Semester Project / Internship: Anomaly Detection for Biosignals Processing for Robots in Healthcare

Modern wearable and nearable sensors allow for continuous 24-hour health monitoring. In clinical settings, such a richness of data is highly desirable to closely monitor the health status of patients, however, often not exploited due to the lack of common interfaces and the overall complexity of integrating the devices into the clinical routine. Moreover, future treatment devices could benefit from interfacing with these sensors through a common interface to perform closed-loop interventions. Based on the three stakeholders in healthcare: patients, clinicians, and engineers, we want to build an architecture that allows the integration of different sensors into a single environment. Engineers should be able to integrate existing sensors into the architecture through wrappers and at the same time be able to access raw data for future applications. Clinicians should also have access to the data collected from the sensors and high-level analysis and modelling. Finally, Patients should have access to their data presented in a concise and meaningful manner.

Modern wearable and nearable sensors allow for continuous 24-hour health monitoring. In clinical settings, such a richness of data is highly desirable to closely monitor the health status of patients, however, often not exploited due to the lack of common interfaces and the overall complexity of integrating the devices into the clinical routine.
Moreover, future treatment devices could benefit from interfacing with these sensors through a common interface to perform closed-loop interventions.
Based on the three stakeholders in healthcare: patients, clinicians, and engineers, we want to build an architecture that allows the integration of different sensors into a single environment.
Engineers should be able to integrate existing sensors into the architecture through wrappers and at the same time be able to access raw data for future applications.
Clinicians should also have access to the data collected from the sensors and high-level analysis and modelling.
Finally, Patients should have access to their data presented in a concise and meaningful manner.

Your task will be to design an architecture for data communication between patients, clinicians, and robots in the context of healthcare applications. 1. Based on use cases from patients, clinicians, and engineers from Swiss Paraplegic Center in Nottwil and National Center for Geriatrics and Gerontology in Japan, develop the automated system for synchronization. 2. Perform a literature review 3. Designing the architecture of human<>sensor<>robot communication within safe network communication for use in healthcare applications 4. Develop a modular method of sensor synchronization, hereby ensuring time synchronization between the different acquisition pipelines in multi-sensory systems. 5. Evaluate the system during inpatient rehabilitation tracking as defined by the clinical teams. 6. Evaluate the clinical applications' ease of integration and use-case assessment. 7. Publishing results (desirable)

Your task will be to design an architecture for data communication between patients, clinicians, and robots in the context of healthcare applications.

1. Based on use cases from patients, clinicians, and engineers from Swiss Paraplegic Center in Nottwil and National Center for Geriatrics and Gerontology in Japan, develop the automated system for synchronization.
2. Perform a literature review
3. Designing the architecture of human<>sensor<>robot communication within safe network communication for use in healthcare applications
4. Develop a modular method of sensor synchronization, hereby ensuring time synchronization between the different acquisition pipelines in multi-sensory systems.
5. Evaluate the system during inpatient rehabilitation tracking as defined by the clinical teams.
6. Evaluate the clinical applications' ease of integration and use-case assessment.
7. Publishing results (desirable)

You will have the unique opportunity of working with researchers from clinical, computer science and robotics for developing this framework applied in robotic sleep therapy and assistive wheelchairs. Moreover, you can implement this framework within an international collaboration with robotic partners in Japan and implement the system on a facility in the partner institution (with potential time abroad).

You will have the unique opportunity of working with researchers from clinical, computer science and robotics for developing this framework applied in robotic sleep therapy and assistive wheelchairs. Moreover, you can implement this framework within an international collaboration with robotic partners in Japan and implement the system on a facility in the partner institution (with potential time abroad).

- ETHZ: D-MAVT, D-INFK / EPFL: IMT, CS Experience with: - Signal processing - Model Optimization - ROS and Gazebo - Experience in package development (preferred) - Knowledge of virtual environments (conda / docker) - Strong experience with Python / C++ - Structured and reliable working style - Ability to work independently on a challenging topic - Understanding of embedded computing (preferable)

- ETHZ: D-MAVT, D-INFK / EPFL: IMT, CS

Experience with:
- Signal processing
- Model Optimization
- ROS and Gazebo
- Experience in package development (preferred)
- Knowledge of virtual environments (conda / docker)
- Strong experience with Python / C++
- Structured and reliable working style
- Ability to work independently on a challenging topic
- Understanding of embedded computing (preferable)

Supervisor: Dr. Jorge Peña (SCAI-Lab, ETHZ) Host: Dr. Diego Paez (SCAI-Lab, ETHZ | SPZ) Please send your CV and latest transcript to: jorge.penaqueralta@hest.ethz.ch

Supervisor: Dr. Jorge Peña (SCAI-Lab, ETHZ)
Host: Dr. Diego Paez (SCAI-Lab, ETHZ | SPZ)
Please send your CV and latest transcript to: jorge.penaqueralta@hest.ethz.ch