Machine Learning in Oesophagogastric Cancer Study (M-LOC Study)
Research type
Research Study
Full title
Application of Machine Learning to multidisciplinary assessment and management in Oesophagogastric Cancer
IRAS ID
319540
Contact name
Tim Underwood
Contact email
Sponsor organisation
University of Southampton
Duration of Study in the UK
2 years, 7 months, 1 days
Research summary
Oesophagogastric cancer (cancer of the food pipe and stomach) is managed by a specialist team of healthcare professionals called a multidisciplinary team (MDT). Only 39% of patients will be potentially curable at the time of first their diagnosis.
The gold standard care for patients with oesophageal and gastric cancer is surgery. Where there is a high chance of cancer recurrence after surgery , the MDT may recommend a patient is given chemotherapy and/or radiotherapy beforehand, called “Neoadjuvant therapy”, to try and minimise this risk. Unfortunately, they also risk potential side effects. Oesophagogastric cancer patients rely on high-quality decision-making in complex clinical scenarios, which impact survival and quality of life. MDTs have been proved to improve patient outcomes such as survival. However, the rising number of patients increases pressures on MDTs potentially leading to variability within this decision-making.
The purpose of this study is to use a branch of artificial intelligence called Machine Learning to develop computer models capable of replicating the human decision-making process of the current Upper Gastrointestinal Surgery MDT. Machine Learning is capable of learning complex patterns or relationships within large amounts of data. We aim to develop an assistive decision tool for use by the MDT and reduce variability therein.
This research involves observing the MDT at Southampton, a leading centre nationwide for upper gastrointestinal cancers, followed by surveying expert clinicians from MDTs all over the country (between 2023-2024). The research team will also speak to patients to gauge how they feel about this type of research in the form of 1-hour focus groups to inform how we develop and put such tools into effect in the future. This lets us unpick the most important aspects of a patient’s cancer journey, which we can use to build a trustworthy computer-based Machine Learning model using these factors.REC name
North West - Preston Research Ethics Committee
REC reference
23/NW/0197
Date of REC Opinion
3 Oct 2023
REC opinion
Further Information Favourable Opinion