Inflammatory Bowel Diseases
Conditions
Keywords
Inflammatory Bowel Disease, Ulcerative Colitis, Crohns Disease, Artificial Intelligence, Endoscopy, Pathology, Radiology
Brief summary
A retrospective and prospective, observational, non-interventional, cohort study to develop quantitative metrics from tools used as standard of care when diagnosing, assessing and monitoring patients with inflammatory bowel disease.
Detailed description
Inflammatory bowel disease is a chronic inflammatory disease of the gastrointestinal tract that is without cure. It comprises two main disorders: Crohn's disease and ulcerative colitis. These conditions can lead to debilitating symptoms of abdominal pain, weight loss, diarrhoea and rectal bleeding. The incidence of inflammatory bowel disease is increasing worldwide and is increasingly considered an emerging global disease. The prevalence of inflammatory bowel disease is currently the highest in North America and Western Europe and up to 600,000 people in the United Kingdom are thought to be affected. The diagnosis and monitoring of inflammatory bowel disease is based on clinical, endoscopic, radiological and histological features. Endoscopy, is considered the current gold standard in inflammatory bowel disease with newer techniques being developed. There are however limitations to these methods, especially regarding the inter and intra observer variability in assessing the validated scoring systems used from the gold standard modalities.
Interventions
None listed
Sponsors
Perspectum
West Hertfordshire Hospitals NHS Trust
Study design
Observational model
COHORT
Time perspective
OTHER
Eligibility
Sex/Gender
ALL
Age
18 Years to 80 Years
Inclusion criteria
* Male or female aged 16 or over * Patients with known or suspected Inflammatory Bowel Disease scheduled to undergo clinical, endoscopic or radiological assessment * Patients with known Inflammatory Bowel Disease under ongoing clinical monitoring, who have previously undergone clinical, endoscopic or radiological assessments. * Participant willing and able to give informed consent for participation in the study
Exclusion criteria
* Participant who is, or suspects that they are, pregnant at the time of assessment * Any other cause, including significant disease or disorder which, in the opinion of the investigator, may either put the participant at risk because of participation in the study, or may influence the participant's ability to participate in the study
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Development and presentation of computer generated metrics derived from endoscopic videos. | 24 months | To develop quantitative metrics from computational modelling of endoscopic, histological and radiological data by comparing clinical scores for inflammatory bowel disease between computer and human generated metrics to aid and improve assessment of inflammatory bowel disease. |
| Development and presentation of computer - generated metrics derived from histology slides. | 24 months | To develop quantitative metrics from computational modelling of endoscopic, histological and radiological data by comparing clinical scores for inflammatory bowel disease between computer and human generated metrics to aid and improve assessment of inflammatory bowel disease. |
| Development and presentation of computer generated metrics derived from Magnetic Resonance Elastography/ Computerised Tomography Enterography / Magnetic Resonance Imaging Pelvis scans. | 24 months | To develop quantitative metrics from computational modelling of endoscopic, histological and radiological data by comparing clinical scores for inflammatory bowel disease between computer and human generated metrics to aid and improve assessment of inflammatory bowel disease. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| The relation between human and computer generated metrics will be assessed using Cohen's Kappa (k) and correlation coefficients (Pearson's/Spearman's). | 24 months | To determine whether quantitative metrics produced from digitised versions of inflammatory bowel disease investigatory tools accurately correspond to current standard care endoscopic, radiological and histological assessment in inflammatory bowel disease. |
| Bland Altman analysis (bias, slope of bias, Limits of Agreement, the corresponding 95% Confidence Interval) and Inter-Class Correlation coefficients will be reported to estimate agreement between variables as endpoints for repeatability/reproducibility. | 24 months | To compare repeatability and reproducibility of humangenerated and computer-generated metrics |
| Statistical performance of computer diagnostic tools compared to clinical diagnoses: sensitivity, specificity, Area Under Receiver Operating Curve, predictive positive/negative value, survival curves with respect to clinical diagnosis/disease severity. | 24 months | To determine the diagnostic and prognostic ability of the novel quantitative metrics to predict/diagnose clinical outcome data in addition to patient response to therapies and surgeries. This will be derived from historical and prospective endoscopic, imaging, clinical and histopathological data. |
| Performance of metrics will be assessed using Cohen's Kappa (k) and Inter-Class Correlation coefficient. | 24 months | To determine inter-observer and intra-observer variability in human-generated and computer generated metrics |
Countries
United Kingdom
Contacts
STUDY_CHAIRJonathan Landy
West Hertfordshire Trust
Outcome results
None listed