Register now After registration you will be able to apply for this opportunity online.
Internship/ Master Thesis: Machine Learning for Assessment of Walking Patterns in the SCI population - Time Series Classification
Gait patterns in multiple impairments present unique and complex patterns, which hinders the proper quantitative assessment of the walking ability for chronic ambulatory conditions when translated to daily living. In this project, we will focus on finding clusters of gait patterns through unsupervised learning from a large dataset of incomplete spinal cord injury individuals. The goal is to investigate hidden patterns in relation to the type of injuries and find their application for future diagnosis and rehabilitation treatment.
Your work will guide future rehabilitation methods in general clinical practice, through applied classification and dimensionality reduction in Biomechanics of walking.
Goal: Develop an unsupervised clustering pipeline for a large dataset of gait patterns from spinal cord injured individuals for class similarity evaluation
Keywords: Medical and health science, computing and computational science, engineering and technology, information, machine learning, data science, data engineering
Gait patterns in multiple impairments present unique and complex patterns, which hinder the proper quantitative assessment of the walking ability for chronic ambulatory conditions when translated to daily living.
In this project, we will focus on finding clusters of gait patterns through unsupervised learning from a large dataset of incomplete spinal cord injury individuals. The goal is to investigate hidden patterns in relation to the type of injuries and find their application for future diagnosis and rehabilitation treatment.
Through gait analysis, multiple valuable biomechanical metrics and features are extracted which can directly be used as a decision-making tool for gait disorders. These metrics are derived through kinematic and kinetic study of the gait patterns. In this study we aim to employ clustering algorithms over the gait computational information, to explore the identifiable quality levels of gait. The clustering can be performed using the biomechanical gait features (feature-based clustering) or the original time series patterns of the gait, derived through motion capture systems (time-series clustering)[1-3].
1. S. Aghabozorgi, A. Seyed Shirkhorshidi, and T. Ying Wah, “Time-series clustering – A decade review,” Inf Syst, vol. 53, pp. 16–38, Oct. 2015, doi: 10.1016/j.is.2015.04.007.
2. K. Kuruvithadam, M. Menner, W. R. Taylor, M. N. Zeilinger, L. Stieglitz, and M. Schmid Daners, “Data-Driven Investigation of Gait Patterns in Individuals Affected by Normal Pressure Hydrocephalus,” Sensors, vol. 21, no. 19, p. 6451, Sep. 2021, doi: 10.3390/s21196451.
3. D. Slijepcevic et al., “Explaining Machine Learning Models for Clinical Gait Analysis,” ACM Trans Comput Healthc, vol. 3, no. 2, pp. 1–27, Apr. 2022, doi: 10.1145/3474121.
Gait patterns in multiple impairments present unique and complex patterns, which hinder the proper quantitative assessment of the walking ability for chronic ambulatory conditions when translated to daily living. In this project, we will focus on finding clusters of gait patterns through unsupervised learning from a large dataset of incomplete spinal cord injury individuals. The goal is to investigate hidden patterns in relation to the type of injuries and find their application for future diagnosis and rehabilitation treatment.
Through gait analysis, multiple valuable biomechanical metrics and features are extracted which can directly be used as a decision-making tool for gait disorders. These metrics are derived through kinematic and kinetic study of the gait patterns. In this study we aim to employ clustering algorithms over the gait computational information, to explore the identifiable quality levels of gait. The clustering can be performed using the biomechanical gait features (feature-based clustering) or the original time series patterns of the gait, derived through motion capture systems (time-series clustering)[1-3].
1. S. Aghabozorgi, A. Seyed Shirkhorshidi, and T. Ying Wah, “Time-series clustering – A decade review,” Inf Syst, vol. 53, pp. 16–38, Oct. 2015, doi: 10.1016/j.is.2015.04.007. 2. K. Kuruvithadam, M. Menner, W. R. Taylor, M. N. Zeilinger, L. Stieglitz, and M. Schmid Daners, “Data-Driven Investigation of Gait Patterns in Individuals Affected by Normal Pressure Hydrocephalus,” Sensors, vol. 21, no. 19, p. 6451, Sep. 2021, doi: 10.3390/s21196451. 3. D. Slijepcevic et al., “Explaining Machine Learning Models for Clinical Gait Analysis,” ACM Trans Comput Healthc, vol. 3, no. 2, pp. 1–27, Apr. 2022, doi: 10.1145/3474121.
- Develop novel features from gait data defined with the clinical team.
- Analyze data quantity and quality through visualization and graphical mapping of features.
- Develop a pipeline for clustering gait metrics and gait pattern time series.
- Developing models for classification exploiting the available data.
- Evaluate supervised and unsupervised learning models using the existing labels.
- Write a report, highlighting the results for each phase.
- Publish your results (optional)
- Develop novel features from gait data defined with the clinical team. - Analyze data quantity and quality through visualization and graphical mapping of features. - Develop a pipeline for clustering gait metrics and gait pattern time series. - Developing models for classification exploiting the available data. - Evaluate supervised and unsupervised learning models using the existing labels. - Write a report, highlighting the results for each phase. - Publish your results (optional)
- Gain unique access and first-hand experience in one of the leading institutions on long-term health management - At the Swiss Paraplegic Center at Nottwil.
- Develop a highly impactful project with direct application to clinical practice.
- Learn and practice unsupervised & supervised learning methods from time-series data.
- Gain unique access and first-hand experience in one of the leading institutions on long-term health management - At the Swiss Paraplegic Center at Nottwil. - Develop a highly impactful project with direct application to clinical practice. - Learn and practice unsupervised & supervised learning methods from time-series data.
- ETHZ: D-MAVT, D-INFK / EPFL: IMT, CS
- A fundamental understanding of machine learning methods.
- Strong understanding of statistics, clustering and unsupervised classification.
- Proven records on some of the following: longitudinal data analysis, sparse feature selection, otr deep learning (preferred).
- Knowledge of virtual environments (conda / docker)
- Strong experience with Python (preferred)
- Structured and reliable working style
- Ability to work independently on a challenging topic
- ETHZ: D-MAVT, D-INFK / EPFL: IMT, CS - A fundamental understanding of machine learning methods. - Strong understanding of statistics, clustering and unsupervised classification. - Proven records on some of the following: longitudinal data analysis, sparse feature selection, otr deep learning (preferred). - Knowledge of virtual environments (conda / docker) - Strong experience with Python (preferred) - Structured and reliable working style - Ability to work independently on a challenging topic
Host: Dr. Diego Paez (SCAI Lab)
Prof. Robert Riener (SMS Lab)
Dr. Med. Inge Eriks (SPZ)
Please send your CV and latest transcript of records from my studies to Dr Diego Paez (diego.paez _at_ hest.ethz.ch)
Host: Dr. Diego Paez (SCAI Lab) Prof. Robert Riener (SMS Lab) Dr. Med. Inge Eriks (SPZ)
Please send your CV and latest transcript of records from my studies to Dr Diego Paez (diego.paez _at_ hest.ethz.ch)