Comprehensive review of publicly available colonoscopic imaging databases for artificial intelligence research: availability, accessibility, and usability.

BACKGROUND AND AIMS: Publicly available databases containing colonoscopic imaging data are valuable resources for artificial intelligence (AI) research. Currently, little is known regarding the available number and content of these databases. This review aimed to describe the availability, accessibility, and usability of publicly available colonoscopic imaging databases, focusing on polyp detection, polyp characterization, and quality of colonoscopy. METHODS: A systematic literature search was performed in MEDLINE and Embase to identify AI studies describing publicly available colonoscopic imaging databases published after 2010. Second, a targeted search using Google's Dataset Search, Google Search, GitHub, and Figshare was done to identify databases directly. Databases were included if they contained data about polyp detection, polyp characterization, or quality of colonoscopy. To assess accessibility of databases, the following categories were defined: open access, open access with barriers, and regulated access. To assess the potential usability of the included databases, essential details of each database were extracted using a checklist derived from the Checklist for Artificial Intelligence in Medical Imaging. RESULTS: We identified 22 databases with open access, 3 databases with open access with barriers, and 15 databases with regulated access. The 22 open access databases contained 19,463 images and 952 videos. Nineteen of these databases focused on polyp detection, localization, and/or segmentation; 6 on polyp characterization, and 3 on quality of colonoscopy. Only half of these databases have been used by other researcher to develop, train, or benchmark their AI system. Although technical details were in general well reported, important details such as polyp and patient demographics and the annotation process were under-reported in almost all databases. CONCLUSIONS: This review provides greater insight on public availability of colonoscopic imaging databases for AI research. Incomplete reporting of important details limits the ability of researchers to assess the usability of current databases.

High quality colonoscopy: using textbook process as a composite quality measure.

BACKGROUND : High quality colonoscopy is fundamental to good patient outcomes. "Textbook outcome" has proven to be a feasible multidimensional measure for quality assurance between surgical centers. In this study, we sought to establish the "textbook process" (TP) as a new composite measure for the optimal colonoscopy process and assessed how frequently TP was attained in clinical practice and the variation in TP between endoscopists. METHODS : To reach consensus on the definition of TP, international expert endoscopists completed a modified Delphi consensus process. The achievement of TP was then applied to clinical practice. Prospectively collected data in two endoscopy services were retrospectively evaluated. Data on colonoscopies performed for symptoms or surveillance between 1 January 2018 and 1 August 2021 were analyzed. RESULTS : The Delphi consensus process was completed by 20 of 27 invited experts (74.1 %). TP was defined as a colonoscopy fulfilling the following items: explicit colonoscopy indication; successful cecal intubation; adequate bowel preparation; adequate withdrawal time; acceptable patient comfort score; provision of post-polypectomy surveillance recommendations in line with guidelines; and the absence of the use of reversal agents, early adverse events, readmission, and mortality. In the two endoscopy services studied, TP was achieved in 5962/8227 colonoscopies (72.5 %). Of 48 endoscopists performing colonoscopy, attainment of TP varied significantly, ranging per endoscopist from 41.0 % to 89.1 %. CONCLUSION : This study proposes a new composite measure for colonoscopy, namely "textbook process." TP gives a comprehensive summary of performance and demonstrates significant variation between endoscopists, illustrating the potential benefit of TP as a measure in future quality assessment programs.

Novel classification for adverse events in GI endoscopy: the AGREE classification.

