Skip to main content

Machine Learning System for Decision Support and Computational Automation of Early Cancer Detection and Categorisation in Colonoscopy (AIdDeCo)

The general objective of the AIdDeCo project is to develop methodologies and software tools to advance the state-of-the-art in analysis of colonoscopy data, aiding further development of efficient colonoscopy screening procedures. The project also aims to create an effective  interdisciplinary research focus around endoscopic data analysis, with a network of collaborators from computing, engineering, physics and clinical disciplines, operating as a hub to co-ordinate exchange of knowledge, people and data between academic and clinical institutions.

For more information about the project and other related research please contact Professor Bogdan Matuszewski or visit the Computer Vision and Machine Learning (CVLM) webpage.

Facts

The Machine LeArnIng System for decision Support and Computational Automation of Early Cancer Detection and Categorisation in Colonoscopy (AIdDeCo) stems from the Engineering and Computational Science for Oncology Network (ECSON), originally funded by the UK Engineering and Physical Sciences Research Council (EPSRC), grant No. EP/F013698/1. ESCON has created a platform for the exchange of ideas and staff, which led to the CVML group developing interest in endoscopic data analysis resulting in successful participation in MICCAI Endoscopic Vision Grand Challenges and the recent completion of a PhD project. All these activities, and the ongoing collaborations within the ECSON, have led to the AIdDeCo project, with the funding awarded by the Science and Technology facilities Council (STFC) Cancer Diagnosis Network+ (CDN+).

Project start date: 1 November 2020

Project end date: 30 April 2022

0

The key objectives and expected project innovations are to advance the current state-of-the art in deep machine learning applied to colonoscopy data. The focus is on developing techniques for automatic detection, segmentation and categorisation of polyps, aiming to reduce chances of polyps being missed during colonoscopy procedures. Furthermore, we aim to detect image artefacts (e.g. saturation, low contrast areas, blur or specularity) and foreign objects to support endoscopists in interpreting the data during examination. We are also looking at developing visual aids supporting colon examination. These will include 3D structure visualisation as well as navigation tools based on visual odometry techniques.

The key expected output of the project is a set of software tools aimed at improving polyp detectability and therefore ultimately leading to a reduced risk of colorectal cancer. The developed software is to be tested and evaluated by clinical endoscopists, preparing the ground for future clinical studies validating the methodology in a hospital environment.

2021

  • 21 April 2021 - B. Matuszewski is to present update on the AIdDeCo at the Digital Health Workshop organised by the STFC HealthTec Cluster.

    31 March 2021 - Seminar on Vision Transformer (ViT) Architectures (part 2)

    24 March 2021 - Seminar on Vision Transformer (ViT) Architectures (part 1)

    17 March 2021 - Seminar on Flow-based generative models (part 2)

    10 March 2021 - Seminar on Flow-based generative models (part 1)

    25 February 2021 - AidDeCO project blog entry

  • Fully funded 3 year PhD Scholarship (UCLan School of Engineering), started in October 2020.

For more information about the project and other related research please contact Professor Bogdan Matuszewski or visit the Computer Vision and Machine Learning (CVLM) webpage.