Education

Internships, Master’s Thesis,
Bachelor & Semester Projects

We offer projects related to (i) the development of new methods in healthcare data; (ii) the derivation of theoretical models, and (iii) the conception and execution of experimental studies for the understanding of SCI secondary conditions. Projects apply advanced machine learning, sensing technology, and robotics towards digital twins in digital health care and rehabilitation. Interested students may check our open student projects or check our research projects and contact the responsible person directly for further questions.

Your own project ideas?

It is always possible to find a project for motivated students with their own ideas in the fields of assistive health care and rehabilitation technologies, advanced machine learning modelling, and applied robotics in health care. Please contact Dr Diego Paez if you would like to pursue a project which is not listed below.

Sirop Links

Currently, the following student projects are available. Please contact the responsible supervisor and apply with your CV and transcripts.

Master Thesis / Internship / Semester Project: Digitization of large 12-lead ECG Image database

12-lead electrocardiograms (ECGs) are still solely documented on paper in many hospitals, especially in the Global South. These physical paper records provide a multitude of conditions recorded in many different countries. Our lab has access to a dataset with more than 8000 patient’s ECG photos / scans of 12-lead signals printed onto physical paper sheets. The dataset comprises 12-lead ECG image records from more than 35 hospital sites across Europe. The primary objective of this project is to develop an automated digitization pipeline from raw image scan in .png format towards 12 vectorized ECG time series in WFDB format.

Keywords

Spinal Cord Injury, Computer Vision, CV, Machine Learning, Deep Learning, AI, Signal Processing, ECG, Medical Data, Healthcare

Labels

Semester Project , Internship , Bachelor Thesis , Master Thesis

Description

**Problem Definition** 12-lead electrocardiograms (ECGs) are still solely documented on paper in many hospitals, especially in the Global South. These physical paper records provide a multitude of conditions recorded in many different countries. Our lab has access to a dataset with more than 8000 patient’s ECG photos / scans of 12-lead signals printed onto physical paper sheets. The dataset comprises 12-lead ECG image records from more than 35 hospital sites across Europe. The primary objective of this project is to develop an automated digitization pipeline from raw image scan in .png format towards 12 vectorized ECG time series in WFDB format. The 12 lead ECG scans are usually noisy and contain a red or black grid depending on the photo format (B / BW). Some ECG images were taken by a camera while the sheet was held in the hand. For these images, the paper is often tilted and the image contains shadow artifacts or wrinkles. Furthermore, the 12 lead signals were not printed in a standardized format. Often, the signals are printed in a 3x4 matrix format onto paper with the lead identifier letters written next to the start loca-tion of the signal. Sometimes, all leads are printed in full length above each other, however. A robust solution for these prob-lems must be developed and a preliminary draft of the pipeline steps is explained in the ‘Your Tasks’ section. **Your Tasks** 1. Conduct extensive literature research on State-of-the-Art ECG digitization systems. 2. Develop an automatic image rotation algorithm for horizontal alignment. 3. Code a network for grid removal. In this step, the grid size should be extracted in pixels for later ECG signal rescaling purposes. 4. Propose a solution for denoising the images after grid-removal. The remaining image should only consist of the lead signals and the lead identifier letters. 5. Apply a pretrained character recognition CV algorithm for character recognition. 6. Detect the relevant image regions for each individual ECG lead signal. 7. Develop an algorithm for conversion of the signal into physical units and potential subsequent post-processing steps. 8. Test the performance of the digitization compared to manually extracted features that have been recorded in the hospital. 9. Digitize the whole dataset including images of more than 8000 patients recorded at 30+ hospital sites across Eu-rope. 10. Make the tool accessible online / via app. (optional) **Your Profile** - Strong experience with Python - Strong background in Computer Vision (preferred) - Background in Signal Processing and Filtering (preferred) - Structured and reliable working style - Ability to work independently on a challenging problem

Goal

Contact Details

More information

Open this project...

Published since: 2024-04-22 , Earliest start: 2024-05-01 , Latest end: 2024-11-01

Organization Spinal Cord Injury & Artificial Intelligence Lab

Hosts Paez Diego, Dr.

