Artificial intelligence-assisted optical diagnosis for the resect-and-discard strategy in clinical practice: the Artificial intelligence BLI Characterization (ABC) study.

BACKGROUND: Optical diagnosis of colonic polyps is poorly reproducible outside of high volume referral centers. The present study aimed to assess whether real-time artificial intelligence (AI)-assisted optical diagnosis is accurate enough to implement the leave-in-situ strategy for diminutive (≤ 5 mm) rectosigmoid polyps (DRSPs). METHODS: Consecutive colonoscopy outpatients with ≥ 1 DRSP were included. DRSPs were categorized as adenomas or nonadenomas by the endoscopists, who had differing expertise in optical diagnosis, with the assistance of a real-time AI system (CAD-EYE). The primary end point was ≥ 90 % negative predictive value (NPV) for adenomatous histology in high confidence AI-assisted optical diagnosis of DRSPs (Preservation and Incorporation of Valuable endoscopic Innovations [PIVI-1] threshold), with histopathology as the reference standard. The agreement between optical- and histology-based post-polypectomy surveillance intervals (≥ 90 %; PIVI-2 threshold) was also calculated according to European Society of Gastrointestinal Endoscopy (ESGE) and United States Multi-Society Task Force (USMSTF) guidelines. RESULTS: Overall 596 DRSPs were retrieved for histology in 389 patients; an AI-assisted high confidence optical diagnosis was made in 92.3 %. The NPV of AI-assisted optical diagnosis for DRSPs (PIVI-1) was 91.0 % (95 %CI 87.1 %-93.9 %). The PIVI-2 threshold was met with 97.4 % (95 %CI 95.7 %-98.9 %) and 92.6 % (95 %CI 90.0 %-95.2 %) of patients according to ESGE and USMSTF, respectively. AI-assisted optical diagnosis accuracy was significantly lower for nonexperts (82.3 %, 95 %CI 76.4 %-87.3 %) than for experts (91.9 %, 95 %CI 88.5 %-94.5 %); however, nonexperts quickly approached the performance levels of experts over time. CONCLUSION: AI-assisted optical diagnosis matches the required PIVI thresholds. This does not however offset the need for endoscopists' high level confidence and expertise. The AI system seems to be useful, especially for nonexperts.

Linked color imaging reduces the miss rate of neoplastic lesions in the right colon: a randomized tandem colonoscopy study.

BACKGROUND: Linked color imaging (LCI) is a newly developed image-enhancing endoscopy technology that provides bright endoscopic images and increases color contrast. We investigated whether LCI improves the detection of neoplastic lesions in the right colon when compared with high definition white-light imaging (WLI). METHODS: Consecutive patients undergoing colonoscopy were randomized (1:1) after cecal intubation into right colon inspection at first pass by LCI or by WLI. At the hepatic flexure, the scope was reintroduced to the cecum under LCI and a second right colon inspection was performed under WLI in previously LCI-scoped patients (LCI-WLI group) and vice versa (WLI-LCI group). Lesions detected on first- and second-pass examinations were used to calculate detection and miss rates, respectively. The primary outcome was the right colon adenoma miss rate. RESULTS: Of the 600 patients enrolled, 142 had at least one adenoma in the right colon, with similar right colon adenoma detection rates (r-ADR) in the two groups (22.7 % in LCI-WLI and 24.7 % in WLI-LCI). At per-polyp analysis, double inspection of the right colon in the LCI-WLI and WLI-LCI groups resulted in an 11.8 % and 30.6 % adenoma miss rate, respectively (P < 0.001). No significant difference in miss rate was found for advanced adenomas or sessile serrated lesions. At per-patient analysis, at least one adenoma was identified in the second pass only (incremental ADR) in 2 of 300 patients (0.7 %) in the LCI - WLI group and in 13 of 300 patients (4.3 %) in the WLI - LCI group (P = 0.01). CONCLUSIONS: LCI could reduce the miss rate of neoplastic lesions in the right colon.

Accuracy of capsule colonoscopy and computed tomographic colonography in individuals with positive results from the fecal occult blood test.

BACKGROUND & AIMS: Computed tomographic colonography (CTC) is a reliable option for screening subjects who are unable or unwilling to undergo optical colonoscopy (OC). A colon capsule (PillCam Colon2 [CC2]; GivenImaging Ltd., Yokneam, Israel) has shown promising results in detecting polyps larger than 6 mm. We compared the accuracy of CC2 and CTC in identifying individuals with at least 1 polyp greater than 6 mm and subjects' attitude toward the procedures. METHODS: Fifty individuals (mean age, 59.2 ± 5.8 y; 58% male) with positive results from the immunochemical fecal occult blood test (iFOBT-positive) underwent CC2, CTC, and OC. The unblinded colonoscopy, integrating OC, CTC, and CC2 results, was used as the reference standard. In a per-patient analysis, the accuracy of CC2 and CTC were assessed for individuals with at least 1 polyp 6 mm or larger. Individuals were asked to choose which procedure they would be willing to repeat between CTC and CC2. RESULTS: The combination of OC, CTC, and CC2 identified 16 cases with at least 1 polyp 6 mm or larger (reference standard). CTC identified the polyps with 88.2% sensitivity, 84.8% specificity, a 3.0 positive likelihood ratio, and a 0.07 negative likelihood ratio. CC2 identified the polyps with 88.2% sensitivity, 87.8% specificity, a 3.75 positive likelihood ratio, and a 0.06 negative likelihood ratio. Thirty-nine subjects (78%) said they preferred CC2 to CTC. CONCLUSIONS: CC2 and CTC detect polyps 6 mm and larger with high levels of accuracy; these techniques are effective in selecting iFOBT-positive individuals who do not need to be referred for colonoscopy. CC2 seems to be better tolerated than CTC, and could be a reliable alternative to CTC for iFOBT-positive individuals who are unable or unwilling to undergo OC. ClinicalTrials.gov number: NCT01744509.

Utility of 3-dimensional image reconstruction in the diagnosis of small-bowel masses in capsule endoscopy (with video).

BACKGROUND: In small-bowel capsule endoscopy (SBCE), differentiating masses (ie, lesions of higher probability for neoplasia) requiring more aggressive intervention from bulges (essentially, false-positive findings) is a challenging task; recently, software that enables 3-dimensional (3D) reconstruction has become available. OBJECTIVE: To evaluate whether "coupling" 3D reconstructed video clips with the standard 2-dimensional (s2D) counterparts helps in distinguishing masses from bulges. DESIGN: Three expert and 3 novice SBCE readers, blind to others and in a random order, reviewed the s2D video clips and subsequently the s2D clips coupled with their 3D reconstruction (2D+3D). SETTING: Multicenter study in 3 community hospitals in Italy and a university hospital in Scotland. PATIENTS: Thirty-two deidentified 5-minute video clips, containing mucosal bulging (19) or masses (13). INTERVENTION: 3D reconstruction of s2D SBCE video clips. MAIN OUTCOME MEASURE: Differentiation of masses from bulges with s2D and 2D+3D video clips, estimated by the area under the receiver operating characteristic curve (AUC); interobserver agreement. RESULTS: AUC for experts and novices for s2D video clips was .74 and .5, respectively (P = .0053). AUC for experts and novices with 2D+3D was .70 (compared with s2D: P = .245) and .57 (compared s2D: P = .049), respectively. AUC for experts and novices with 2D+3D was similar (P = .1846). The interobserver agreement was good for both experts and novices with the s2D (k = .71 and .54, respectively) and the 2D+3D video clips (k = .58 in both groups). LIMITATIONS: Few, short video clips; fixed angle of 3D reconstruction. CONCLUSIONS: The adjunction of a 3D reconstruction to the s2D video reading platform does not improve the performance of expert SBCE readers, although it significantly increases the performance of novices in distinguishing masses from bulging.

White light computer-aided optical diagnosis of diminutive colorectal polyps in routine clinical practice.

Background and study aims Artificial Intelligence (AI) systems could make the optical diagnosis (OD) of diminutive colorectal polyps (DCPs) more reliable and objective. This study was aimed at prospectively evaluating feasibility and diagnostic performance of AI-standalone and AI-assisted OD of DCPs in a real-life setting by using a white light-based system (GI Genius, Medtronic Co, Minneapolis, Minnesota, United States). Patients and methods Consecutive colonoscopy outpatients with at least one DCP were evaluated by 11 endoscopists (5 experts and 6 non-experts in OD). DCPs were classified in real time by AI (AI-standalone OD) and by the endoscopist with the assistance of AI (AI-assisted OD), with histopathology as the reference standard. Results Of the 480 DCPs, AI provided the outcome "adenoma" or "non-adenoma" in 81.4% (95% confidence interval [CI]: 77.5-84.6). Sensitivity, specificity, positive and negative predictive value, and accuracy of AI-standalone OD were 97.0% (95% CI 94.0-98.6), 38.1% (95% CI 28.9-48.1), 80.1% (95% CI 75.2-84.2), 83.3% (95% CI 69.2-92.0), and 80.5% (95% CI 68.7-82.8%), respectively. Compared with AI-standalone, the specificity of AI-assisted OD was significantly higher (58.9%, 95% CI 49.7-67.5) and a trend toward an increase was observed for other diagnostic performance measures. Overall accuracy and negative predictive value of AI-assisted OD for experts and non-experts were 85.8% (95% CI 80.0-90.4) vs. 80.1% (95% CI 73.6-85.6) and 89.1% (95% CI 75.6-95.9) vs. 80.0% (95% CI 63.9-90.4), respectively. Conclusions Standalone AI is able to provide an OD of adenoma/non-adenoma in more than 80% of DCPs, with a high sensitivity but low specificity. The human-machine interaction improved diagnostic performance, especially when experts were involved.

