Computer-aided classification of colorectal segments during colonoscopy: a deep learning approach based on images of a magnetic endoscopic positioning device

OBJECTIVE

Assessment of the anatomical colorectal segment of polyps during colonoscopy is important for treatment and follow-up strategies, but is largely operator dependent. This feasibility study aimed to assess whether, using images of a magnetic endoscope imaging (MEI) positioning device, a deep learning approach can be useful to objectively divide the colorectum into anatomical segments.

METHODS

Models based on the VGG-16 based convolutional neural network architecture were developed to classify the colorectum into anatomical segments. These models were pre-trained on ImageNet data and further trained using prospectively collected data of the POLAR study in which endoscopists were using MEI (3930 still images and 90,151 video frames). Five-fold cross validation with multiple runs was used to evaluate the overall diagnostic accuracies of the models for colorectal segment classification (divided into a 5-class and 2-class colorectal segment division). The colorectal segment assignment by endoscopists was used as the reference standard.

RESULTS

For the 5-class colorectal segment division, the best performing model correctly classified the colorectal segment in 753 of the 1196 polyps, corresponding to an overall accuracy of 63%, sensitivity of 63%, specificity of 89% and kappa of 0.47. For the 2-class colorectal segment division, 1112 of the 1196 polyps were correctly classified, corresponding to an accuracy of 93%, sensitivity of 93%, specificity of 90% and kappa of 0.82.

CONCLUSION

The diagnostic performance of a deep learning approach for colorectal segment classification based on images of a MEI device is yet suboptimal (clinicaltrials.gov: NCT03822390).

Linked Colour imaging for the detection of polyps in patients with Lynch syndrome: a multicentre, parallel randomised controlled trial

OBJECTIVE

Despite regular colonoscopy surveillance, colorectal cancers still occur in patients with Lynch syndrome. Thus, detection of all relevant precancerous lesions remains very important. The present study investigates Linked Colour imaging (LCI), an image-enhancing technique, as compared with high-definition white light endoscopy (HD-WLE) for the detection of polyps in this patient group.

DESIGN

This prospective, randomised controlled trial was performed by 22 experienced endoscopists from eight centres in six countries. Consecutive Lynch syndrome patients ≥18 years undergoing surveillance colonoscopy were randomised (1:1) and stratified by centre for inspection with either LCI or HD-WLE. Primary outcome was the polyp detection rate (PDR).

RESULTS

Between January 2018 and March 2020, 357 patients were randomised and 332 patients analysed (160 LCI, 172 HD-WLE; 6 excluded due to incomplete colonoscopies and 19 due to insufficient bowel cleanliness). No significant difference was observed in PDR with LCI (44.4%; 95% CI 36.5% to 52.4%) compared with HD-WLE (36.0%; 95% CI 28.9% to 43.7%) (p=0.12). Of the secondary outcome parameters, more adenomas were found on a patient (adenoma detection rate 36.3%; vs 25.6%; p=0.04) and a colonoscopy basis (mean adenomas per colonoscopy 0.65 vs 0.42; p=0.04). The median withdrawal time was not statistically different between LCI and HD-WLE (12 vs 11 min; p=0.16).

CONCLUSION

LCI did not improve the PDR compared with HD-WLE in patients with Lynch syndrome undergoing surveillance. The relevance of findings more adenomas by LCI has to be examined further.

TRIAL REGISTRATION NUMBER

NCT03344289.

Real-time diagnostic accuracy of blue light imaging, linked color imaging and white-light endoscopy for colorectal polyp characterization

Background and study aims  Fujifilm has developed a novel ELUXEO 7000 endoscope system that employs light-emitting diodes (LEDs) at four different wavelengths as light sources that enable blue light imaging (BLI), linked color imaging (LCI), and high-definition white-light endoscopy (HD-WLE). The aim of this study was to address the diagnostic accuracy of real-time polyp characterization using BLI, LCI and HD-WLE (ELUXEO 7000 endoscopy system). Patients methods  This is a prespecified post-hoc analysis of a prospective study in which 22 experienced endoscopists (> 2,000 colonoscopies) from eight international centers participated. Using a combination of BLI, LCI, and HD-WLE, lesions were endoscopically characterized including a high- or low-confidence statement. Per protocol, digital images were created from all three imaging modalities. Histopathology was the reference standard. Endoscopists were familiar with polyp characterization, but did not take dedicated training for purposes of this study. Results  Overall, 341 lesions were detected in 332 patients. Of the lesions, 269 histologically confirmed polyps with an optical diagnosis were included for analysis (165 adenomas, 27 sessile serrated lesions, and 77 hyperplastic polyps). Overall, polyp characterization was performed with high confidence in 82.9 %. The overall accuracy for polyp characterization was 75.1 % (95 % confidence interval [CI] 69.5-80.1 %), compared with an accuracy of 78.0 % (95 % CI 72.0-83.2 %) for high confidence assignments. The accuracy for endoscopic characterization for diminutive polyps was 74.7 % (95 %CI 68.4-80.3 %), compared with an accuracy of 78.2 % (95 % CI 71.4-84.0 %) for high-confidence assignments. Conclusions  The diagnostic accuracy of BLI, LCI, and HD-WLE by experienced endoscopist for real-time polyp characterization seems limited (NCT03344289).