Topics Medical and Health Sciences , Information, Computing and Communication Sciences , Engineering and Technology

Master Thesis / Internship: Automated Time Series Analysis in Urinary Tract Assessment in Spinal Cord Injury

The primary objective of this project is to develop an automated pipeline for the identification and recognition of patterns within urodynamic recordings, utilizing urodynamic recording data in conjunction with annotated patterns provided by experts. This endeavor seeks to reduce the susceptibility of interpreting urodynamic recordings to potential errors arising from human judgment and inaccuracies, thereby improving the management of urinary tract complications in patients with spinal cord injury. By implementing a systematic approach to pattern recognition in Bladder Valomue/Pressure Time Series Measurements of urodynamic data, the potential for error in decision-making can be significantly reduced.

Keywords

Spinal Cord Injury, Machine Learning, Deep Learning, Pattern Recognition, Feature Engineering, Time Series Analysis, Signal Processing

Labels

Semester Project , Internship , Master Thesis

Description

**Background** Spinal cord injury (SCI) is a severe medical condition that affects millions of people worldwide. These patients often suffer from other health problems and comorbidities such as urinary tract complications which should be monitored so that the risk is minimized. A urodynamic test is a diagnostic tool used to evaluate the function of the bladder and urethra. The test measures flow and pressure in the urinary tract and can help determine the underlying cause of urinary dysfunction. The major complication of neurogenic lower urinary tract dysfunction is a risk to the upper tracts, so baseline urodynamic studies are performed in patients with SCI to assess symptoms and risk to the upper tracts, so that management can be planned accordingly. The interpretation of urodynamic measurements and the identification of patterns in the resulting graphs are critical components of the decision-making process regarding the administration of the test. However, this process is subject to a significant degree of error due to the involvement of human judgment. There is a pressing need for the development of an automated, systematic approach to pattern recognition in urodynamic data that would reduce the potential for error in decision-making. **Your Tasks** - Retrieve the urodynamic recordings data in conjunction with annotations and patterns provided by experts. - Conduct an extensive review of the current literature on pattern recognition in time series data. - Perform pre-processing procedures on the time-series recordings to eliminate artifacts and enhance data quality. - Develop and implement pipelines utilizing deep learning or feature extraction techniques for accurate pattern recognition in time series data (employing supervised pattern recognition methods). - Fine-tune the developed pipeline to ensure its user-friendly nature, enabling seamless adoption in future studies. - Compile a comprehensive report that highlights the achievements, methodologies, and outcomes of the study. - (Optional) Prepare a manuscript summarizing the project's findings for submission to a prominent and relevant academic journal, aiming to disseminate the research outcomes to the wider scientific community. **Your Benefits** - Gain unique access and first-hand experience in one of the leading institutions on long-term health management - At the Swiss Paraplegic Centre at Nottwil. - Working with the potential of long-term applications from the results of this thesis work. **Your Profile** - Proven understanding of signal analysis and pattern recognition. - Strong experience with Python (preferred) - Structured and reliable working style - Ability to work independently on a challenging topic.

Goal

Contact Details

More information

Open this project...

Published since: 2024-04-21 , Earliest start: 2024-05-19 , Latest end: 2024-12-31

Applications limited to Agroscope , Berner Fachhochschule , CERN , Corporates Switzerland , CSEM - Centre Suisse d'Electronique et Microtechnique , Department of Quantitative Biomedicine , Eawag , Empa , EPFL - Ecole Polytechnique Fédérale de Lausanne , ETH Zurich , Fernfachhochschule , Forschungsinstitut für biologischen Landbau (FiBL) , Friedrich Miescher Institute , Hochschulmedizin Zürich , IBM Research Zurich Lab , Institute for Research in Biomedicine , Lucerne University of Applied Sciences and Arts , NCCR Democracy , NGOs Switzerland , Pädagogische Hochschule St.Gallen , Paul Scherrer Institute , Physikalisch-Meteorologisches Observatorium Davos , Sirm Institute for Regenerative Medicine , Swiss Federal Institute for Forest, Snow and Landscape Research , Swiss Institute of Bioinformatics , Swiss National Science Foundation , SystemsX.ch , Università della Svizzera italiana , Université de Neuchâtel , University of Basel , University of Berne , University of Fribourg , University of Geneva , University of Lausanne , University of Lucerne , University of St. Gallen , University of Zurich , Wyss Translational Center Zurich , Zurich University of Applied Sciences , Zurich University of the Arts , University of Konstanz , Technische Universität München , TU Berlin , Eberhard Karls Universität Tübingen , European Molecular Biology Laboratory (EMBL) , FH Aachen , Humboldt-Universität zu Berlin , Justus Liebig University, Gießen , Ludwig Maximilians Universiy Munich , Martin Luther Universitat, Halle , Max Delbruck Center for Molecular Medicine (MDC) , Max Planck Society , Otto Von Guericke Universitat, Magdeburg , RWTH Aachen University , Social Science Research Center Berlin , Technische Universität Hamburg , TU Darmstadt , TU Dresden , Universität der Bundeswehr München , Universität Ulm , Universität zu Lübeck , University of Cologne , University of Erlangen-Nuremberg , University of Hamburg , Universtity of Bayreuth , Delft University of Technology , Maastricht Science Programme , Radboud University Nijmegen , Utrecht University , Max Planck ETH Center for Learning Systems , European Molecular Biology Laboratory , IEE S.A. Luxembourg , Istituto Italiano di Tecnologia , Technical University of Denmark , Technion - Israel Institute of Technology , University of Southern Denmark , Imperial College London , UCL - University College London , University of Oxford , University of Cambridge , National Institute for Medical Research

