The goal of this clinical trial is to assess the efficacy of hyperspectral endoscopic imaging to distinguish neoplasia from non-dysplastic Barrett's oesophagus in vivo. The main outcome measures are: 1) Classification of spectral patterns corresponding to neoplastic vs non-neoplastic Barrett's; 2) Image quality assessed by the endoscopist by VAS rating the level of confidence in delineating the area of interest; 3) Time to perform hyperspectral imaging; 4) Quantification of molecular biomarkers in endoscopic areas with neoplastic spectral patterns.

Participants will firstly undergo a standard endoscopy, followed by another endoscopy using the experimental HySe device. Subsequently, patients will receive biopsies according to Seattle protocol, and up to 6 additional (optional) snap frozen research biopsies.

This trial is designed to evaluate the feasibility of hyperspectral endoscopic imaging in vivo.

The team has developed a custom hyperspectral endoscope based around a commercially available, standard-of-care endoscopy stack from Olympus (Tokyo, Japan). In this system, the original gastroscope, video processor and display are used with a modified light source that incorporates a laser with tunable wavelength for illuminating the scene through the endoscope. A live video feed from the gastroscope is displayed on the endoscopy theatre monitors and can be acquired and saved in the native video acquisition system.

Given that it is essential in the pilot study to assess the detection of neoplastic lesions in a small patient cohort, we will artificially enrich for patients with early neoplasia by including patients with previously recognized early neoplasia within Barrett's oesophagus (high grade dysplasia and intramucosal adenocarcinoma). These patients will receive a clinically indicated endoscopy to be assessed to determine eligibility for endoscopic treatment, and imaging would be done at this time point. This will allow the team to have sufficient number of outcomes to have preliminary data on diagnostic accuracy and be able to power future studies in non-enriched patient cohorts.

Inclusion Criteria:

• Male or female subjects over 18 years.
• Previous diagnosis of Barrett's oesophagus, with an endoscopic length of at least 2 cm if circumferential (C≥2 according to Prague classification) or 3 cm if not circumferential (maximal extent M≥3, according to Prague classification).
• Previous diagnosis of oesophageal glandular dysplasia or early oesophageal adenocarcinoma for consideration of endoscopic therapy.

Exclusion Criteria:

• History of oesophageal stricture precluding passage of the endoscope.
• Pregnancy, or planned pregnancy during the course of the study.
• Currently breastfeeding.
• Any history of oesophageal varices, liver impairment of moderate or worse severity (Child's- Pugh class B & C) or evidence of varices on initial treatment endoscopy.
• Any history of oesophageal surgery, except for uncomplicated fundoplication.
• History of coagulopathy, with INR>1.3 and/or platelet count of <75,000.
• On clopidogrel, and/or warfarin for high risk condition and unable to withhold temporarily the medication.