Hussein, Mohamed; (2025) Endoscopic innovations for the detection and haemostasis of oesophageal cancer. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text Hussein_10214423_redacted_sigs_removed.pdf Download (5MB) | Preview

Abstract

Despite advances in endoscopic imaging techniques in the last decade there is still a significant miss rate of early oesophageal cancer in Barrett’s oesophagus. Several adjuncts have been developed to support endoscopists with early detection. Artificial intelligence has been shown to play an important role in increased polyp detection during colonoscopies in both pre-clinical studies and randomised control trials with several CE marked technologies now developed to be used real time in endoscopy. Research on the use of Artificial intelligence in Barrett’s oesophagus has been more limited and there is a great potential role for it to play in this area. Early detection of cancer in Barrett’s oesophagus is key as patients can then be offered curative endoscopic resection therapy. Also, if not detected early enough advanced oesophageal malignancy can have a poor prognosis. Both these scenarios can also be associated with risks of bleeding which can be managed with several standard endoscopic haemostatic modalities. Treatment options are less favourable for an oesophageal malignancy using standard techniques due to the friable nature of the mucosa. Novel haemostatic powders such as TC-325 have been developed and provide non-contact treatment options for malignancy related bleeding as well as for the treatment of difficult post endoscopic therapy bleeding. This thesis consists of a literature review around detection and treatment strategies for early cancer in Barrett’s oesophagus, and haemostatic treatment options for bleeding following endoscopic therapy in the oesophagus and secondary to advanced oesophageal malignancies. The first three studies have the overarching theme of early detection strategies for Barrett’s oesophagus. The first study sets the scene with a retrospective analysis of the accuracy of detection of low-grade dysplasia in Barrett’s oesophagus, its natural history and risk factors which help predict the risk of progression to a cancer. The second study shows the development of a novel artificial intelligence algorithms for the detection of dysplasia in Barrett’s oesophagus using targeted biopsies on overview images which is compared to the accuracy of non-expert endoscopists. The third study supports the second study where a separate artificial intelligence system is developed for the characterisation of dysplasia on magnification endoscopy which has the added benefit of supporting experts in delineating resection margins for a curative endoscopic resection. The potential benefits of the second and third studies are to provide adjuncts for earlier detection of cancer which can then be endoscopically resected. There are associated risks of bleeding following a resection. The fourth study in this thesis assesses the outcomes of the treatment of intraprocedural bleeding following endoscopic therapy in the oesophagus with TC-325 haemostatic powder as a monotherapy, part of a combination therapy or as a rescue treatment. Adjuncts like artificial intelligence will be important in minimising an increased miss rate of early cancer leading to advanced oesophageal malignancy which also have an associated complication risk of bleeding. Therefore, the fifth and final study in this thesis assesses the outcomes of the treatment of bleeding advanced oesophageal malignancies with TC-325 as part of a monotherapy, combination therapy or rescue therapy and assesses the impact of this treatment on transfusion requirements. These are commonly difficult to treat, and I assess how TC-325 can bridge this difficulty. The studies presented in this thesis collectively aim to improve the early detection of cancer in Barrett’s oesophagus and improve haemostasis techniques used during endoscopic therapy. Together, these studies contribute to both the diagnostic and therapeutic progress in the endoscopic management of oesophageal disease.

Download activity - last month

Export as CSVXMLJSON

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item