Organization Spinal Cord Injury & Artificial Intelligence Lab

Hosts Paez Diego, Dr. , Paez Diego, Dr.

Topics Medical and Health Sciences , Information, Computing and Communication Sciences , Engineering and Technology

Student Research Assistance for App development in Biosensing and Healthcare Data (~12 months)

Join a team of scientists improving the long-term prognosis and treatment of Spinal Cord Injury (SCI) through mobile and wearable systems and personalized health monitoring. Joining the SCAI Lab part of the Sensory-Motor Systems Lab at ETH, you will have the unique opportunity of working at one of the largest and most prestigious health providers in Switzerland: Swiss Paraplegic Center (SPZ) in Nottwil (LU).

Keywords

App development, Machine Learning, Data bases, Data engineering, Systems Engineering, Data Modelling

Labels

Internship , Lab Practice , Student Assistant / HiWi , ETH Zurich (ETHZ)

Description

Goal

Contact Details

More information

Open this project...

Published since: 2024-04-11 , Earliest start: 2024-06-01 , Latest end: 2025-06-30

Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne , IBM Research Zurich Lab , Institute for Research in Biomedicine , Hochschulmedizin Zürich , Swiss Institute of Bioinformatics , University of Lucerne , University of Zurich , Zurich University of Applied Sciences , Zurich University of the Arts , Lucerne University of Applied Sciences and Arts , Berner Fachhochschule

Organization Spinal Cord Injury & Artificial Intelligence Lab

Hosts Paez Diego, Dr. , Paez Diego, Dr. , Paez Diego, Dr.

Topics Information, Computing and Communication Sciences

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

Labels

Internship , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)

Project Background

Your Task

Your Benefits

Your Profile

Contact Details

More information

Open this project...

Published since: 2024-04-03 , Earliest start: 2024-06-01 , Latest end: 2025-03-31

Applications limited to EPFL - Ecole Polytechnique Fédérale de Lausanne , ETH Zurich , CERN , Corporates Switzerland , IBM Research Zurich Lab , NGOs Switzerland , Zurich University of Applied Sciences , Wyss Translational Center Zurich , University of Zurich , University of St. Gallen , University of Lucerne , University of Lausanne , University of Geneva , University of Fribourg , University of Berne , University of Basel , Université de Neuchâtel , Università della Svizzera italiana , Swiss National Science Foundation , Swiss Institute of Bioinformatics , Empa , Eawag , TU Berlin , Technische Universität München , Technische Universität Hamburg , RWTH Aachen University , Max Delbruck Center for Molecular Medicine (MDC) , Delft University of Technology , UCL - University College London , University of Cambridge , University of Oxford , University of Leeds , University of Manchester , University of Nottingham , National Institute for Medical Research , Imperial College London , Radboud University Nijmegen , Maastricht Science Programme

Organization Sensory-Motor Systems Lab

Hosts Paez Diego, Dr. , Paez Diego, Dr.

