{"id":{"repo_id":"middlesex","oai_identifier":"oai:repository.mdx.ac.uk:27y427"},"canonical_url":"https://search.dev.ndltd.org/etd/middlesex/oai:repository.mdx.ac.uk:27y427","repository":{"repo_id":"middlesex","name":"Middlesex University","base_url":"https://repository.mdx.ac.uk/oai2"},"display":{"title":"Early detection of oesophageal cancer based on colour appearance and colour-based data augmentation","abstract":"Oesophagus cancer is the 9th most common cancer and the 6th leading cause of cancer-related death. Globally, the estimated number of new cases was 572 000, of which approximately 509 000 persons died from oesophageal cancer in 2018. While the overall five-year survival rate of oesophagus cancer is less than 20%, this figure can be improved significantly to more than 90% if oesophageal cancer (OC) is detected in its intramucosal stage when lymph node metastasis is unlikely and it still can be treated endoscopically. Hence early detection of oesophageal cancer plays a crucial role. The challenge here is that early stage of caner present subtle changes in comparison with normal skill tissue and can easily missed being diagnosed (1 in 4). This research investigates the contribution of colour attribute in diagnosis of OC. Two studies are conducted. One is to distinguish colour differences between images of low grade, high grade of OC and cancer as well as normal skin colours. The second part of study focuses on the colour-based data augmentation for an artificial intelligence (AI) enhanced computer diagnosis system to enlarge data set. With regard to colour differences, all four categories overlap to a large extent. Hence it is unlikely that determination of early stage of OC can be made accurately purely based on colour changes. When training an AI system for diagnosing OC, the averaged sensitivity, specificity, and accuracy are 90.2%, 97.2% , and 95.8% respectively for classification in comparison with 85.7%, 95.6%, and 93.3% for the system trained without colour-based augmentation. Significantly, for low grade dysplasia (LGD), sensitivity and accuracy improved by over 8.1% and 4.3%, respectively.","abstract_html":"Oesophagus cancer is the 9th most common cancer and the 6th leading cause of cancer-related death. Globally, the estimated number of new cases was 572 000, of which approximately 509 000 persons died from oesophageal cancer in 2018. While the overall five-year survival rate of oesophagus cancer is less than 20%, this figure can be improved significantly to more than 90% if oesophageal cancer (OC) is detected in its intramucosal stage when lymph node metastasis is unlikely and it still can be treated endoscopically. Hence early detection of oesophageal cancer plays a crucial role. The challenge here is that early stage of caner present subtle changes in comparison with normal skill tissue and can easily missed being diagnosed (1 in 4). This research investigates the contribution of colour attribute in diagnosis of OC. Two studies are conducted. One is to distinguish colour differences between images of low grade, high grade of OC and cancer as well as normal skin colours. The second part of study focuses on the colour-based data augmentation for an artificial intelligence (AI) enhanced computer diagnosis system to enlarge data set. With regard to colour differences, all four categories overlap to a large extent. Hence it is unlikely that determination of early stage of OC can be made accurately purely based on colour changes. When training an AI system for diagnosing OC, the averaged sensitivity, specificity, and accuracy are 90.2%, 97.2% , and 95.8% respectively for classification in comparison with 85.7%, 95.6%, and 93.3% for the system trained without colour-based augmentation. Significantly, for low grade dysplasia (LGD), sensitivity and accuracy improved by over 8.1% and 4.3%, respectively.","abstract_has_math":false,"creators":["Cui, D."],"institution":"Middlesex University","degree_name":"PhD","degree_level":"PhD thesis","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2024,"date_issued":"2024","date_published":"2024","updated_at":"2026-07-24T03:03:20Z","subjects":[],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:repository.mdx.ac.uk:27y427"],"render_values":[{"text":"oai:repository.mdx.ac.uk:27y427","href":null,"code":true}]}]},"links":{"outbound_url":null,"outbound_label":null,"outbound_source":null},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["Cui, D."