wakefulness, and Circadian Functions in Neurologic Disorders. Sleep Medicine Clinics 4, 16 (2021)

Markerless Motion Tracking

Monitoring Human Movement by Markerless Motion Tracking

Current Project Members:
Kevin Möri (Master's Student), Lena Bruhin (PhD Student), Michael Single (PhD Student), & Stephan Gerber (Post-doc)
Project Start: 20.03.2023

Analyzing human movement is part of today's diagnostics in neurodegenerative diseases, where the progression of diseases is accompanied by a change in pathological movement patterns. One possible option to assess human movement patterns (i.e., digital biomarkers) is motion tracking (i.e., camera systems).

There exist two types of motion tracking systems. On one hand, there is marker-based motion tracking, which allows for great flexibility in marker positioning and enables reliable measurement results. However, physical objects attached to the body are likely to obstruct a person in certain movements and therefore alter the natural motion of that person. Furthermore, it is restricted to a laboratory setting as it requires markers to be attached to specific locations on the body. On the other hand, markerless motion tracking has the great advantage of contactless acquisition of motion, which is independent of its environment and allows the subject to move freely without any attached sensors or other objects. However, markerless motion tracking is less accurate.

Therefore, the project aims to develop digital biomarkers to assess human movement patterns based on maker-based and markerless motion tracking systems.

Digital Measures of MS

Digital Measures of Multiple Sclerosis from Multimodal Sensor Recordings

Current Project Members:
Michael Single (PhD Student), Lena Bruhin (PhD Student), Kevin Möri (Master's Student), Aileen Naef (PhD Student), & Stephan Gerber (Post-doc)
Project Start: 01.08.2020

The autoimmune disease multiple sclerosis (MS) is one of the most prominent causes of nontraumatic neurodegenerative disorder with a typical onset in young adults. Currently, in Switzerland, approximately 110 out of 100’000 people suffering from MS, whereas the prevalence in the last years was on the rise. The severity of MS symptoms considerably varies between patients, occurring in relapsing or progressive forms, and can be classified in motor (e.g., unsteady gait, tremor, weakness) and as well non-motor symptoms (e.g., vision issues, sensory disturbances, fatigue). Therefore, with the progression of the disease, the quality of life decreases and the risk of the need of institutional care increases.

Diagnosis and management routines are based on clinical findings like established scores, including the EDSS (Expanded Disability Status scale6) and the MSFC (Multiple Sclerosis Functional composite). However, the treatment regime is patient- and disease-stage specific and thus requires detailed information about motor and non-motor symptoms in a patient’s everyday life. The quality and number of activities of daily living (ADLs, such as cooking, going to bed, doing housework) that an MS patient can perform relates with the progression of the disease. Therefore, ADLs represent a promising digital measure, which can be exploited to extract information for offering patient-tailored treatment. There is, therefore, a strong clinical need for an objective and continuous method for monitoring of non-motor and motor symptoms related to ADLs in MS patients in a home-like environment for improving patients’ therapeutic regimens. In order to address this need, motor and non-motor parameters will be measured with an unobtrusive sensor system in an instrumented home (Neurotec Loft in Bern) and in the homes of participating MS patients.

 

Related Publications:

  • Gerber, S.M., Single, M., Knobel, S.E., Schütz, N., Bruhin, L.C., Botros, A., Naef, A.C., Schindler, K.A. and Nef, T., 2022. An Instrumented Apartment to Monitor Human Behavior: A Pilot Case Study in the NeuroTec Loft. Sensors, 22(4), p.1657.
  • Schütz, N., Knobel, S.E., Botros, A., Single, M., Pais, B., Santschi, V., Gatica-Perez, D., Buluschek, P., Urwyler, P., Gerber, S.M. and Müri, R.M., 2022. A systems approach towards remote health-monitoring in older adults: Introducing a zero-interaction digital exhaust. NPJ digital medicine, 5(1), pp.1-13.
  • Botros, A., Gyger, N., Schütz, N., Single, M., Nef, T. and Gerber, S.M., 2021. Contactless Gait Assessment in Home-like Environments. Sensors, 21(18), p.6205.
  • Schindler, K. A. et al. NeuroTec Sitem-Insel bern: Closing the last mile in Neurology. Clinical and translational neuroscience 5, 13 (2021)

Predicting Health Changes in PD

Digital Biomarkers To Predict Health Changes of Parkinson’s Disease Patients – The Closed Loop

Current Project Members:
Matilde Castelli (PhD Student), & Stephan Gerber (Post-doc)
Project Start: Fall 2022

To minimize Parkinson's Disease symptoms and thus increase patients' quality of life, the project's purpose is to develop a closed-loop control system, where the treatment (e.g., medication) is dynamically adapted to the patient's current need.    In this case, the current state or specifically motor- and non-motor symptoms (e.g., tremor, emotions, gait) are assessed through ambient and object sensors (e.g., inertial measurement unit, microphone) and then analyzed by machine learning algorithms (i.e., digital biomarkers) to predict health change and finally the needed adjustment in treatment.

 

Related Publications: