Exploring the Relationship Between Functional Disorders and Physiological Anomalies in gastrointestinal disorders using Multimodal Data Analysis

Context :

Gastrointestinal disorders pose a significant health challenge, impacting a notable portion of
the population. Despite its prevalence, there remains uncertainty regarding the connection
between functional disorders and physiological anomalies. This internship opportunity offers
a unique chance to delve into this complex interplay using cutting-edge data analysis
techniques.

Project Overview :

This internship focuses on leveraging a substantial dataset consisting of defecography X-ray films and associated medical reports from a retired radiologist. The objective is to investigate the potential link between functional disorders and physiological anomalies in fecal evacuation.

Keys Tasks :

1. Dataset Curation : 
   • Digitization of defecography X-ray films for efficient data handling and analysis.
   • Establishing a well-organized database of digitized films linked to medical reports.
2. Image Segmentation :
   • Implementing image segmentation techniques to identify relevant structures and features in defecography X-ray images.
   • Extracting key quantitative information related to bowel movements and anatomical structures.
3. Natural Language Processing (NLP):
   • Employing NLP algorithms to analyze and extract valuable insights from the medical reports provided by the radiologist.
   • Correlating textual information with image-derived data to enrich the understanding of each case.
4. Audio Processing : 
   • Exploring the potential inclusion of audio data (if available) during defecation to identify patterns or anomalies.
   • Integrating audio features into the multimodal analysis framework.
5. Collaboration : 
   • Collaborating with a digestive surgeon from CHUGA (University Hospital Grenoble Alpes) for medical insights and guidance.
   • It is highly probable that we will engage with an AI-specialized company to implement cutting-edge techniques and methodologies (collaboration under discussion).

Expected Learning Outcomes :

- Proficient use of image segmentation techniques in medical imaging.

- Hands-on experience with Natural Language Processing for medical text analysis.

- Exposure to audio processing and integration of diverse data modalities.

- Collaboration with medical professionals and AI experts.

Benefits :

- Gain practical experience in working with real-world healthcare datasets.

- Contribute to advancing our understanding of gastrointestinal disorders.

- Network with professionals in both medical and AI domains.