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2024"]},{"key":"dc:date.issued","label":"Date","values":["2024"]},{"key":"dc:publisher","label":"Institution","values":["Middlesex University Research Repository"]},{"key":"dc:publisher.department","label":"Dc Publisher Department","values":["Computer Science","Science and Technology"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["Middlesex University"]},{"key":"dc:relation","label":"Dc Relation","values":["https://repository.mdx.ac.uk/item/27y427"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://repository.mdx.ac.uk/item/27y427"]},{"key":"dc:type","label":"Dc Type","values":["Thesis or dissertation"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["PhD thesis"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["PhD"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:repository.mdx.ac.uk:27y427"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://repository.mdx.ac.uk/download/522e1e57575859be55d4edda2b8700900761cbeb07de583338f84a37c57e88ee/8311386/DCui%20thesis.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Oesophagus cancer is the 9th most common cancer and the 6th leading cause of cancer-related death. Globally, the estimated number of new cases was 572 000, of which approximately 509 000 persons died from oesophageal cancer in 2018. While the overall five-year survival rate of oesophagus cancer is less than 20%, this figure can be improved significantly to more than 90% if oesophageal cancer (OC) is detected in its intramucosal stage when lymph node metastasis is unlikely and it still can be treated endoscopically. Hence early detection of oesophageal cancer plays a crucial role. The challenge here is that early stage of caner present subtle changes in comparison with normal skill tissue and can easily missed being diagnosed (1 in 4). This research investigates the contribution of colour attribute in diagnosis of OC. Two studies are conducted. One is to distinguish colour differences between images of low grade, high grade of OC and cancer as well as normal skin colours. The second part of study focuses on the colour-based data augmentation for an artificial intelligence (AI) enhanced computer diagnosis system to enlarge data set. With regard to colour differences, all four categories overlap to a large extent. Hence it is unlikely that determination of early stage of OC can be made accurately purely based on colour changes. When training an AI system for diagnosing OC, the averaged sensitivity, specificity, and accuracy are 90.2%, 97.2% , and 95.8% respectively for classification in comparison with 85.7%, 95.6%, and 93.3% for the system trained without colour-based augmentation. Significantly, for low grade dysplasia (LGD), sensitivity and accuracy improved by over 8.1% and 4.3%, respectively."]},{"key":"dc:description.abstract","label":"Abstract","values":["Oesophagus cancer is the 9th most common cancer and the 6th leading cause of cancer-related death. Globally, the estimated number of new cases was 572 000, of which approximately 509 000 persons died from oesophageal cancer in 2018. While the overall five-year survival rate of oesophagus cancer is less than 20%, this figure can be improved significantly to more than 90% if oesophageal cancer (OC) is detected in its intramucosal stage when lymph node metastasis is unlikely and it still can be treated endoscopically. Hence early detection of oesophageal cancer plays a crucial role. The challenge here is that early stage of caner present subtle changes in comparison with normal skill tissue and can easily missed being diagnosed (1 in 4). This research investigates the contribution of colour attribute in diagnosis of OC. Two studies are conducted. One is to distinguish colour differences between images of low grade, high grade of OC and cancer as well as normal skin colours. The second part of study focuses on the colour-based data augmentation for an artificial intelligence (AI) enhanced computer diagnosis system to enlarge data set. With regard to colour differences, all four categories overlap to a large extent. Hence it is unlikely that determination of early stage of OC can be made accurately purely based on colour changes. When training an AI system for diagnosing OC, the averaged sensitivity, specificity, and accuracy are 90.2%, 97.2% , and 95.8% respectively for classification in comparison with 85.7%, 95.6%, and 93.3% for the system trained without colour-based augmentation. Significantly, for low grade dysplasia (LGD), sensitivity and accuracy improved by over 8.1% and 4.3%, respectively."]