Retrieval of chronic hepatitis C patients. A manifesto for action to eliminate hepatitis C in the Netherlands: the CELINE project

Chronic hepatitis C virus (HCV) infection is a global public health issue, which is associated with high rates of morbidity and mortality. The development of direct acting antivirals (DAAs) has transformed treatment: they offer us highly-effective therapy with superior tolerability compared to interferon-containing regimens. In 2016, the World Health Organization (WHO) therefore adopted several ambitious viral hepatitis elimination targets, aiming for a 90% reduction in new infections and a 65% reduction in mortality by 2030. The ultimate goal is to eliminate HCV completely. It is reasonable that these goals may be achieved in the Netherlands due to the low prevalence of chronic HCV, the availability of DAAs, and excellent healthcare infrastructure. This paper describes a national effort to curtail the HCV epidemic in the Netherlands through an HCV retrieval and linkage to care project (CELINE: Hepatitis C Elimination in the Netherlands).

Mutation prediction models in Lynch syndrome: evaluation in a clinical genetic setting

BACKGROUND/AIMS

The identification of Lynch syndrome is hampered by the absence of specific diagnostic features and underutilisation of genetic testing. Prediction models have therefore been developed, but they have not been validated for a clinical genetic setting. The aim of the present study was to evaluate the usefulness of currently available prediction models.

METHODS

The authors collected data of 321 index probands who were referred to the department of clinical genetics of the Erasmus Medical Center because of a family history of colorectal cancer. These data were used as input for five previously published models. External validity was assessed by discriminative ability (AUC: area under the receiver operating characteristic curve) and calibration. For further insight, predicted probabilities were categorised with cut-offs of 5%, 10%, 20% and 40%. Furthermore, costs of different testing strategies were related to the number of extra detected mutation carriers.

RESULTS

Of the 321 index probands, 66 harboured a germline mutation. All models discriminated well between high risk and low risk index probands (AUC 0.82-0.84). Calibration was well for the Premm(1,2) and Edinburgh model, but poor for the other models. Cut-offs could be found for the prediction models where costs could be saved while missing only few mutations.

CONCLUSIONS

The Edinburgh and Premm(1,2) model were the models with the best performance for an intermediate to high risk setting. These models may well be of use in clinical practice to select patients for further testing of mismatch repair gene mutations.

A high incidence of MSH6 mutations in Amsterdam criteria II-negative families tested in a diagnostic setting

BACKGROUND AND AIMS

In Lynch syndrome, the clinical phenotype in MSH6 mutation families differs from that in MLH1 and MSH2 families. Therefore, MSH6 mutation families are less likely to fulfil diagnostic criteria such as the Amsterdam II criteria (AC II) and the revised Bethesda guidelines (rBG), and will be underdiagnosed. The aim of the present study was to evaluate the contribution of MSH6 gene mutations in families that were analysed for Lynch syndrome in a diagnostic setting.

METHODS

Families that had molecular analysis for Lynch syndrome were included in this study. Complete molecular screening of the MLH1, MSH2 and MSH6 genes was performed in all families. Microsatellite instability (MSI) and immunohistochemical (IHC) analysis was performed in almost all families. Clinical data were collected from medical records and family pedigrees.

RESULTS

A total of 108 families were included. MSI and IHC analysis was performed in 97 families, and in 40 an MSI-high phenotype with absent protein expression was found. Germline mutation analysis detected mutations in 23 families (7 MLH1, 4 MSH2 and 12 MSH6). The majority of MSH6 families were AC II negative, but fulfilled the rBG.

CONCLUSIONS

There is a high incidence of MSH6 mutations in families tested for Lynch syndrome in a diagnostic setting. Many of these families remain underdiagnosed using the AC II. The rBG are more useful to select these families for further analysis. However, to optimise the detection of MSH6 families, MSI and IHC analysis should also be performed in families with clustering of late-onset endometrial carcinoma.

A high incidence of MSH6 mutations in Amsterdam criteria II-negative families tested in a diagnostic setting

BACKGROUND AND AIMS

In Lynch syndrome, the clinical phenotype in MSH6 mutation families differs from that in MLH1 and MSH2 families. Therefore, MSH6 mutation families are less likely to fulfil diagnostic criteria such as the Amsterdam II criteria (AC II) and the revised Bethesda guidelines (rBG), and will be underdiagnosed. The aim of the present study was to evaluate the contribution of MSH6 gene mutations in families that were analysed for Lynch syndrome in a diagnostic setting.

