Automatic recognition of irregularities in the oesophagus

Esophageal cancer is one of the most lethal tumors worldwide. When esophageal cancer is diagnosed, it is often in a late stage and only half of the patients undergo curative surgical removal of the esophagus. When discovered in an early stage, it can be treated minimally invasive without removing the esophagus with excellent outcome. Patients with Barrett’s
esophagus (BE) have an increased risk of esophageal adenocarcinoma (EAC). Therefore, they undergo regular
endoscopy for early cancer detection.

EAC in a Barrett’s esophagus is difficult to detect during surveillance endoscopies. This is partly because of its subtle appearance and partly because most endoscopists rarely encounter early BE neoplasia and therefore are unfamiliar with its endoscopic appearance . A computer aided detection (CAD) system might assist endoscopists in the recognition and subsequent characterization of early BE neoplasia, thereby improving efficacy of BE surveillance. The aim of this study is to develop a CAD algorithm using high quality endoscopic imagery of BE neoplasia.

The development of a computer aided detection (CAD) system might be able to assist endoscopists in the recognition and subsequent characterization of early Barrett's neoplasia, thereby improving efficacy of BE surveillance.

not applicable

No interventions will take place.

For this study, extra endoscopic imagery will be obtained by making endoscopic images and (zoom) videos.