Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Internship/Master thesis - Automatic gait event detection for pathological gait: comparison between velocity based & frequency based detection methods
6 Month internship, in which different methods for automatic gait event detection for pathological gait will be compared. 3D motion capture data was already collected at the University Children's Hospital Basel and baseline codes are available. Skills in Matlab or Python are required.
**Background:**
In the last decade, clinical gait analyses (CGA) has become a standard assessment for lower extremity neural musculoskeletal disorders. While CGA provides a lot of information to clinicians, analyzing the results is very time consuming which prevents further implementation of CGA into the clinical setting. To allow for faster analyses of the data, automatic gait event (heel strike, toe off) detection is necessary.
Multiple automatic gait event detection methods have been developed for healthy gait. However, their performance on pathological gait is often poor or unknown. Therefore, we would like to evaluate the performance of different automatic gait event detection methods and compare them to the manual gait event localization by trained experts on the pathological gait data of children with various neural musculoskeletal disorders.
**Tasks & task content:**
10% Literature review, 10% Data preparation, 50% Programming, 20% Analysis & interpretation, 10% Reporting.
**Skillset:**
Matlab/Python;
**Project duration:**
February 2019 – August 2019
**Background:** In the last decade, clinical gait analyses (CGA) has become a standard assessment for lower extremity neural musculoskeletal disorders. While CGA provides a lot of information to clinicians, analyzing the results is very time consuming which prevents further implementation of CGA into the clinical setting. To allow for faster analyses of the data, automatic gait event (heel strike, toe off) detection is necessary.
Multiple automatic gait event detection methods have been developed for healthy gait. However, their performance on pathological gait is often poor or unknown. Therefore, we would like to evaluate the performance of different automatic gait event detection methods and compare them to the manual gait event localization by trained experts on the pathological gait data of children with various neural musculoskeletal disorders.
**Tasks & task content:** 10% Literature review, 10% Data preparation, 50% Programming, 20% Analysis & interpretation, 10% Reporting.
**Skillset:** Matlab/Python;
**Project duration:** February 2019 – August 2019
**Aim:**
1. Develop algorithm which is able to detect gait events in pathological gait automatically
2. Compare velocity based and frequency based detection methods with manual gait event detection by trained experts
**Aim:**
1. Develop algorithm which is able to detect gait events in pathological gait automatically
2. Compare velocity based and frequency based detection methods with manual gait event detection by trained experts
**Research group:**
Laboratory for Movement Biomechanics – Professorship William R. Taylor, PhD
Supervisor – Navrag Singh, PhD
Direct Advisors – Rosa Visscher & Deepak Ravi
**Location:**
HCP H16.1, Leopold-Ruzicka-Weg 4, 8093 Zurich, Switzerland
(UKBB, Spitalstresse 33, 4056 Basel, Switzerland)
**Contact:**
Rosa Visscher, Rosa.visscher@hest.ethz.ch
Deepak Ravi, Deepak.ravi@hest.ethz.ch
Institute of Biomechanics,HCP 16.1, ETH Zurich
Professorship Bill Taylor
**Research group:** Laboratory for Movement Biomechanics – Professorship William R. Taylor, PhD