BACKGROUND AND AIMS: Standardized registration and evaluation of adverse events (AEs) are essential to assess the safety of endoscopic procedures. We propose a novel classification system, named adverse events in GI endoscopy (AGREE), adapted from a widely accepted surgical tool. METHODS: The Clavien-Dindo classification for surgical AEs was adapted for endoscopy. To validate the novel classification, we assessed if the severity of AEs, as perceived by 10 endoscopists, 10 endoscopy nurses, and 10 patients, corresponded with the severity grading used in the AGREE classification in 10 pairwise comparisons. We additionally assessed the correlation between the AGREE classification and the American Society for Gastrointestinal Endoscopy (ASGE) classification. The acceptability of the AGREE classification was evaluated through an international questionnaire. RESULTS: The perception of endoscopists, endoscopy nurses, and patients corresponded with the severity grading of the AGREE classification in 80% of cases (238/299). The AGREE classification significantly correlated with the ASGE classification (ρ = .760). Fifty-seven of 84 experts (68%) completed a questionnaire regarding the acceptability of the AGREE classification. The experts consulted considered the AGREE classification as simple (86%), reproducible (98%), logical (98%), and useful (96%). Most case presentations (84%) were correctly graded according to the AGREE classification. CONCLUSIONS: The AGREE classification provides a standardized and reproducible approach to the assessment of AEs in diagnostic and therapeutic GI endoscopy. Broad implementation of the AGREE classification may facilitate the evaluation of AEs across different endoscopists, disciplines, endoscopy services, and regions. This standardization of AE reporting will support improved quality assurance in GI endoscopy.

Case-mix adjustment to compare colonoscopy performance between endoscopy centers: a nationwide registry study.

BACKGROUND : Nonmodifiable patient and endoscopy characteristics might influence colonoscopy performance. Differences in these so-called case-mix factors are likely to exist between endoscopy centers. This study aimed to examine the importance of case-mix adjustment when comparing performance between endoscopy centers. METHODS : Prospectively collected data recorded in the Dutch national colonoscopy registry between 2016 and 2019 were retrospectively analyzed. Cecal intubation rate (CIR) and adequate bowel preparation rate (ABPR) were analyzed. Additionally, polyp detection rate (PDR) was studied in screening colonoscopies following a positive fecal immunochemical test (FIT). Variation in case-mix factors between endoscopy centers and expected outcomes for each performance measure were calculated per endoscopy center based on case-mix factors (sex, age, American Society of Anesthesiologist [ASA] score, indication) using multivariable logistic regression. RESULTS: 363 840 colonoscopies were included from 51 endoscopy centers. Mean percentages per endoscopy center were significantly different for age > 65 years, male patients, ASA ≥ III, and diagnostic colonoscopies (all P < 0.001). In the FIT-positive screening population, significant differences were observed between endoscopy centers for age > 65 years, male patients, and ASA ≥ III (all P ≤ 0.001). The expected CIR, ABPR, and PDR ranged from 95.0 % to 96.9 %, from 93.6 % to 96.4 %, and from 76.2 % to 79.1 %, respectively. Age, sex, ASA classification, and indication were significant case-mix factors for CIR and ABPR. In the FIT-positive screening population, age, sex, and ASA classification were significant case-mix factors for PDR. CONCLUSION: Our findings emphasize the importance of considering case-mix adjustment when comparing colonoscopy performance measures between endoscopy centers.

Continuous monitoring of colonoscopy performance in the Netherlands: first results of a nationwide registry.

BACKGROUND: To optimize colonoscopy quality, several performance measures have been developed. These are usually assessed without distinction between the indications for colonoscopy. This study aimed to assess the feasibility of linking two national registries (one for colonoscopy and one for adverse events of gastrointestinal endoscopies in the Netherlands), and to describe the results of colonoscopy quality per indication. METHODS: This retrospective study was conducted with prospectively collected data of the Dutch Gastrointestinal Endoscopy Audit (DGEA) and the Dutch Registration of Complications in Endoscopy (DRCE). Data between 01-01-2016 and 01-01-2019 were analyzed. To calculate adverse event rates, data were linked at the level of endoscopy service. RESULTS: During the 3-year study period, 266 981 colonoscopies were recorded in DGEA. Of all indications, cecal intubation rate was highest in fecal immunochemical test (FIT)-positive screening colonoscopies (97.1 %), followed by surveillance (93.2 %), diagnostic (90.7 %), and therapeutic colonoscopies (83.1 %). The highest rate of adequate bowel preparation was observed in FIT-positive screening colonoscopies (97.1 %). A total of 1540 colonoscopy-related adverse events occurred (0.58 % of all colonoscopies). Bleeding and perforation and rates were highest for therapeutic (1.56 % and 0.51 %, respectively) and FIT-positive screening (0.72 % and 0.06 %, respectively) colonoscopies. The colonoscopy-related mortality was 0.006 %. CONCLUSION: This study describes the first results of the Dutch national colonoscopy registry, which was successfully linked to data from the national registry for adverse events of gastrointestinal endoscopies. In this large dataset, performance varied between indications. Our results emphasize the importance of defining benchmarks per indication in future guidelines.

Performance measures for colonoscopy in inflammatory bowel disease patients: European Society of Gastrointestinal Endoscopy (ESGE) Quality Improvement Initiative.

The European Society of Gastrointestinal Endoscopy (ESGE) presents a short list of performance measures for colonoscopy in inflammatory bowel disease (IBD) patients. Current performance measures for colonoscopy mainly focus on detecting (pre)malignant lesions. However, these performance measures are not relevant for all colonoscopy indications in IBD patients. Therefore, our aim was to provide endoscopy services across Europe and other interested countries with a tool for quality monitoring and improvement in IBD colonoscopy. Eight key performance measures and one minor performance measure were recommended for measurement and evaluation in daily endoscopy practice.

Impact of sedation on the Performance Indicator of Colonic Intubation.

BACKGROUND: The Performance Indicator of Colonic Intubation (PICI) is a new measure of high-quality colonic intubation. Adequate PICI was defined as cecal intubation without significant discomfort and use of minimal sedation. This study assessed achievement of PICI within the Dutch colorectal cancer (CRC) screening program, and determined the association between PICI and adenoma detection rate (ADR). PICI achievement when using the Dutch median midazolam dose was also assessed. METHODS: This retrospective study was conducted within the Dutch fecal immunochemical test-based CRC screening program. Colonoscopy and pathology data were prospectively collected in a national database. Data between January 2016 through January 2018 were analyzed. Adequate PICI was defined as successful cecal intubation, Gloucester Comfort Scale (GCS) of 1 - 3, and use of ≤ 2.5 mg midazolam. RESULTS: 107 328 colonoscopies were performed during the study period. Adequate PICI was achieved in 49 500 colonoscopies (46.1 %). In colonoscopies with inadequate PICI, inadequacy was due to higher sedation doses in 87.8 %. Adequate PICI was associated with higher ADR (odds ratio 1.16, 95 % confidence interval 1.12 - 1.20). When using a cutoff of 5 mg midazolam, median dose in this Dutch population, adequate PICI was achieved in 95 410 colonoscopies (88.9  %). CONCLUSION: PICI appeared to be heavily dependent on sedation practice. Because of wide variation in sedation practice between individual endoscopists and countries, the benefit of PICI as a quality indicator is limited.

Plasma lecithin:cholesterol acyltransferase and phospholipid transfer protein activity independently associate with nonalcoholic fatty liver disease.

BACKGROUND: Nonalcoholic fatty liver disease (NAFLD) is a highly prevalent condition which contributes to atherogenic apolipoprotein B dyslipoproteinemias. Lecithin:cholesterol acyltransferase (LCAT) and phospholipid transfer protein (PLTP) are both synthesized by the liver and are important in lipid metabolism. Here, we interrogated the impact of NAFLD on plasma LCAT and PLTP activities. METHODS: Plasma LCAT activity (exogenous substrate assay) and PLTP activity (phospholipid vesicles-HDL assay) were determined in 348 subjects (279 men; 81 subjects with type 2 diabetes (T2DM); 123 with metabolic syndrome (MetS)). A Fatty Liver Index (FLI) ≥60 was used as a proxy of NAFLD. Insulin resistance was determined by homoeostasis model assessment (HOMA-IR). RESULTS: A total of 147 participants had an FLI ≥60 coinciding with T2DM and MetS (P < 0.001 for each). Plasma LCAT activity and PLTP activity were on average 12% and 5% higher, respectively, in subjects with an FLI ≥ 60 (P < 0.001 for each). In age- and sex-adjusted partial linear regression analysis, LCAT activity and PLTP activity were positively related to various obesity measures and HOMA-IR (P < 0.001 for each). In multivariable linear regression analyses adjusted for age and sex, LCAT activity was associated with an FLI ≥ 60 independent of T2DM and MetS, the waist/hip ratio, or HOMA-IR (ß = 0.307 to 0.366, P < 0001 for all models). PLTP activity was also associated with an FLI ≥ 60 independent of these variables (ß = 0.151 to 0223, P = 0.013 to 0.001). CONCLUSION: NAFLD, as inferred from an FLI≥60, confers higher plasma LCAT and to a lesser extent PLTP activity, even when taking account of T2DM, MetS, central obesity and insulin resistance.