Topics Medical and Health Sciences , Information, Computing and Communication Sciences , Engineering and Technology

Master Thesis/ Internship: Quantifying Biomechanics of Gait from IMU Data Simulation

Gait analysis is crucial for evaluating walking ability in individuals with ambulatory conditions e.g., Stroke or Parkinson’s disease. Traditional marker-based motion capture systems face limitations in real-life scenarios. This project proposes using wearable IMUs for gait analysis due to their portability. The goal is use datasets already recorded in our labs to model and synchronize IMU data and with 3D motion capture recordings, extract meaningful gait features for different pathologies. The extracted gait features will be validated against, and validate them against a motion capture-based ground truth features calculated for the same patients. This research aims to enhance gait analysis outside of labs and provide valuable insights for decision-making in gait disorders.

Keywords

Gait Analysis, Inertial Measurement Unit, Wearable Sensors, Signal Processing, Pattern Recognition, Machine Learning, Time Series Analysis

Labels

Internship , Master Thesis

PLEASE LOG IN TO SEE DESCRIPTION

More information

Open this project...

Published since: 2024-04-03 , Earliest start: 2024-06-01 , Latest end: 2025-03-31

Applications limited to Agroscope , Berner Fachhochschule , CERN , Corporates Switzerland , CSEM - Centre Suisse d'Electronique et Microtechnique , Department of Quantitative Biomedicine , Eawag , Empa , EPFL - Ecole Polytechnique Fédérale de Lausanne , ETH Zurich , Fernfachhochschule , Forschungsinstitut für biologischen Landbau (FiBL) , Friedrich Miescher Institute , Hochschulmedizin Zürich , IBM Research Zurich Lab , Institute for Research in Biomedicine , Lucerne University of Applied Sciences and Arts , NCCR Democracy , NGOs Switzerland , Pädagogische Hochschule St.Gallen , Paul Scherrer Institute , Physikalisch-Meteorologisches Observatorium Davos , Sirm Institute for Regenerative Medicine , Swiss Federal Institute for Forest, Snow and Landscape Research , Swiss Institute of Bioinformatics , Swiss National Science Foundation , SystemsX.ch , Università della Svizzera italiana , Université de Neuchâtel , University of Basel , University of Berne , University of Fribourg , University of Geneva , University of Lausanne , University of Lucerne , University of St. Gallen , University of Zurich , Wyss Translational Center Zurich , Zurich University of Applied Sciences , Zurich University of the Arts , Eberhard Karls Universität Tübingen , European Molecular Biology Laboratory (EMBL) , FH Aachen , Humboldt-Universität zu Berlin , Justus Liebig University, Gießen , Ludwig Maximilians Universiy Munich , Martin Luther Universitat, Halle , Max Delbruck Center for Molecular Medicine (MDC) , Max Planck Society , Otto Von Guericke Universitat, Magdeburg , RWTH Aachen University , Social Science Research Center Berlin , Technische Universität Hamburg , Technische Universität München , TU Berlin , TU Darmstadt , TU Dresden , Universität der Bundeswehr München , Universität Ulm , Universität zu Lübeck , University of Cologne , University of Erlangen-Nuremberg , University of Hamburg , University of Konstanz , Universtity of Bayreuth , Delft University of Technology , Maastricht Science Programme , Radboud University Nijmegen , Utrecht University , Chalmers University of Technology , Champalimaud Foundation , CNRS - Centre national de la recherche scientifique , European Molecular Biology Laboratory , Grenoble Institute of Technology (G-INP) - Phelma , IDEA League , IEE S.A. Luxembourg , Max Planck ETH Center for Learning Systems , Politecnico di Milano , Research Internships at HU Berlin , Technical University of Denmark , Technion - Israel Institute of Technology , The Microsoft Research – University of Trento Centre for Computational and Systems Biology (COSBI) , Université de Strasbourg , Universiteit Stellenbosch , University College Dublin , University of Southern Denmark , Vienna Biocenter - Scientific Training , Uppsala Universitet

Organization Spinal Cord Injury & Artificial Intelligence Lab

Hosts Paez Diego, Dr.

Topics Medical and Health Sciences , Information, Computing and Communication Sciences , Engineering and Technology

Master Thesis/ Internship: Causal Machine Learning with Experts in the Loop for Spinal Cord Injury (SCI) Comorbitities

Despite the growing amount of work on applying causal discovery method with expert knowledge to areas of interest, few of them inspect the uncertainty of expert knowledge (what if the expert goes wrong?). This is highly important since that in scientific fields, causal discovery with expert knowledge should be cautious and an approach taking expert uncertainty into account will be more robust to potential bias induced by individuals. Therefore, we aim to develop an iterative causal discovery method with experts in the loop to enable continual interaction and calibration between experts and data. Based on the qualifications of the candidates, we can arrange a subsidy/allowance for covering traveling or living costs.

