One-fourth of colorectal neoplasias are missed during screening colonoscopies; these can develop into colorectal cancer (CRC). Deep learning systems allow for real-time computer-aided detection ...(CADe) of polyps with high accuracy. We performed a multicenter, randomized trial to assess the safety and efficacy of a CADe system in detection of colorectal neoplasias during real-time colonoscopy.
We analyzed data from 685 subjects (61.32 ± 10.2 years old; 337 men) undergoing screening colonoscopies for CRC, post-polypectomy surveillance, or workup due to positive results from a fecal immunochemical test or signs or symptoms of CRC, at 3 centers in Italy from September through November 2019. Patients were randomly assigned (1:1) to groups who underwent high-definition colonoscopies with the CADe system or without (controls). The CADe system included an artificial intelligence–based medical device (GI-Genius, Medtronic) trained to process colonoscopy images and superimpose them, in real time, on the endoscopy display a green box over suspected lesions. A minimum withdrawal time of 6 minutes was required. Lesions were collected and histopathology findings were used as the reference standard. The primary outcome was adenoma detection rate (ADR, the percentage of patients with at least 1 histologically proven adenoma or carcinoma). Secondary outcomes were adenomas detected per colonoscopy, non-neoplastic resection rate, and withdrawal time.
The ADR was significantly higher in the CADe group (54.8%) than in the control group (40.4%) (relative risk RR, 1.30; 95% confidence interval CI, 1.14–1.45). Adenomas detected per colonoscopy were significantly higher in the CADe group (mean, 1.07 ±1.54) than in the control group (mean 0.71 ± 1.20) (incidence rate ratio, 1.46; 95% CI, 1.15–1.86). Adenomas 5 mm or smaller were detected in a significantly higher proportion of subjects in the CADe group (33.7%) than in the control group (26.5%; RR, 1.26; 95% CI, 1.01–1.52), as were adenomas of 6 to 9 mm (detected in 10.6% of subjects in the CADe group vs 5.8% in the control group; RR, 1.78; 95% CI, 1.09–2.86), regardless of morphology or location. There was no significant difference between groups in withdrawal time (417 ± 101 seconds for the CADe group vs 435 ± 149 for controls; P = .1) or proportion of subjects with resection of non-neoplastic lesions (26.0% in the CADe group vs 28.7% of controls; RR, 1.00; 95% CI, 0.90–1.12).
In a multicenter, randomized trial, we found that including CADe in real-time colonoscopy significantly increases ADR and adenomas detected per colonoscopy without increasing withdrawal time. ClinicalTrials.gov no: 04079478
Display omitted
Use of artificial intelligence may increase detection of colorectal neoplasia at colonoscopy by improving lesion recognition (CADe) and reduce pathology costs by improving optical diagnosis (CADx).
A ...multicenter library of ≥ 200 000 images from 1572 polyps was used to train a combined CADe/CADx system. System testing was performed on two independent image sets (CADe: 446 with polyps, 234 without; CADx: 267) from 234 polyps, which were also evaluated by six endoscopists (three experts, three non-experts).
CADe showed sensitivity, specificity, and accuracy of 92.9 %, 90.6 %, and 91.7 %, respectively. Experts showed significantly higher accuracy and specificity, and similar sensitivity, while non-experts + CADe showed comparable sensitivity but lower specificity and accuracy than CADe and experts. CADx showed sensitivity, specificity, and accuracy of 85.0 %, 79.4 %, and 83.6 %, respectively. Experts showed comparable performance, whereas non-experts + CADx showed comparable accuracy but lower specificity than CADx and experts.
The high accuracy shown by CADe and CADx was similar to that of experts, supporting further evaluation in a clinical setting. When using CAD, non-experts achieved a similar performance to experts, with suboptimal specificity.
