Feasibility of CADU system and bio-impedence in Barrett's Oesophagus
Research type
Research Study
Full title
Prospective Pilot/Feasibility study on the diagnostic applicability of a Computer-aided detection device (CADU system) for detection of neoplasia in Barrett’s Oesophagus and the use of bio-impedance in risk-stratification.
IRAS ID
310124
Contact name
Venkat Subramanian
Contact email
Sponsor organisation
Leeds Teaching Hospitals NHS Trust
Duration of Study in the UK
1 years, 0 months, 1 days
Research summary
Barrett’s Oesophagus (BE) is characterised by the presence of metaplastic columnar epithelium in the oesophagus. The significance of BE is in its progression through a sequence of low-grade dysplasia to high-grade dysplasia and eventually oesophageal adenocarcinoma (OAC).
The incidence of OAC is thought to be increasing in the Western world and BE is a well-recognised pre-malignant stage of OAC. Current recommended guidelines suggest 4 random quadratic biopsies every 2cm intervals of the total MCE length (Seattle Protocol) with the use of White Light Endoscopy (WLE). However, the multifocal and patchy distribution of dysplastic lesions often leads to sampling errors and low yield.
Whilst the Seattle protocol remains the current “Gold” standard, there is a conscious need for further technological advances for early detection of oesophageal neoplasia and thus, research has been driven in this field over the past decade. Of these technologies, we aim to study two novel advancements which have the potential to improve BE surveillance – Artificial Intelligence (CADU system) and Bio-Impedance Spectroscopy.
The CADU system, developed by ODIN Vision, is a UKCA and CE marked computer aided detection device (CADe) for the oesophagus. For the detection of dysplasia in BE, preliminary data on the “Test phase” of the CADe system on 33 patients achieved an area under the ROC (AUROC) of 84.4%, with a sensitivity of 91.7% and specificity of 74.4%. When the CADU system is used with enhanced imaging (I-scan Mode 1), in 45 patients the AUROC increased to 93.5%, with sensitivity and specificity of 90.5% and 80.4% respectively.
We aim to assess the in-vivo performance and diagnostic accuracy, sensitivity, and specificity of the novel CADU system and comparison with High-definition WLE and NBI in neoplastic detection in BE. Furthermore we will also look into the feasibility of Bio-impedance Spectroscopy in the risk-stratification of BE patients.REC name
London - Chelsea Research Ethics Committee
REC reference
22/PR/0184
Date of REC Opinion
23 Feb 2022
REC opinion
Favourable Opinion