Keywords

Causal Discovery, Expert Knowledge, Iterative Algorithm, Spinal Cord Injury

Labels

Semester Project , Internship , Master Thesis

Project Background

Causal graphical models (CGM) can be determined via randomized controlled experiments (and this is the golden standard to determine the causal relations and quantify the causal effects), learned purely from data[1](i.e., structural learning or causal discovery) or elicited by domain experts based on their knowledge and experiences. It is also feasible to learn the CGM from data with prior expert knowledge when data become not reliable due to insufficient sample size or measurement errors. Early works in causal discovery utilized constraint-based approaches, such as the PC algorithm and the FCI algorithm. These methods exploit conditional independence relationships among variables to infer causal structures. They can restrict the output graph space through conditional independence assumptions or background knowledge about causal relationships [2]. Score-based approaches, such as the GES (Greedy Equivalence Search) algorithm, aim to find the best causal structure that optimizes a scoring criterion, e.g., the BIC (Bayesian Information Criterion) score. These methods often incorporate expert knowledge through prior specifications or domain-specific constraints on the model parameters [3, 4, 5]. Recent advancements in causal discovery involve hybrid approaches that combine elements of constraint-based and score-based methods. These approaches aim to leverage the strengths of both approaches to improve the accuracy and efficiency of causal structure learning. They may incorporate expert knowledge through the integration of domain-specific constraints and scoring functions. In addition to observational data, causal discovery methods have also been developed to handle experimental data, where interventions or randomized experiments are conducted. These methods utilize the causal effects of interventions to infer causal relationships. Expert knowledge can be incorporated by designing well-controlled experiments or interventions based on domain expertise. Despite the growing amount of work on structural learning with expert knowledge in scientific fields, few of them inspect the uncertainty of expert knowledge (what if the expert goes wrong?). This is highly important since in medical applications, causal discovery with expert knowledge should be taken cautiously and sometimes experts can give false knowledge, leading to serious mistakes. An approach considering the expert uncertainty during CGM construction will be more robust to potential bias induced by individuals. This idea is made possible by recent progress toward evaluating and falsifying a given CGM from observational data without ground truth[6]. Therefore, we hope to develop an iterative causal discovery method with experts in the loop for continual interaction and calibration between experts and data. In the meantime, we are also trying to develop an interactive user interface for medical experts to input their knowledge and edit the learned graph. References [1] Glymour, C., Zhang, K., & Spirtes, P. (2019). Review of causal discovery methods based on graphical models. Frontiers in genetics, 10, 524. [2] Flores, M. J., Nicholson, A. E., Brunskill, A., Korb, K. B., & Mascaro, S. (2011). Incorporating expert knowledge when learning Bayesian network structure: a medical case study. Artificial intelligence in medicine, 53(3), 181-204. [3] O’Donnell, R. T., Nicholson, A. E., Han, B., Korb, K. B., Alam, M. J., & Hope, L. R. (2006). Causal discovery with prior information. In AI 2006: Advances in Artificial Intelligence: 19th Australian Joint Conference on Artificial Intelligence, Hobart, Australia, December 4-8, 2006. Proceedings 19 (pp. 1162-1167). Springer Berlin Heidelberg. [4] Perković, E., Kalisch, M., & Maathuis, M. H. (2017). Interpreting and using CPDAGs with background knowledge. arXiv preprint arXiv:1707.02171. [5] Kleinegesse, S., Lawrence, A. R., & Chockler, H. (2022). Domain Knowledge in A*-Based Causal Discovery. arXiv preprint arXiv:2208.08247. [6] Eulig, E., Mastakouri, A. A., Blöbaum, P., Hardt, M., & Janzing, D. (2023). Toward Falsifying Causal Graphs Using a Permutation-Based Test. arXiv preprint arXiv:2305.09565.

