P198 Practical deep learning tool for the scoring of ulcerative colitis disease activity in central reading

Abstract Background Central Reading Org.& the pharma industry employ subject matter experts(SMEs)to score videos from sites participating in clinical trials for Ulcerative Colitis(UC).As we are developing Artificial Intelligence(AI)models for scoring purposes,we need to build a new software interface that can incorporate these AI models to aid SMEs,making their determination of the scores for each segment of the video & for the video as a whole.We propose a system that reduces the time for SMEs to review & score videos,improving the accuracy of scoring,with the help of our AI models. Methods We built a web-based interface supported by our AI models which can read,write multiple databases & data stores to read & display videos to be scored by a central reader, as well as the associated metadata required to improve the process.User interface shows a timeline with markers for the segments of the colon,with sections that are blurry,poorly prepped,or unscorable highlighted in different colours.While we could also highlight sections of the video with the precise score assigned to it by the AI,this would bias the central reader’s opinion.We hide the precise score generated by our AI models & instead display 3 colours for low,medium or high disease activity.When a video is loaded to be read by the user,the playback marker is set to the first high disease activity section based on known medical indexes such as the Mayo Endoscopic Subscore(MES) & UCEIS(Ulcerative Colitis Endoscopic Index of Severity),usually consists of a few seconds of video & that video is played back continuously in a loop until the reader selects the appropriate score for that section.When the reader saves the section,the software immediately moves the video cursor to the highest scored section of the video.That way the central reader can review only the relevant portions of the video to confirm the score assigned to each segment.If the central reader’s scores do not align well with the AI scores then the software continues to show more sections of the video to the user,including sections it may have labeled as unscorable,that may be scorable. Results The review of the system by 3 key opinion leaders,user experience was positive.Not only does the system allow the reader’s attention to be more efficiently used,but the interface allows both AI & central reader scores to be saved,allowing for the latter to be used iteratively to re-train & improve the underlying AI model(s).Our tool was also used by a gastroenterologist specialist in order to perform video quality assessment & colon sections scoring. Conclusion We developed an AI tool that can be used to improve the efficiency & accuracy of the central reading process in clinical trials for UC.Further work is ongoing to improve the interface.

Download Full-text

Systemic Lupus Erythematosus Disease Activity Index 2000 Responder Index-50 Enhances the Ability of SLE Responder Index to Identify Responders in Clinical Trials

The Journal of Rheumatology ◽

10.3899/jrheum.110550 ◽

2011 ◽

Vol 38 (11) ◽

pp. 2395-2399 ◽

Cited By ~ 31

Author(s):

ZAHI TOUMA ◽

DAFNA D. GLADMAN ◽

DOMINIQUE IBAÑEZ ◽

SHAHRZAD TAGHAVI-ZADEH ◽

MURRAY B. UROWITZ

Keyword(s):

Systemic Lupus Erythematosus ◽

Clinical Trials ◽

Disease Activity ◽

Lupus Erythematosus ◽

Activity Index ◽

Disease Activity Index ◽

Systemic Lupus ◽

Original Definition ◽

Definition Of

Objective.To evaluate the performance of the Systemic Lupus Erythematosus (SLE) Responder Index (SRI) when the SLE Disease Activity Index 2000 (SLEDAI-2K) is substituted with SLEDAI-2K Responder Index-50 (SRI-50), a valid and reliable index of disease activity improvement. Also, to determine whether the SRI-50 will enhance the ability of SRI in detecting responders.Methods.Our study was conducted on patients who attended the Lupus Clinic from September 2009 to September 2010. SLEDAI-2K, SRI-50, the British Isles Lupus Assessment Group measure, and the Physician’s Global Assessment were determined initially and at followup. SRI was determined at the followup visit according to its original definition using the SLEDAI-2K score and by substituting SLEDAI-2K with SRI-50.Results.A total of 117 patients with SLEDAI-2K ≥ 4 at baseline were studied. Patients had 1 followup visit over a 3-month period. Twenty-nine percent of patients met the original definition of SRI and 35% of patients met the definition of SRI when SLEDAI-2K was substituted with SRI-50. The use of SRI-50 allowed determination of significant improvement in 7 additional patients. This improvement could not be discerned with the use of SLEDAI-2K as a component of SRI. At followup visits that showed improvement, SRI-50 scores decreased to a greater extent than SLEDAI-2K scores (p < 0.0001).Conclusion.SRI-50 enhances the ability of SRI to identify patients with clinically important improvement in disease activity. SRI-50 was superior to SLEDAI-2K in detecting partial clinical improvement, ≥ 50%, between visits. These properties of the SRI-50 enable it to be used as an independent outcome measure of improvement or as a component of SRI in clinical trials.

