Modeling of bolus movement in intra-esophageal manometry/impedance measurements

Master’s thesis in Medical Data Analysis

Background:

Patients with benign esophageal diseases often endure lengthy medical journeys before receiving a definitive diagnosis. The diagnostic process entails a spectrum of examinations, with high-resolution manometry (HRM) emerging as the foremost method for detecting esophageal motility disorders. However, the movement of a swallowed bolus is currently still primarily assessed through videofluoroscopy, an imaging technique that exposes patients to radiation and requires specialized equipment.

Goal:

The project aims to develop a computational (potentially AI-based) model to analyze the movement of a bolus through the esophagus during swallowing, using high-resolution manometry (HRM) data augmented with impedance measurements (data is available). By integrating pressure profiles and impedance-derived bolus presence data, the model will characterize bolus transit dynamics and identify physiological patterns. Special focus will be placed on the precise detection and modeling of bolus clearance, defined as the point at which the entire bolus has been fully cleared from the esophagus, indicating the true end of the swallow event. The outcome will enhance understanding of esophageal function and support clinical diagnostics through visual and quantitative analysis of bolus flow. Our long-term goal is to provide a non-radiative, data-driven alternative to videofluoroscopy by leveraging HRM and impedance data to characterize bolus dynamics.

Tasks:

• Conduct a literature review on existing research work and analogous tools and solutions
• Comprehensive analysis of provided data and identification of relevant features
• Develop method to accurately detect and model bolus movement (especially clearance) based on manometry/impedance data
• Written report documenting the methods and outcomes (thesis + potentially publication)

Requirements:

• Strong experience with Python
• Expertise in data analysis
• Preferred: Experience in Machine Learning and corresponding libraries (e.g. PyTorch)
• Keen interest in medical applications
• Self-motivation and ability to work independently

Please contact us if you are interested in this topic.

Alexander Geiger, M.Sc.
Email: alexander.geiger@tum.de

 

Date: May 20, 2025