Your Task

1. Literature Review on Causal Discovery: You will study classic and state-of-the-art causal discovery methods (preferably on real datasets with mixed-type variables) that incorporate expert knowledge, leading to a systematic review of causal discovery methods in different categories (constraint-based vs. score-based) with different assumptions (e.g., latent variables, causal mechanism/function forms, and cycles) for different settings (e.g., static data or time series). 2. Data Exploration and Analysis: You will explore analyze and datasets collected for HAPI (Hospital-Acquired Pressure Injury) and Hospital Stay after PI Surgeries. The data record hundreds of patients into dozens of mixed-type variables including patients’ demographic features, lab testing values, and health conditions observed during their stay. Apart from clinical data, you will also have access to multivariate time series extracted from bio-signals on SCI people under intervention experiments. We aim to learn the causal relations among these variables of interest from the above data with experts’ knowledge. Additionally, we will probe some public datasets of other fields such as heart diseases and earth science with ground-truth graphs. 3. Methodology Development and Deployment: You will help develop and validate an iterative algorithm for causal discovery with the expert in the loop against benchmarks using the simulated and real datasets. There are two potential exploring directions: 1. How will the correct expert knowledge help reduce the search space of graphs to improve the accuracy and efficiency of the algorithm? 2. How to encode prior knowledge or design a post-modification strategy with uncertainty from experts’ input to increase the robustness of the algorithm? Based on the methodology, an interactive web/application interface for doctors to input their knowledge and edit the learned graph is expected. If possible, you will assist in deriving the mathematical guarantee for the iterative algorithm in terms of convergence and stability properties. 4. Presentation and Documentation: You will prepare a high-quality manuscript for publication with a clear and engaging presentation of the results and methodology.

Your Benefits

Your Profile

Contact Details

More information

Open this project...

Published since: 2024-03-13 , Earliest start: 2024-04-15 , Latest end: 2024-10-15

Organization Sensory-Motor Systems Lab

Hosts Paez Diego, Dr. , Paez Diego, Dr.

Topics Medical and Health Sciences , Mathematical Sciences , Information, Computing and Communication Sciences

Temporal Graphical Modeling for Understanding and Preventing Autonomic Dysreflexia

This project will be based on the preliminary results obtained from a previous master project in causal graphical modeling of autonomous dysreflexia (AD). The focus of the extension would be two-fold. One is improving the temporal graphical reconstruction for understanding the mechanism of AD. The other one is building a forecasting framework for the early detection and prevention of AD using the graph structure we constructed before. Please refer to the attached document for more details about the task description. Based on the candidate's qualifications, funding/allowance can be provided.

Keywords

Graphical Modeling; Graph Neural Networks; Multivariate Time Series; Spinal Cord Injuries; Autonomic Dysreflexia; Wearable Sensing

Labels

Semester Project , Internship , Master Thesis , ETH Zurich (ETHZ)

Description

Graphical modeling (GM) refers to a statistical modeling framework that uses probabilistic graphs to encode conditional dependencies and independences among variables of interest [1]. Due to its ability to represent complex probability distributions, GM has been developed and adapted to various fields including earth science [2], biology [3], health [4], and medicine [5] to dissect the structure of targeted phenomena to improve the understanding of the mechanism and decision-making process. Multivariate Time series (MTS), characterized by sequential observations of several variables over time, is pervasive in various disciplines. Understanding variables' temporal evolution and spatial dependencies is essential for uncovering patterns, trends, and anomalies that unfold over time. Time series analysis provides insights into the underlying dynamics of phenomena, enabling researchers and practitioners to make informed predictions, learn temporal dependencies, and formulate effective strategies. The exploration of temporal graphical models (TGMs) in the realm of time series analysis is paramount, given the crucial role that time series data plays in various fields. In this context, TGMs emerge as a sophisticated approach to modeling time series data, capturing the dynamic and evolving relationships among variables. As a validation scenario, we are working with the Swiss Paraplegic Centre (SPZ) to improve the long-term prognosis of individuals with spinal cord injury (SCI). Secondary health conditions (SHC) such as pulmonary infections, pressure injuries (PI), AD, and cardiovascular dysregulation are highly prevalent among the SCI population, imposing a tremendous risk on the patient’s health and the healthcare system [6,7]. Through the medical records and multimodal-sensory monitoring system, we expect to develop an intelligent system that can fuse different data sources, interact with clinical operation and expertise, and guide the medical treatment and decision-making for SHC management of SCI individuals. This project will focus on the TGM of a multivariate time series extracted from a wearable multi-modal sensing system. We aim to learn the high-resolution temporal dynamics together with spatial dependencies among biometrics, based on which the early detection and inference of AD episodes could be enabled. For more details about the task description, please refer to the attached document!

Goal

Contact Details

More information

Open this project...