Download Full-text

P233 Evaluating the association between faecal calprotectin and endoscopic outcomes in ulcerative colitis using the HICKORY open-label induction cohort

Journal of Crohn s and Colitis ◽

10.1093/ecco-jcc/jjz203.362 ◽

2020 ◽

Vol 14 (Supplement_1) ◽

pp. S259-S259

Author(s):

W Reinisch ◽

B El Azzouzi ◽

R Li ◽

S Lacey ◽

M Daperno ◽

...

Keyword(s):

Ulcerative Colitis ◽

Clinical Trials ◽

Disease Activity ◽

Clinical Remission ◽

Area Under The Curve ◽

Percentage Change ◽

Open Label ◽

Faecal Calprotectin ◽

Using Data ◽

Factor Α

Abstract Background In clinical practice, faecal calprotectin (FC) is used to monitor disease activity in ulcerative colitis (UC); however, there is no consensus on optimal cut-off values of FC for predicting endoscopic outcomes. FC performance has not been extensively assessed in the context of clinical trials using central endoscopy. This study aimed to evaluate the association between FC and endoscopic disease activity and to propose a meaningful cut-off FC value to predict endoscopic outcomes using data from the open-label induction (OLI) cohort of HICKORY (NCT02100696). Methods HICKORY is a Phase 3 study evaluating etrolizumab in anti-tumour necrosis factor α-experienced patients with moderate-to-severe UC. The study included patients who received ≥1 dose of etrolizumab 105 mg subcutaneously every 4 weeks during a 14-week induction period. Percentage change in FC was calculated at week 14. The endoscopic activity was measured by Mayo Clinic score (MCS) endoscopic subscore (ES) using a robust central-reading model. Endoscopic improvement was defined as ES=0/1; clinical remission as MCS ≤2 and no individual subscore >1. FC analysis was performed by Covance® (Bühlman FC ELISA assay). Receiver operator characteristic (ROC) curve analyses were used to calculate cut-off FC values. Results A total of 97 patients (mean age [standard deviation], 41.2 ± 13.4 years) were included in the analysis. Median (interquartile range [IQR]) baseline duration of disease was 6.3 (3.2–12.3) years with a median (IQR) MCS of 9 (8–10). Median (IQR) baseline FC and ES were 254 (156–455) µg/g and 3 (3-3). At week 14, median (IQR) FC percentage change was −13 (−57 to 112). A numerical association between changes in FC level and ES was observed (Table). A cut-off FC value of 159 µg/g was observed to predict endoscopic improvement with >70% sensitivity and specificity; ROC area under the curve was 0.78 (Figure). Similar results were observed for clinical remission. Conclusion In this exploratory analysis using HICKORY OLI cohort data, changes in FC appear to associate with changes in ES. A cut-off FC value of 159 µg/g predicted endoscopic improvement. In UC, FC may be a useful non-invasive biomarker for ascertaining endoscopic disease activity in clinical trials; however, further clinical studies validating FC cut-offs against centrally read endoscopy are needed.

Download Full-text

Automatic, computer-aided determination of endoscopic and histological inflammation in patients with mild to moderate ulcerative colitis based on red density

Gut ◽

10.1136/gutjnl-2019-320056 ◽

2020 ◽

Vol 69 (10) ◽

pp. 1778-1786 ◽

Cited By ~ 10

Author(s):

Peter Bossuyt ◽

Hiroshi Nakase ◽

Séverine Vermeire ◽

Gert de Hertogh ◽

Tom Eelbode ◽

...

