Dataset Information

Effectiveness of computer-aided diagnosis of colorectal lesions using novel software for magnifying narrow-band imaging: a pilot study.

ABSTRACT: Magnifying narrow-band imaging (M-NBI) enables detailed observation of microvascular architecture and can be used in endoscopic diagnosis of colorectal lesion. However, in clinical practice, differential diagnosis and estimation of invasion depth of colorectal lesions based on M-NBI findings require experience. Therefore, developing computer-aided diagnosis (CAD) for M-NBI would be beneficial for clinical practice. The aim of this study was to evaluate the effectiveness of software for CAD of colorectal lesions.In collaboration with Yamaguchi University, we developed novel software that enables CAD of colorectal lesions using M-NBI images. This software for CAD further specifically divides original Sano's colorectal M-NBI classification into 3 groups (group A, capillary pattern [CP] type I; group B, CP type II + CP type IIIA; group C, CP type IIIB), which describe hyperplastic polyps (HPs), adenoma/adenocarcinoma (intramucosal [IM] to submucosal [SM]-superficial) lesions, and SM-deep lesions, respectively. We retrospectively reviewed 121 lesions evaluated using M-NBI.The 121 reviewed lesions included 21 HP, 80 adenoma/adenocarcinoma (IM to SM-superficial), and 20 SM-deep lesions. The concordance rate between the CAD and the diagnosis of the experienced endoscopists was 90.9 %. The sensitivity, specificity, positive and negative predictive values, and accuracy of the CAD for neoplastic lesions were 83.9 %, 82.6 %, 53.1 %, 95.6 %, and 82.8 %, respectively. The values for SM-deep lesions were 83.9 %, 82.6 %, 53.1 %, 95.6 %, and 82.8 %, respectively. Relatively high diagnostic values were obtained using CAD. This software for CAD could possibly lead to a wider use of M-NBI in the endoscopic diagnosis of colorectal lesions.

SUBMITTER: Tamai N

PROVIDER: S-EPMC5542817 | biostudies-other | 2017 Aug

REPOSITORIES: biostudies-other

ACCESS DATA

Similar Datasets

Project description:BackgroundIn recent years, two new narrow-band imaging (NBI) classifications have been proposed: The NBI international colorectal endoscopic (NICE) classification and Japanese NBI expert team (JNET) classification. Most validation studies of the two new NBI classifications were conducted in classification setting units by experienced endoscopists, and the application of use in different centers among endoscopists with different endoscopy skills remains unknown.AimTo evaluate clinical application and possible problems of NICE and JNET classification for the differential diagnosis of colorectal cancer and precancerous lesions.MethodsSix endoscopists with varying levels of experience participated in this study. Eighty-seven consecutive patients with a total of 125 lesions were photographed during non-magnifying conventional white-light colonoscopy, non-magnifying NBI, and magnifying NBI. The three groups of endoscopic pictures of each lesion were evaluated by the six endoscopists in randomized order using the NICE and JENT classifications separately. Then we calculated the six endoscopists' sensitivity, specificity, accuracy, positive predictive value, and negative predictive value for each category of the two classifications.ResultsThe sensitivity, specificity, and accuracy of JNET classification type 1 and 3 were similar to NICE classification type 1 and 3 in both the highly experienced endoscopist (HEE) and less-experienced endoscopist (LEE) groups. The specificity of JNET classification type 1 and 3 and NICE classification type 3 in both the HEE and LEE groups was > 95%, and the overall interobserver agreement was good in both groups. The sensitivity of NICE classification type 3 lesions for diagnosis of SM-d carcinoma in the HEE group was significantly superior to that in the LEE group (91.7% vs 83.3%; P = 0.042). The sensitivity of JNET classification type 2B lesions for the diagnosis of high-grade dysplasia or superficial submucosal invasive carcinoma in the HEE and LEE groups was 53.8% and 51.3%, respectively. Compared with other types of JNET classification, the diagnostic ability of type 2B was the weakest.ConclusionThe treatment strategy of the two classification type 1 and 3 lesions can be based on the results of endoscopic examination. JNET type 2B lesions need further examination.

Project description:OBJECTIVES: Magnifying narrow-band imaging (M-NBI) is more accurate than white-light imaging for diagnosing small gastric cancers. However, it is uncertain whether moving M-NBI images have additional effects in the diagnosis of gastric cancers compared with still images. DESIGN: A prospective multicenter cohort study. METHODS: To identify the additional benefits of moving M-NBI images by comparing the diagnostic accuracy of still images only with that of both still and moving images. Still and moving M-NBI images of 40 gastric lesions were obtained by an expert endoscopist prior to this prospective multicenter cohort study. Thirty-four endoscopists from ten different Japanese institutions participated in the prospective multicenter cohort study. Each study participant was first tested using only still M-NBI images (still image test), then tested 1 month later using both still and moving M-NBI images (moving image test). The main outcome was a difference in the diagnostic accuracy of cancerous versus noncancerous lesions between the still image test and the moving image test. RESULTS: Thirty-four endoscopists were analysed. There were no significant difference of cancerous versus noncancerous lesions between still and moving image tests in the diagnostic accuracy (59.9% versus 61.5%), sensitivity (53.4% versus 55.9%), and specificity (67.0% versus 67.6%). And there were no significant difference in the diagnostic accuracy between still and moving image tests of demarcation line (65.4% versus 65.5%), microvascular pattern (56.7% versus 56.9%), and microsurface pattern (48.1% versus 50.9%). Diagnostic accuracy showed no significant difference between the still and moving image tests in the subgroups of endoscopic findings of the lesions. CONCLUSIONS: The addition of moving M-NBI images to still M-NBI images does not improve the diagnostic accuracy for gastric lesions. It is reasonable to concentrate on taking sharp still M-NBI images during endoscopic observation and use them for diagnosis. TRIAL REGISTRATION: Umin.ac.jp UMIN-CTR000008048.