Published since: 2024-02-28 , Earliest start: 2024-04-01 , Latest end: 2024-10-01

Organization Spinal Cord Injury & Artificial Intelligence Lab

Hosts Paez Diego, Dr. , Li Yanke , Paez Diego, Dr. , Paez Diego, Dr.

Topics Medical and Health Sciences , Information, Computing and Communication Sciences

Disease Onset Forecasting through Graphical Modeling Based Digital Twin from Biomedical Data for Spinal Cord Injury Individuals

This project focuses on developing an explainable Artificial Intelligence (xAI) framework based on graphical modeling (GM), to enhance the capacity and capability of medical AI. Collaborating with the Swiss Paraplegic Centre (SPZ) for validation, our goal is to improve the long-term prognosis of spinal cord injury (SCI) individuals. Through medical records and a multimodal sensory monitoring system, we aim to create digital twins capable of integrating diverse data sources, guiding medical treatment, and addressing common secondary health conditions in the SCI population. The envisioned GM-based digital twin (GMDT) will represent hierarchical relations across demographic features, functional abilities, daily activities, and health conditions for SCI individuals, allowing for downstream tasks such as prediction, causal inference, and counterfactual reasoning. The assimilation and evolution between the physical and digital twins will be implemented under the GM framework, promising advancements in personalized healthcare strategies and improved outcomes for SCI people. Please refer to the attached document for more details about the task description. Based on the candidate's qualifications, funding/allowance can be provided.

Keywords

Graphical Modelling, Digital Twins, Causal Inference, Data Fusion, Multimodal Learning, Physiological Modelling, Spinal Cord Injuries, Digital Healthcare

Labels

Semester Project , Internship , Master Thesis , ETH Zurich (ETHZ)

Description

Graphical modeling (GM) was originally proposed as a statistical modeling framework that uses probabilistic graphs to encode conditional dependences and independence among variables of interest [1]. It is straightforward for stakeholders to understand the interaction among different factors and reasons using the structure and relations. Due to the strong ability of GM to represent complex probability distributions, it has been developed and adapted to various fields including earth science [2], biology [3], and healthcare [4] to dissect the structure and sparsity of targeted phenomenon to improve the understanding of mechanism and decision-making process. In digital healthcare and medicine with a high-stake nature, it is imperative to enable a transparent and accessible modeling paradigm such that both modeling technicians and medical experts can comprehend, trust, and improve the digital models for decision-making. Given the transparent structural representation of GM, we aim to develop an explainable Artificial intelligence (xAI) framework based on CGM (Causal Graphical Models, a special type of GM where edges represent causal relations) to cope with the challenges of real-world medical data and improve both the capacity and capability of medical AI. As a validation scenario of our approach, we are working with the Swiss Paraplegic Centre (SPZ) to improve the long-term prognosis of individuals with spinal cord injury (SCI). Secondary health conditions such as pulmonary infections, pressure injuries (PI), autonomic dysreflexia (AD), and cardiovascular dysregulation are highly prevalent among the SCI population, imposing a tremendous risk on the patient’s health and the healthcare system [5]. Through the medical records and multimodal-sensory monitoring system, we expect to end up with the digital twins of SCI individuals that can fuse different data sources, interact with clinical operations and expertise, and guide medical treatment. Your Tasks: 1. Literature Review on Digital Twins and Graphical Models: You will study literature related to graphical models with applications, digital twins with applications, and preferably the combination of both fields (GMDT). 2. Data Exploration and Analysis: You will explore multimodal datasets collected for PI and AD. The PI data record the conditions of hundreds of SCI individuals during their hospital stay. There are mixed-type variables including demographic features, lab testing values, and health conditions observed during their stay. Additionally, you will have access to multivariate time series extracted from bio-signals on SCI people under intervention experiments. 3. Methodology Development and Deployment: You will help define what data to use, design the experiments, develop, and validate both the modeling foundation and implementation pipeline for the GMDT. A graphical user interface is expected to be built based on the current framework we are developing in the lab. 4. Presentation and Documentation: You will prepare a high-quality manuscript for publication with a clear and engaging presentation of the methodology and validation results. References: [1] Glymour, C., Zhang, K. & Spirtes, P. Review of Causal Discovery Methods Based on Graphical Models. Front. Genet. 10, 524 (2019). [2] Runge, J. et al. Inferring causation from time series in Earth system sciences. Nat. Commun. 10, 2553 (2019). [3] Friedman, N. Inferring Cellular Networks Using Probabilistic Graphical Models. Science 303, 799–805 (2004). [4] Richens, J. G., Lee, C. M. & Johri, S. Improving the accuracy of medical diagnosis with causal machine learning. Nat. Commun. 11, 3923 (2020). [5] Haisma, J. et al. Complications following spinal cord injury: occurrence and risk factors in a longitudinal study during and after inpatient rehabilitation. J. Rehabil. Med. 39 5, 393–8 (2007).