Constipation in the elderly: management strategies.

Constipation is a highly prevalent and bothersome disorder that negatively affects patients' social and professional lives and places a great economic burden on both patients and national health services. An accurate determination of the prevalence of constipation is difficult because of the various definitions used, but many epidemiological studies have shown that it affects up to 20% of the population at any one time. Although constipation is not a physiological consequence of normal aging, decreased mobility and other co-morbid medical conditions may contribute to its prevalence in older adults. Functional constipation is diagnosed when no secondary causes can be identified. Patients have some unusual beliefs about their bowel habits. Systematic attention to history, examination and investigation, especially in older people, can be highly effective in resolving problems and in enhancing quality of life. There is a considerable range of treatment modalities available for patients with constipation, but the clinical evidence supporting their use varies widely. However, if constipation is not managed proactively, patients can experience negative consequences, such as anorexia, nausea, bowel impaction or bowel perforation. The clinical benefits of various traditional pharmacological and non-pharmacological agents remain unclear. The first steps in the treatment of simple constipation include increasing intake of dietary fibre and the use of a fibre supplement. Patients with severe constipation or those unable to comply with the recommended intake of fibre may benefit from the addition of laxatives. More recently, newer agents (e.g. tegaserod and lubiprostone), have been approved for the treatment of patients with chronic constipation. Additional work is needed to determine what role, if any, these agents may play in the treatment of patients with chronic constipation. The purpose of this review is to identify evidence-based interventions for the prevention and management of constipation in the elderly.

Educational strategies for colonoscopy bowel prep overcome barriers against split-dosing: A randomized controlled trial.

BACKGROUND: A split-dose (SD) regimen is crucial for colonoscopy quality. Compliance with SD for early morning colonoscopy is generally poor. The present study evaluated whether pre-colonoscopy counselling, in addition to a dedicated leaflet, might increase SD uptake. METHODS: Consecutive 50-69-year-old patients undergoing screening colonoscopy before 10 a.m. were randomized to either receive written information only on bowel preparation (Written Group, WG) or written and oral instructions (Written and Oral Group, WaOG). The leaflet strongly encouraged SD adoption. The primary endpoint was the number of patients adopting SD in each group. The secondary endpoints were predictors of SD uptake, compliance with preparation schemes and cleansing adequacy. RESULTS: A total of 286 patients (143 WG, 143 WaOG) were enrolled (mean age 59.6 ± 6.1 years, men 49.3%). SD was adopted by 114 and 125 patients in the WG and WaOG, respectively (79.7% versus 87.4%, p = 0.079). No significant differences were observed for the proportion of patients with full compliance with preparation scheme (97.9% versus 97.2%, p = 0.99) and of procedures with adequate bowel cleansing (95.6% versus 95.1%, p = 0.77). At multivariate analysis, a > 1 h travel time to the endoscopy service was inversely correlated with SD uptake (odds ratio (OR) 0.30, 95% confidence interval (CI) 0.09-0.98). CONCLUSIONS: Our leaflet guaranteed satisfactory uptake of SD and excellent adherence to the preparation scheme for early morning colonoscopy. Its use might marginalize the need for additional oral instructions, particularly in open-access settings.

Evaluation of 4 three-dimensional representation algorithms in capsule endoscopy images.