},{"key":"dc:title","label":"Title","values":["Early detection of oesophageal cancer based on colour appearance and colour-based data augmentation"]}]}],"canonical_facts":{"dc:creator":["Cui, D."],"dc:date":["2024"],"dc:date.issued":["2024"],"dc:description":["Oesophagus cancer is the 9th most common cancer and the 6th leading cause of cancer-related death. Globally, the estimated number of new cases was 572 000, of which approximately 509 000 persons died from oesophageal cancer in 2018. While the overall five-year survival rate of oesophagus cancer is less than 20%, this figure can be improved significantly to more than 90% if oesophageal cancer (OC) is detected in its intramucosal stage when lymph node metastasis is unlikely and it still can be treated endoscopically. Hence early detection of oesophageal cancer plays a crucial role. The challenge here is that early stage of caner present subtle changes in comparison with normal skill tissue and can easily missed being diagnosed (1 in 4). This research investigates the contribution of colour attribute in diagnosis of OC. Two studies are conducted. One is to distinguish colour differences between images of low grade, high grade of OC and cancer as well as normal skin colours. The second part of study focuses on the colour-based data augmentation for an artificial intelligence (AI) enhanced computer diagnosis system to enlarge data set. With regard to colour differences, all four categories overlap to a large extent. Hence it is unlikely that determination of early stage of OC can be made accurately purely based on colour changes. When training an AI system for diagnosing OC, the averaged sensitivity, specificity, and accuracy are 90.2%, 97.2% , and 95.8% respectively for classification in comparison with 85.7%, 95.6%, and 93.3% for the system trained without colour-based augmentation. Significantly, for low grade dysplasia (LGD), sensitivity and accuracy improved by over 8.1% and 4.3%, respectively."],"dc:description.abstract":["Oesophagus cancer is the 9th most common cancer and the 6th leading cause of cancer-related death. Globally, the estimated number of new cases was 572 000, of which approximately 509 000 persons died from oesophageal cancer in 2018. While the overall five-year survival rate of oesophagus cancer is less than 20%, this figure can be improved significantly to more than 90% if oesophageal cancer (OC) is detected in its intramucosal stage when lymph node metastasis is unlikely and it still can be treated endoscopically. Hence early detection of oesophageal cancer plays a crucial role. The challenge here is that early stage of caner present subtle changes in comparison with normal skill tissue and can easily missed being diagnosed (1 in 4). This research investigates the contribution of colour attribute in diagnosis of OC. Two studies are conducted. One is to distinguish colour differences between images of low grade, high grade of OC and cancer as well as normal skin colours. The second part of study focuses on the colour-based data augmentation for an artificial intelligence (AI) enhanced computer diagnosis system to enlarge data set. With regard to colour differences, all four categories overlap to a large extent. Hence it is unlikely that determination of early stage of OC can be made accurately purely based on colour changes. When training an AI system for diagnosing OC, the averaged sensitivity, specificity, and accuracy are 90.2%, 97.2% , and 95.8% respectively for classification in comparison with 85.7%, 95.6%, and 93.3% for the system trained without colour-based augmentation. Significantly, for low grade dysplasia (LGD), sensitivity and accuracy improved by over 8.1% and 4.3%, respectively."],"dc:identifier":["oai:repository.mdx.ac.uk:27y427"],"dc:identifier.uri":["https://repository.mdx.ac.uk/download/522e1e57575859be55d4edda2b8700900761cbeb07de583338f84a37c57e88ee/8311386/DCui%20thesis.pdf"],"dc:publisher":["Middlesex University Research Repository"],"dc:publisher.department":["Computer Science","Science and Technology"],"dc:publisher.institution":["Middlesex University"],"dc:relation":["https://repository.mdx.ac.uk/item/27y427"],"dc:relation.isreferencedby":["https://repository.mdx.ac.uk/item/27y427"],"dc:title":["Early detection of oesophageal cancer based on colour appearance and colour-based data augmentation"],"dc:type":["Thesis or dissertation"],"dc:type.qualificationlevel":["PhD thesis"],"dc:type.qualificationname":["PhD"]},"updated_at":"2026-07-24T03:03:20Z"}