[How important is the quality of colonoscopy, considering the population screening for colorectal cancer?] / Hoe belangrijk is de kwaliteit van coloscopie?

Colonoscopy is the reference standard for the detection of polyps and colorectal cancer (CRC). If during colonoscopy, all colorectal lesions are detected and completely removed, this individual will be long-term protected from CRC. The quality of the colonoscopy procedure is essential for an optimal protective effect. Recently published data of negative colonoscopies within the Polish Colonoscopy Screening Program, with a maximum follow-up of 17.4 years, demonstrated that high-quality colonoscopy was associated with a lower CRC incidence and mortality compared to low-quality colonoscopy. Colonoscopy quality was defined by completeness of colonoscopy, quality of the bowel preparation and number of detected colorectal lesions. These results suggest that the interval after a negative colonoscopy for the next screening might be safely prolonged, preventing unnecessary costs and risks for the patient. The quality of the initial colonoscopy is essential and a high quality will be fundamental for surveillance guidelines in the near future.

High prevalence of apolipoprotein B dyslipoproteinemias in non-alcoholic fatty liver disease: The lifelines cohort study.

OBJECTIVE: Cardiovascular disease (CVD) is a major adverse consequence of non-alcoholic fatty liver disease (NAFLD). The association of NAFLD with various apolipoprotein B (apoB) dyslipoproteinemias is unclear. We determined the prevalence of specific apoB dyslipoproteinemias in subjects with suspected NAFLD. METHODS: This study was conducted among 22,865 fasting adults living in the northern part of the Netherlands (Lifelines Cohort Study). Six apoB dyslipoproteinemias were defined using an algorithm derived from apoB, total cholesterol and triglycerides. NAFLD was defined as Fatty Liver Index (FLI) ≥60. Advanced hepatic fibrosis was defined as NAFLD fibrosis score (NFS) ≥0.676. RESULTS: 4790 participants (20.9%) had an FLI≥60. NAFLD subjects were older, more likely to be men, more obese and more often had diabetes and metabolic syndrome (P<0.001 for each). Among NAFLD subjects, any apoB dyslipoproteinemia was present in 61.5% vs. 16.5% in subjects without NAFLD (P<0.001). Elevated chylomicrons were not observed in NAFLD. In univariate analysis, NAFLD was associated with a higher prevalence of each apoB dyslipoproteinemia vs. subjects with an FLI<60 (P<0.001), except for low density lipoprotein (LDL) dyslipoproteinemia. Additionally, each apoB dyslipoproteinemia was independently associated with NAFLD in age- and sex-adjusted logistic regression analysis, including the apoB dyslipoproteinemias together (P<0.001). The prevalence of apoB dyslipoproteinemias was not altered in subjects with NFS ≥0.676. CONCLUSIONS: NAFLD rather than advanced hepatic fibrosis is independently associated with increased prevalence of chylomicrons+very low-density lipoproteins (VLDL) remnants, VLDL, LDL and VLDL+LDL dyslipoproteinemias. ApoB dyslipoproteinemias may contribute to increased CVD risk associated with NAFLD.