Keyword(s):

Ulcerative Colitis ◽

Disease Activity ◽

Objective Evaluation ◽

Second Phase ◽

Endoscopic Images ◽

Severity Scores ◽

Computer Based ◽

Index Of Severity ◽

Using Data ◽

Histological Factors

BackgroundThe objective evaluation of endoscopic disease activity is key in ulcerative colitis (UC). A composite of endoscopic and histological factors is the goal in UC treatment. We aimed to develop an operator-independent computer-based tool to determine UC activity based on endoscopic images.MethodsFirst, we built a computer algorithm using data from 29 consecutive patients with UC and 6 healthy controls (construction cohort). The algorithm (red density: RD) was based on the red channel of the red-green-blue pixel values and pattern recognition from endoscopic images. The algorithm was refined in sequential steps to optimise correlation with endoscopic and histological disease activity. In a second phase, the operating properties were tested in patients with UC flares requiring treatment escalation. To validate the algorithm, we tested the correlation between RD score and clinical, endoscopic and histological features in a validation cohort.ResultsWe constructed the algorithm based on the integration of pixel colour data from the redness colour map along with vascular pattern detection. These data were linked with Robarts histological index (RHI) in a multiple regression analysis. In the construction cohort, RD correlated with RHI (r=0.74, p<0.0001), Mayo endoscopic subscores (r=0.76, p<0.0001) and UC Endoscopic Index of Severity scores (r=0.74, p<0.0001). The RD sensitivity to change had a standardised effect size of 1.16. In the validation set, RD correlated with RHI (r=0.65, p=0.00002).ConclusionsRD provides an objective computer-based score that accurately assesses disease activity in UC. In a validation study, RD correlated with endoscopic and histological disease activity.

Download Full-text

P225 Using a standardised reporting proforma is associated with improved endoscopic assessment in UC

Journal of Crohn s and Colitis ◽

10.1093/ecco-jcc/jjz203.354 ◽

2020 ◽

Vol 14 (Supplement_1) ◽

pp. S255-S255

Author(s):

R KADER ◽

P Middleton ◽

O Ahmad ◽

R Dart ◽

J McGuire ◽

...

Keyword(s):

Ulcerative Colitis ◽

Disease Activity ◽

Site Factors ◽

Multi Centre Study ◽

Gi Endoscopy ◽

Mayo Score ◽

Chi Squared ◽

Index Of Severity ◽

Level Of Training ◽

Endoscopic Assessment

Abstract Background Repeated endoscopic assessments are an essential part of ulcerative colitis (UC) disease management and current guidelines recommend the use of an endoscopic activity score, either the endoscopic Mayo score or Ulcerative Colitis Endoscopic Index of Severity (UCEIS) as treatment targets. These indices have prognostic value, with endoscopic healing associated with favourable short- and long-term outcomes. This multi-centre study aimed to assess the frequency of using endoscopic disease activity scores in UC patients undergoing lower GI endoscopy. Methods Lower GI endoscopy reports from patients with UC were retrospectively reviewed from 7 sites in London between April and October 2018. Endoscopy reports were assessed based on the BRIDGe endoscopic reporting criteria including the use of Mayo or UCEIS score. The comparison was made between site factors (specialist IBD centres/non-specialist centres, use of reporting proforma), endoscopist speciality (gastroenterology, surgery or nurse endoscopist), level of training (consultant, registrar or nurse endoscopist) and interest in IBD. Chi-squared was used to compare groups. Results 899 lower GI endoscopy reports were reviewed. Mayo or UCEIS was used in 51% of cases (453/899). The use of endoscopic scores were significantly higher in gastroenterologists than in surgeons and nurse endoscopists respectively (401/762 (53%) vs. 22/54 (41%) vs. 30/83 (36%)), and higher in registrar trainees than consultants and nurse endoscopists (175/251 (70%) vs. 248/565 (44%) vs. 30/83 (36%)) and in those with a specialist interest in IBD compared with those without (237/409 (58%) vs. 216/490 (44%), p < 0.0001). The use of endoscopic scores was more frequent in specialist IBD centres than in non-specialist centres (417/728 (58%) vs. 36/172 (21%), p < 0.001). One centre used a reporting proforma which was associated with a significantly higher frequency of score use compared with centres without a proforma (202/260 (78%) vs. 251/639 (39%), p < 0.0001). Conclusion Reporting of endoscopic disease activity using a standardised scoring system occurs in only half of cases from this large multi-centre cohort. Frequency of use is higher in specialist IBD centres and when performed by gastroenterology specialists. Endoscopy reports from a site that used a standardised reporting proforma were significantly more likely to include an endoscopic index as well as a range of other reporting items. This suggests, at least in part, that endoscopy reporting may be optimised by the introduction of a proforma. Integration of a standardised proforma into reporting software would target all endoscopists performing UC endoscopies regardless of speciality, site or IBD interest.