AIM: To evaluate the three-dimensional (3-D) representation performance of 4 publicly available Shape-from-Shading (SfS) algorithms in small-bowel capsule endoscopy (SBCE). METHODS: SfS techniques recover the shape of objects using the gradual variation of shading. There are 4 publicly available SfS algorithms. To the best of our knowledge, no comparative study with images obtained during clinical SBCE has been performed to date. Three experienced reviewers were asked to evaluate 54 two-dimensional (2-D) images (categories: protrusion/inflammation/vascular) transformed to 3-D by the aforementioned SfS 3-D algorithms. The best algorithm was selected and inter-rater agreement was calculated. RESULTS: Four publicly available SfS algorithms were compared. Tsai's SfS algorithm outperformed the rest (selected as best performing in 45/54 SBCE images), followed by Ciuti's algorithm (best performing in 7/54 images) and Torreão's (in 1/54 images). In 26/54 images; Tsai's algorithm was unanimously selected as the best performing 3-D representation SfS software. Tsai's 3-D algorithm superiority was independent of lesion category (protrusion/inflammatory/vascular; P = 0.678) and/or CE system used to obtain the 2-D images (MiroCam/PillCam; P = 0.558). Lastly, the inter-observer agreement was good (kappa = 0.55). CONCLUSION: 3-D representation software offers a plausible alternative for 3-D representation of conventional capsule endoscopy images (until optics technology matures enough to allow hardware enabled-"real" 3-D reconstruction of the gastrointestinal tract).

Hyoscine N-butylbromide for adenoma detection during colonoscopy: a randomized, double-blind, placebo-controlled study.

BACKGROUND: Hyoscine N-butylbromide (HBB), commonly used during colonoscopy to facilitate cecal intubation, has been proposed to increase the adenoma detection rate (ADR). AIMS: To evaluate whether HBB administration increases the adenoma detection rate and influences patients' tolerance. METHODS: Consecutive colonoscopy outpatients were randomized after cecal intubation to receive either 20mg HBB or placebo i.v. The number, size, histology and location of polyps were recorded. The air retained in the abdomen was either indirectly estimated by ΔAC (difference in the abdominal circumference measured before and after colonoscopy) or directly evaluated by patients' perception (visual analogic scale, range 0-100). RESULTS: 402 patients (44% male; mean age 57.7±12.5years) received either HBB or placebo. No differences in ADR (31.7% vs. 28%, p=0.48), advanced-ADR (7.4% vs. 10.5%, p=0.35) were observed between HBB and placebo group, respectively. A significantly lower detection rate of flat/depressed lesions was observed in the HBB group (0.5% vs. 5.5%, p=0.003). The ΔAC and the bloating perception were comparable between the two groups (p=0.22 and p=0.48, respectively). CONCLUSIONS: HBB administered before colonoscope withdrawal does not increase adenoma detection rate and seems to hamper the visualization of flat/depressed lesions. This finding raises concerns on the indiscriminate use of HBB during colonoscopy.

Can we improve the detection rate and interobserver agreement in capsule endoscopy?

BACKGROUND: Data about strategies for improving the diagnostic ability of capsule endoscopy readers are lacking. AIM: (1) To evaluate the detection rate and the interobserver agreement among readers with different experience; (2) to verify the impact of a specific training (hands-on training plus expert tutorial) on these parameters. METHODS: 17 readers reviewed 12 videos twice; between the two readings they underwent the training. The identified small bowel findings were described by a simplified version of Structured Terminology and classifies as clinically significant/non-significant. Findings identified by the readers were compared with those identified by three experts (Reference Standard). RESULTS: The Reference Standard identified 26 clinically significant findings. The mean detection rate of overall readers for significant findings was low (about 50%) and did not change after the training (46.2% and 46.4%, respectively). There was no difference in the detection rate among readers with different experience. The interobserver agreement with the Reference Standard in describing significant findings was moderate (k = 0.44; CI95%: 0.39-0.50) and did not change after the training (k = 0.44; CI95%: 0.38-0.49) or stratifying readers according to their experience. CONCLUSIONS: Both the interobserver agreement and the detection rate of significant findings are low, regardless of the readers' experience. Our training did not significantly increase the performance of readers with different experience.

Anticoagulant or aspirin treatment does not affect the positive predictive value of an immunological fecal occult blood test in patients undergoing colorectal cancer screening: results from a nested in a cohort case-control study.

BACKGROUND AND AIM: The immunochemical fecal occult blood test (i-FOBT) is widely used as a recommended screening strategy for colorectal cancer (CRC). A growing number of patients potentially targeted by CRC screening programs are on oral anticoagulant or chronic low-dose aspirin therapy, mainly for primary or secondary cardiovascular prophylaxis. This study aims at evaluating whether the use of these medications may impact on the diagnostic performances of i-FOBT for CRC screening. METHODS: All i-FOBT-positive patients on anticoagulant or chronic low-dose aspirin therapy recorded in a regional mass screening program database were enrolled as cases. Control groups were derived from the same database and included drug-naive i-FOBT-positive patients, matched in a ratio of 1 : 2 for age (± 3 years of age), sex, date of colonoscopy, and practice site. Information about the use of medications was collected by cross-checking patients' interview before colonoscopy and data recorded in the provincial electronic registry of medical prescriptions. The positive predictive value of i-FOBT for significant neoplasia (high-risk adenoma and CRC) was calculated in the case and control groups. RESULTS: In a 2-year study period, 2376 patients were recorded in the regional database. Of these patients, 53 (2%) were on anticoagulation (control group of 106 patients) and 172 (6.6%) were on chronic low-dose aspirin treatment (control group of 344 patients). Significant neoplasia was detected in 15 (28.3%) patients on anticoagulants and in 37 (34.9%) corresponding controls (P=0.45). Significant neoplasia was detected in 50 (29.1%) patients on chronic low-dose aspirin and in 107 (31.1%) corresponding controls (P=0.64). CONCLUSION: The positive predictive value of i-FOBT for significant neoplasia is not affected by ongoing anticoagulant or chronic low-dose aspirin therapy. This finding suggests that there is no need to interrupt these treatments before i-FOBT for CRC screening.

Small bowel capsule endoscopy in clinical practice: a multicenter 7-year survey.

BACKGROUND AND AIM: Data about small bowel capsule endoscopy (SBCE) come from studies involving small and highly selected populations. The study aim was to describe extent of use, indications, results, complications, and practical issues of SBCE in clinical practice in a Northern Italian Region (Lombardia). MATERIALS AND METHODS: Twenty-three out of 29 invited centers fulfilled a specific questionnaire. RESULTS: Between 2001 and 2008, 2921 procedures were performed and both the number of centers performing SBCE (from 5 to 29) and the number of SBCE (from 7.2 to 69.2 per month) increased steadily. The main indications for SBCE were: obscure gastrointestinal bleeding (OGIB) (43.4%), unexplained anemia (23.9%), suspected Crohn's disease (7.8%) and abdominal pain (5.3%). Overall, SBCE was positive in 50% of cases, negative in 36% and undefined in 14%. The highest diagnostic yields were observed in patients with OGIB (62.5%), polypoid syndromes (74.1%), known (54.8%) or suspected (47.3%) inflammatory bowel disease, while the yields were low in patients examined for chronic diarrhea (27.4%) and abdominal pain (14.9%), 61 patients (2.1%) experienced capsule retention. Thirty-two of them eventually excreted the capsule naturally while endoscopic or surgical retrieval was necessary in 29 (1%) (in two because of obstruction). CONCLUSION: Over a period of 7 years the use of SBCE in Lombardia increased steadily confirming, in clinical practice, a high diagnostic yield and an acceptable safety profile.

Single bolus of midazolam versus bolus midazolam plus meperidine for colonoscopy: a prospective, randomized, double-blind trial.

BACKGROUND: The aim of this study was to determine whether a single bolus of meperidine in addition to midazolam improves patient tolerance during colonoscopy. METHODS: Consecutive patients undergoing outpatient colonoscopy were randomly assigned in double-blind fashion to receive a single rapid intravenous bolus of 5 mg of midazolam and placebo (Group A, n = 125) or 5 mg midazolam plus 50 mg meperidine (Group B, n = 128). Tolerance (4-point scale: 1 excellent, 4 unbearable), pain (4-point scale: 1 none, 4 severe) and willingness to undergo another colonoscopy were assessed 24 to 48 hours later in a telephone interview conducted by an independent observer blinded to the regimen of sedative medication. RESULTS: Significantly more patients in Group A reported moderate or severe pain (28% vs. 9%; p < 0.001), poor or unbearable tolerance (18% vs. 6%; p < 0.01) and unwillingness to undergo colonoscopy again in the future (14% vs. 5%; p < 0.05). By multivariate analysis, randomization to the midazolam group and younger age were the only variables independently associated with the risk of reporting at least one of these outcomes. Recovery time, frequency of oxygen desaturation, and need for supplemental oxygen were not significantly different between the 2 groups. CONCLUSIONS: The addition of a single bolus of meperidine to midazolam improves patient tolerance and lessens pain during colonoscopy without significantly increasing the frequency of side effects or prolonging recovery time.