Goal

Contact Details

More information

Open this project...

Published since: 2024-02-28 , Earliest start: 2024-03-15 , Latest end: 2024-09-30

Applications limited to TU Dresden , TU Darmstadt , TU Berlin , Technische Universität München , Technische Universität Hamburg , RWTH Aachen University , Max Planck Society , Ludwig Maximilians Universiy Munich , Humboldt-Universität zu Berlin , Eberhard Karls Universität Tübingen , Universität zu Lübeck , Imperial College London , UCL - University College London , University of Oxford , University of Cambridge , Delft University of Technology , Zurich University of Applied Sciences , Wyss Translational Center Zurich , University of Zurich , Swiss Institute of Bioinformatics , IBM Research Zurich Lab , ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne , Empa , Corporates Switzerland , Zurich University of the Arts , University of St. Gallen , University of Lausanne , University of Geneva , University of Fribourg , University of Berne , University of Basel , Swiss National Science Foundation , Swiss Federal Institute for Forest, Snow and Landscape Research , Paul Scherrer Institute , CERN , Department of Quantitative Biomedicine , Eawag , University of Konstanz , University of Cologne , University of Erlangen-Nuremberg , University of Hamburg , Universtity of Bayreuth , Universität Ulm , Universität der Bundeswehr München , Social Science Research Center Berlin , National Institute for Medical Research , Royal College of Art , University of Leeds , University of Manchester , University of Nottingham , University of Aberdeen , Utrecht University , Radboud University Nijmegen , Maastricht Science Programme , Stanford University , Yale University , CNRS - Centre national de la recherche scientifique , Massachusetts Institute of Technology , Max Planck ETH Center for Learning Systems , The University of Tokyo , Tsinghua University , Peking University , Politecnico di Milano , Princeton University , Harvard , University of Toronto , University of Copenhagen , University of California, Berkeley , The University of Edinburgh , Technical University of Denmark , The University of Melbourne , The Australian National University , National University of Singapore , Nanyang Technological University

Organization Spinal Cord Injury & Artificial Intelligence Lab

Hosts Li Yanke , Paez Diego, Dr. , Paez Diego, Dr.

Topics Information, Computing and Communication Sciences , Engineering and Technology , Behavioural and Cognitive Sciences

Education

Internships, Master’s Thesis, Bachelor & Semester Projects

Your own project ideas?

Sirop Links

Master Thesis / Internship / Semester Project: Digitization of large 12-lead ECG Image database

Keywords

Labels

Description

Goal

Contact Details

More information

Master Thesis / Internship: Automated Time Series Analysis in Urinary Tract Assessment in Spinal Cord Injury

Keywords

Labels

Description

Goal

Contact Details

More information

Student Research Assistance for App development in Biosensing and Healthcare Data (~12 months)

Keywords

Labels

Description

Goal

Contact Details

More information

Internship/ Master Thesis: Machine Learning for Assessment of Walking Patterns in the SCI population - Time Series Classification

Keywords

Labels

Project Background

Your Task

Your Benefits

Your Profile

Contact Details

More information

Master Thesis/ Internship: Quantifying Biomechanics of Gait from IMU Data Simulation

Keywords

Labels

PLEASE LOG IN TO SEE DESCRIPTION

More information

Master Thesis/ Internship: Causal Machine Learning with Experts in the Loop for Spinal Cord Injury (SCI) Comorbitities

Keywords

Labels

Project Background

Your Task

Your Benefits

Your Profile

Contact Details

More information

Temporal Graphical Modeling for Understanding and Preventing Autonomic Dysreflexia

Keywords

Labels

Description

Goal

Contact Details

More information

Disease Onset Forecasting through Graphical Modeling Based Digital Twin from Biomedical Data for Spinal Cord Injury Individuals

Keywords

Labels

Description

Goal

Contact Details

More information

Internships, Master’s Thesis,
Bachelor & Semester Projects