Assessing the use of artificial intelligence in rectal MRI
Research type
Research Study
Full title
Quantitative assessment of image quality in rectal cancer MR images when using artificial intelligence reconstruction techniques
IRAS ID
345225
Contact name
Anita Wale
Contact email
Sponsor organisation
St George’s, University Hospitals NS Foundation Trust
Duration of Study in the UK
1 years, 0 months, 31 days
Research summary
Magnetic resonance imaging (MRI) is important in pre- and post-treatment evaluation of rectal cancer and is key for determining the staging and thus the most appropriate treatment pathway for the patient. It is imperative that images are of high enough quality that the tumour location and morphology can be characterised, and that surrounding tissues can be differentiated and identified easily. Furthermore, it is essential that measurements of tumour size can be accurately determined as this is an important biomarker for rectal cancer staging and influences the patient's clinical pathway.
Recent advancements in MRI include artificial intelligence (AI) reconstruction techniques. These techniques, initially introduced nationally to combat Covid backlogs, are becoming widespread due to their ability to maintain or increase MRI image quality whilst decreasing image acquisition time. This is advantageous for the clinical service, as decreased scanning time directly affects the number of patients that can be scanned in a list. It is also advantageous for the patients, particularly those who are claustrophobic, have learning difficulties, or who are in pain or discomfort. This has the effect of decreasing patient burden and increasing patient cooperation, vital for image quality.
Protocols for rectal cancer MR imaging can be long and may benefit from being shortened using these AI techniques. It is however important that image quality is validated, and that measurements are validated quantitatively, before implementing clinically. This project aims to optimise and validate these AI techniques in volunteers first, followed by qualitative and quantitative assessment by radiologists. These AI sequences will then be introduced alongside existing clinical protocols so that essential measurements can be validated quantitatively.
REC name
East of England - Cambridge South Research Ethics Committee
REC reference
24/EE/0247
Date of REC Opinion
16 Dec 2024
REC opinion
Further Information Favourable Opinion