Download Full-text

Direct Determination of Colonic Nitric Oxide Level--A Sensitive Marker of Disease Activity in Ulcerative Colitis

The American Journal of Gastroenterology ◽

10.1016/s0002-9270(98)00009-4 ◽

1998 ◽

Vol 93 (3) ◽

pp. 409-412 ◽

Cited By ~ 1

Author(s):

D Rachmilewitz

Keyword(s):

Nitric Oxide ◽

Ulcerative Colitis ◽

Disease Activity ◽

Direct Determination ◽

Nitric Oxide Level ◽

Sensitive Marker

Download Full-text

P253 The prevalence of type 6 herpes with polyps in patients with inflammatory bowel disease

Journal of Crohn s and Colitis ◽

10.1093/ecco-jcc/jjz203.382 ◽

2020 ◽

Vol 14 (Supplement_1) ◽

pp. S276-S276

Author(s):

G Babayeva-Sadigova ◽

Z Babayev ◽

U Mahmudov ◽

G Asadova ◽

F Quliyev ◽

...

Keyword(s):

Ulcerative Colitis ◽

Inflammatory Bowel Disease ◽

Crohn’S Disease ◽

Disease Activity ◽

Pyruvate Kinase ◽

Bowel Disease ◽

Herpes Viruses ◽

Intestinal Polyps ◽

Inflammatory Bowel

Abstract Background The prevalence of opportunistic infections in individuals with inflammatory bowel disease (IBD) remains relevant. Viral infections are a common cause of systemic inflammation of the digestive tract. Clinically expressed herpes infections are a serious problem, especially for individuals with prolonged immunosuppression. The aim of this study was t o assess the frequency of detection of herpes-6 type with polyps in patients with IBD. Methods 189 patients with IBD, 102 with ulcerative colitis and 87 with Crohn’s disease were examined. Of the number of patients: 98women, 91man. The age of the patients is from 16 to 63 years (mean age 41.4 ± 4.8), out of 189 patients intestinal polyps were identified in 61 patients with a colonoscopy. In addition to standard clinical and endoscopic examinations, disease activity was assessed by indicators of highly sensitive C-reactive protein, homocysteine, vitamin D in blood serum, albumin in urine, calprotectin, lactoferrin and pyruvate kinase in faeces. All patients underwent a serological blood test by ELISA for specific antibodies to herpesvirus type 1–6 IgG/IgM antibodies, as well as avidity, determination of DNA to herpes viruses of types 1–6 species by PCR in blood and mucosal biopsy. Results Among patients with IBD (n = 189), active herpes viruses in the blood were found in 65 patients (34.4%); of them, 39 patients (20.6% (n = 189) and 60%(n = 65),respectively) were diagnosed with monoherpesvirus infections, and 26 (13.7%(n = 189) and 40%(n = 65), respectively) of myxedvirus infections. PCR of the biopsy material was positive in isolation for type 6 herpes virus in 22 cases (11.6% (n = 189) and 33.8% (n = 65)) and in 34 cases (17.9% (n = 189) and 52.3% (n = 65)) a myxedvirus lesion was detected in the tissue (HV6 + EPB, HV6 + CMV), these indicators were detected in patients with intestinal polyps. Although clinical and endoscopic and laboratory indicators of disease activity were significantly higher in patients with mixed tissue lesions (p˂0.05), the level of faecal pyruvate kinase was higher in patients with polyps, indicating a negative effect of herpes type 6 on the course of inflammatory bowel diseases. At the same time, no significant differences were found between patients with ulcerative colitis and Crohn’s disease. Conclusion The frequency of occurrence of herpes type 6 on PCR biopsy in patients with polyps with IBD is more than 91.8% (36.5% in isolation and 55.7% of mixed lesions) and is accompanied by higher rates of faecal pyruvate kinase, which is already an indicator of emerging malignancy. The authors propose the determination of polyps in patients with IBD HV6 by PCR biopsy of intestinal tissue.

Download Full-text