Background and Aims Flexible endoscopic septum division (FESD) is a rapidly evolving technique for the treatment of Zenker's diverticulum (ZD). The aim was to perform a systematic review and ...meta-analysis of the literature focusing on FESD for ZD, including an in-depth evaluation of its efficacy, safety, and limitations. Methods A comprehensive literature search was completed to identify papers that examined the efficacy and safety of FESD for ZD. Demographic, clinical, and technical information was retrieved. Main outcomes were extracted, pooled, and analyzed. Heterogeneity among studies was assessed using the I2 statistic. A random effect model was used as the pooling method in cases of high heterogeneity; otherwise the fixed effect model was applied. Meta-regression was also performed. Main outcomes such as rates of success, adverse events, and recurrences were evaluated. Results Twenty studies with a total of 813 patients were selected. The pooled success, adverse events, and recurrence rates were 91% (95% confidence interval CI, 86%–95%; I2 = 69.5%), 11.3% (95% CI, 8%–16%; I2 = 64%), and 11% (95% CI, 8%–15%; I2 = 38.4%), respectively. Substantial heterogeneity across studies was found. However, for success rates, excluding 3 studies reduced heterogeneity to non-significant rates I2 = 25.6%; P = .154. Adverse event rates decreased with larger samples (coefficient, −0.0123; 95% CI, −0.03 to −0.003; P = .017), whereas recurrence rates increased (coefficient, 0.006; 95% CI, −0.0010 to 0.0125; P = .093). Year of publication was negatively associated with success rate, whereas the opposite pattern was found for recurrence rates. Conclusions FESD is a feasible, safe, and effective treatment for symptomatic ZD, with low adverse event and recurrence rates.
Colon capsule endoscopy (CCE) is a noninvasive technique used to explore the colon without sedation or air insufflation. A second-generation capsule was recently developed to improve accuracy of ...detection, and clinical use has expanded globally. We performed a systematic review and meta-analysis to assess the accuracy of CCE in detecting colorectal polyps.
We searched MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, and other databases from 1966 through 2015 for studies that compared accuracy of colonoscopy with histologic evaluation with CCE. The risk of bias within each study was ascertained according to Quality Assessment of Diagnostic Accuracy in Systematic Reviews recommendations. Per-patient accuracy values were calculated for polyps, overall and for first-generation (CCE-1) and second-generation (CCE-2) capsules. We analyzed data by using forest plots, the I2 statistic to calculate heterogeneity, and meta-regression analyses.
Fourteen studies provided data from 2420 patients (1128 for CCE-1 and 1292 for CCE-2). CCE-2 and CCE-1 detected polyps >6 mm with 86% sensitivity (95% confidence interval CI, 82%–89%) and 58% sensitivity (95% CI, 44%–70%), respectively, and 88.1% specificity (95% CI, 74.2%–95.0%) and 85.7% specificity (95% CI, 80.2%–90.0%), respectively. CCE-2 and CCE-1 detected polyps >10 mm with 87% sensitivity (95% CI, 81%–91%) and 54% sensitivity (95% CI, 29%–77%), respectively, and 95.3% specificity (95% CI, 91.5%–97.5%) and 97.4% specificity (95% CI, 96.0%–98.3%), respectively. CCE-2 identified all 11 invasive cancers detected by colonoscopy.
The sensitivity in detection of polyps >6 mm and >10 mm increased substantially between development of first-generation and second-generation colon capsules. High specificity values for detection of polyps by CCE-2 seem to be achievable with a 10-mm cutoff and in a screening setting.
Abstract
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 endoscopistsʼ high level confidence and expertise. The AI system seems to be useful, especially for nonexperts.
False positive (FP) results by computer-aided detection (CADe) hamper the efficiency of colonoscopy by extending examination time. Our aim was to develop a classification of the causes and clinical ...relevance of CADe FPs, and to assess the relative distribution of FPs in a real-life setting.
In a post-hoc analysis of a randomized trial comparing colonoscopy with and without CADe (NCT: 04079478), we extracted 40 CADe colonoscopy videos. Using a modified Delphi process, 4 expert endoscopists identified the main domains for the reasons and clinical relevance of FPs. Then, 2 expert endoscopists manually examined each FP and classified it according to the proposed domains. The analysis was limited to the withdrawal phase.
The 2 main domains for the causes of CADe FPs were identified as artifacts due to either the mucosal wall or bowel content, and clinical relevance was defined as the time spent on FPs and the FP rate per minute. The mean number of FPs per colonoscopy was 27.3 ± 13.1, of which 24 ± 12 (88%) and 3.2 ± 2.6 (12%) were due to artifacts in the bowel wall and bowel content, respectively. Of the 27.3 FPs per colonoscopy, 1.6 (5.7%) required additional exploration time of 4.8 ± 6.2 seconds per FP (ie, 0.7% of the mean withdrawal time). In detail, 15 (24.2%), 33 (53.2%), and 14 (22.6%) FPs were classified as being of mild, moderate, or severe clinical relevance. The rate of FPs per minute of withdrawal time was 2.4 ± 1.2, and was higher for FPs due to artifacts from the bowel wall than for those from bowel content (2.4 ± 0.6 vs 0.3 ± 0.2, P < .001).
FPs by CADe are primarily due to artifacts from the bowel wall. Despite a high frequency, FPs result in a negligible 1% increase in the total withdrawal time because most of them are immediately discarded by the endoscopist.
Background Differential diagnosis of pancreatic solid masses with EUS-guided FNA (EUS-FNA) is still challenging in about 15% of cases. Mutation of the K- ras gene is present in over 75% of pancreatic ...adenocarcinomas (PADC). Objective To assess the accuracy of K- ras gene mutation analysis for diagnosing PADC. Design We systematically searched the electronic databases for relevant studies published. Data from selected studies underwent meta-analysis by use of a bivariate model providing a pooled value for sensitivity, specificity, diagnostic odds ratio, and summary receiver operating characteristic curve. Setting Meta-analysis of 8 prospective studies. Patients Total of 931 patients undergoing EUS-FNA for diagnosis of pancreatic solid masses. Intervention K- ras mutation analysis. Main Outcome Measurements Diagnostic accuracy of K- ras mutation analysis and of combined diagnostic strategy by using EUS-FNA and K- ras mutation analysis in the diagnosis of PADC. Results The pooled sensitivity of EUS-FNA for the differential diagnosis of PADC was 80.6%, and the specificity was 97%. Estimated sensitivity and specificity were 76.8% and 93.3% for K- ras gene analysis, respectively, and 88.7% and 92% for combined EUS-FNA plus K- ras mutation analysis. Overall, K- ras mutation testing applied to cases that were inconclusive by EUS-FNA reduced the false-negative rate by 55.6%, with a false-positive rate of 10.7%. Not repeating EUS-FNA in cases in which mutation testing of the K- ras gene is inconclusive would reduce the repeat-biopsy rate from 12.5% to 6.8%. Limitations Small number of studies and between-study heterogeneity. Conclusion K- ras mutation analysis can be useful in the diagnostic work-up of pancreatic masses, in particular when tissue obtained by EUS-FNA is insufficient, and the diagnosis inconclusive.
Outcomes of endoscopic surveillance after surgery for colorectal cancer (CRC) vary with the incidence and timing of CRC detection at anastomoses or non-anastomoses in the colorectum. We performed a ...systematic review and meta-analysis to evaluate the incidence of CRCs identified during surveillance colonoscopies of patients who have already undergone surgery for this cancer.
We searched PubMed, EMBASE, SCOPUS, and the Cochrane Central Register of Clinical Trials through January 1, 2018 to identify studies investigating rates of CRCs at anastomoses or other locations in the colorectum after curative surgery for primary CRC. We collected data from published randomized controlled, prospective, and retrospective cohort studies. Data were analyzed by multivariate meta-analytic models.
From 2373 citations, we selected 27 studies with data on 15,803 index CRCs for analysis (89% of patients with stage I–III CRC). Overall, 296 CRCs at non-anastomotic locations were reported over time periods of more than 16 years (cumulative incidence, 2.2% of CRCs; 95% confidence interval CI, 1.8%–2.9%). The risk of CRC at a non-anastomotic location was significantly reduced more than 36 months after resection compared with before this time point (odds ratio for non-anastomotic CRCs at 36–48 months vs 6–12 months after surgery, 0.61; 95% CI, 0.37–0.98; P = .031); 53.7% of all non-anastomotic CRCs were detected within 36 months of surgery. One hundred and fifty-eight CRCs were detected at anastomoses (cumulative incidence of 2.7%; 95% CI, 1.9%–3.9%). The risk of CRCs at anastomoses was significantly lower 24 months after resection than before (odds ratio for CRCs at anastomoses at 25–36 months after surgery vs 6–12 months, 0.56; 95% CI, 0.32–0.98; P = .036); 90.8% of all CRCs at anastomoses were detected within 36 months of surgery.
After surgery for CRC, the highest risk of CRCs at anastomoses and at other locations in the colorectum is highest during 36 months after surgery—risk decreases thereafter. Patients who have undergone CRC resection should be evaluated by colonoscopy more closely during this time period. Longer intervals may be considered thereafter.
Display omitted
EUS elastography is useful in characterizing solid pancreatic lesions (SPLs), and fractal analysis–based technology has been used to evaluate geometric complexity in oncology. The aim of this study ...was to evaluate EUS elastography (strain ratio) and fractal analysis for the characterization of SPLs.
Consecutive patients with SPLs were prospectively enrolled between December 2015 and February 2017. Elastographic evaluation included parenchymal strain ratio (pSR) and wall strain ratio (wSR) and was performed with a new compact US processor. Elastographic images were analyzed using a computer program to determine the 3-dimensional histogram fractal dimension. A composite cytology/histology/clinical reference standard was used to assess sensitivity, specificity, positive predictive value, negative predictive value, and area under the receiver operating curve.
Overall, 102 SPLs from 100 patients were studied. At final diagnosis, 69 (68%) were malignant and 33 benign. At elastography, both pSR and wSR appeared to be significantly higher in malignant as compared with benign SPLs (pSR, 24.5 vs 6.4 P < .001; wSR, 56.6 vs 15.3 P < .001). When the best cut-off levels of pSR and wSR at 9.10 and 16.2, respectively, were used, sensitivity, specificity, positive predictive value, negative predictive value, and area under the receiver operating curve were 88.4%, 78.8%, 89.7%, 76.9%, and 86.7% and 91.3%, 69.7%, 86.5%, 80%, and 85.7%, respectively. Fractal analysis showed a significant statistical difference (P = .0087) between the mean surface fractal dimension of malignant lesions (D = 2.66 ± .01) versus neuroendocrine tumor (D = 2.73 ± .03) and a statistical difference for all 3 channels red, green, and blue (P < .0001).
EUS elastography with pSR and fractal-based analysis are useful in characterizing SPLs. (Clinical trial registration number: NCT02855151.)
Display omitted
Endoscopic ultrasound-guided sampling (EUS sampling) is a safe and effective technique. The study aim was to evaluate the presence of a histological core from pancreatic lesions using a new 25G ...fork-tip needle.
Observational multicenter prospective and analytical study, including consecutive patients with solid pancreatic masses referred for EUS-guided sampling. At each needle pass, the endoscopist performed macroscopic on-site evaluation (MOSE). The primary outcome was the histological core procurement rates. Secondary outcomes were the evaluation of interobserver agreement between endoscopists and pathologists, adequacy of EUS samples for the diagnosis and post-procedure adverse events.
100 patients were enrolled in 3 centers. The mean size of the lesions was 28.5 mm (SD 11.7). Final diagnoses were adenocarcinoma (68%), neuroendocrine tumor (21%), inflammatory mass/benign lesions (8.0%), and pancreatic metastasis (3.0%). The pathologists described the presence of a core in 67 samples (67.0% of patients), with poor agreement with MOSE (kappa, 0. 12; 95% CI: 0.03–0.28). The diagnostic accuracy was 93%. We observed 6% of mild adverse events.
The new 25-gauge core needle showed good overall adequacy and a good rate of histological specimens during EUS sampling of solid pancreatic masses, with a minimum number of passes and no major complications. Clinicaltrial.gov number, NCT02946840.