METHODS

Families that had molecular analysis for Lynch syndrome were included in this study. Complete molecular screening of the MLH1, MSH2 and MSH6 genes was performed in all families. Microsatellite instability (MSI) and immunohistochemical (IHC) analysis was performed in almost all families. Clinical data were collected from medical records and family pedigrees.

RESULTS

A total of 108 families were included. MSI and IHC analysis was performed in 97 families, and in 40 an MSI-high phenotype with absent protein expression was found. Germline mutation analysis detected mutations in 23 families (7 MLH1, 4 MSH2 and 12 MSH6). The majority of MSH6 families were AC II negative, but fulfilled the rBG.

CONCLUSIONS

There is a high incidence of MSH6 mutations in families tested for Lynch syndrome in a diagnostic setting. Many of these families remain underdiagnosed using the AC II. The rBG are more useful to select these families for further analysis. However, to optimise the detection of MSH6 families, MSI and IHC analysis should also be performed in families with clustering of late-onset endometrial carcinoma.

Review article: Detection and management of hereditary non-polyposis colorectal cancer (Lynch syndrome)

BACKGROUND

The most common hereditary colorectal cancer syndrome is hereditary non-polyposis colorectal cancer (HNPCC), also known as Lynch syndrome. Diagnosis of this syndrome is difficult, because of lack of specific diagnostic features.

AIM

To discuss the diagnostic criteria and laboratory work up for HNPCC. Furthermore, surveillance programs for HNPCC and treatment of HNPCC associated colorectal cancer are discussed.

RESULTS

Current diagnostic criteria, including the Amsterdam II and Bethesda criteria, are suboptimal for the detection of HNPCC. Molecular screening by microsatellite instability (MSI) and immunohistochemistry (IHC) is useful in the diagnosis of HNPCC. Both techniques have a higher sensitivity compared to the Amsterdam II and Bethesda criteria. A combination of both MSI and IHC provides the most optimal selection for mutation analysis. After identification of a mutation in an affected individual, genetic counselling and presymptomatic mutation analysis should be offered to relatives. Furthermore, colonoscopic surveillance should be performed in proven mutation carriers.

CONCLUSIONS

Identification of HNPCC is a clinical challenge involving many clinicians. Identification of persons at risk can be achieved by a combination of a detailed family history, testing with molecular and mutation analysis.

Population screening for colorectal cancer: faeces, endoscopes or X-rays?

Colorectal carcinoma (CRC) is a common cancer and the second most common cause of death. The therapeutic costs for this disease will continue to rise due to an increasing incidence and the introduction of new chemotherapeutic modalities. Colorectal carcinoma is preceded by precursor lesions, which can be used as a target for early detection and therapy. Biennial population screening with faecal occult blood tests (FOBT) lowers CRC mortality with 14-18%. Five year screening with flexible sigmoidoscopy is a cost-effective alternative, which yields a higher preventive effect when similar participation rates are achieved. Screening colonoscopy has the advantage of examination of the complete colon but disadvantages are the high participant burden and the higher demand for endoscopic personnel and endoscopy units. Future screening modalities like faecal DNA markers and CT colonography are promising but need further improvement. In Europe, faecal occult blood testing and flexible sigmoidoscopy are currently the most suitable screening modalities for colorectal cancer screening.

The assessment of the gingival capillary density with orthogonal spectral polarization (OPS) imaging

OBJECTIVE

In this study we evaluated the inter-observer agreement in the assessment of gingival capillary density using Orthogonal Polarization Spectral Imaging.

METHODS

In this study gingival capillary density of 100 healthy subjects was determined by 2 independent observers. Agreement was quantified by calculation of the mean differences between the observers and the standard deviation of this difference and the limits of agreement. Reliability was quantified by means of the intraclass correlation coefficient (ICC).

RESULTS

Fifty males and 50 females were included in the study. The mean age for males was 20+/-1.2 years and for females 20+/-1.4. OPS images showed remarkable good quality images of the gingival microcirculation. The interclass correlation between the 2 observers was 0.63 while the interclass correlation for the 6 measurements in observer 1 was 0.95 and 0.94 for observer 2. The mean capillary density for females in observer 1 was 83.69+/-16.4 and 83+/-16.0 in observer 2, versus 60.55+/-12.3 for observer 1 and 60.4+/-12.1 for males. The mean quantitative functional capillary density in male students was 60.48+/-10.7, compared to 83.45+/-13.5 in female students.

CONCLUSIONS

OPS imaging enabled for the first time direct in vivo visualization and quantification of human functional gingival capillary density in healthy medical students. The inter-observer agreement was found to be good to fair on the quantification of gingival capillary density between the two independent observers. The intracorrelation coefficient (0.95) was excellent when assessing the reliability of one observer.