Project description:AIM:To assess the utility of modified Sano's (MS) vs the narrow band imaging international colorectal endoscopic (NICE) classification in differentiating colorectal polyps. METHODS:Patients undergoing colonoscopy between 2013 and 2015 were enrolled in this trial. Based on the MS or the NICE classifications, patients were randomised for real-time endoscopic diagnosis. This was followed by biopsies, endoscopic or surgical resection. The endoscopic diagnosis was then compared to the final (blinded) histopathology. The primary endpoint was the sensitivity (Sn), specificity (Sp), positive predictive value (PPV) and negative predictive value (NPV) of differentiating neoplastic and non-neoplastic polyps (MS II/IIo / IIIa / IIIb vs I or NICE 1 vs 2/3). The secondary endpoints were "endoscopic resectability" (MS II/IIo/IIIa vs I/IIIb or NICE 2 vs 1/3), NPV for diminutive distal adenomas and prediction of post-polypectomy surveillance intervals. RESULTS:A total of 348 patients were evaluated. The Sn, Sp, PPV and NPV in differentiating neoplastic polyps from non-neoplastic polyps were, 98.9%, 85.7%, 98.2% and 90.9% for MS; and 99.1%, 57.7%, 95.4% and 88.2% for NICE, respectively. The area under the receiver operating characteristic curve (AUC) for MS was 0.92 (95%CI: 0.86-0.98); and AUC for NICE was 0.78 (95%CI: 0.69, 0.88). The Sn, Sp, PPV and NPV in predicting "endoscopic resectability" were 98.9%, 86.1%, 97.8% and 92.5% for MS; and 98.6%, 66.7%, 94.7% and 88.9% for NICE, respectively. The AUC for MS was 0.92 (95%CI: 0.87-0.98); and the AUC for NICE was 0.83 (95%CI: 0.75-0.90). The AUC values were statistically different for both comparisons (P = 0.0165 and P = 0.0420, respectively). The accuracy for diagnosis of sessile serrated adenoma/polyp (SSA/P) with high confidence utilizing MS classification was 93.2%. The differentiation of SSA/P from other lesions achieved Sp, Sn, PPV and NPV of 87.2%, 91.5%, 89.6% and 98.6%, respectively. The NPV for predicting adenomas in diminutive rectosigmoid polyps (n = 150) was 96.6% and 95% with MS and NICE respectively. The calculated accuracy of post-polypectomy surveillance for MS group was 98.2% (167 out of 170) and for NICE group was 92.1% (139 out of 151). CONCLUSION:The MS classification outperformed the NICE classification in differentiating neoplastic polyps and predicting endoscopic resectability. Both classifications met ASGE PIVI thresholds.

Project description:Endometriosis is a common gynaecological condition characterised by the growth of endometrial tissue outside the uterus and is associated with pain and infertility. Currently, the gold standard for endometriosis diagnosis is laparoscopic excision and histological identification of endometrial epithelial and stromal cells. There is, however, currently no known association between the histological appearance, size, morphology, or subtype of endometriosis and disease prognosis. In this study, we used histopathological software to identify and quantify the number of endometrial epithelial and stromal cells within excised endometriotic lesions and assess the relationship between the cell contents and lesion subtypes. Prior to surgery for suspected endometriosis, patients provided menstrual and abdominal pain and dyspareunia scores. Endometriotic lesions removed during laparoscopic surgery were collected and prepared for immunohistochemistry from 26 patients. Endometrial epithelial and stromal cells were identified with Cytokeratin and CD10 antibodies, respectively. Whole slide sections were digitised and the QuPath software was trained to automatically detect and count epithelial and stromal cells across the whole section. Using this classifier, we identified a significantly larger number of strongly labelled CD10 stromal cells (p = 0.0477) in deeply infiltrating lesions (99,970 ± 2962) compared to superficial lesions (2456 ± 859). We found the ratio of epithelial to stromal cells was inverted in deeply infiltrating endometriosis lesions compared to superficial peritoneal and endometrioma lesions and we subsequently identified a correlation between total endometrial cells and abdominal pain (p = 0.0005) when counted via the automated software. Incorporating histological software into current standard diagnostic pipelines may improve endometriosis diagnosis and provide prognostic information in regards to severity and symptoms and eventually provide the potential to personalise adjuvant treatment decisions.

Dataset Information

Effectiveness of computer-aided diagnosis of colorectal lesions using novel software for magnifying narrow-band imaging: a pilot study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets