Classification criteria for advanced adenomas of the colon by using probe-based confocal laser endomicroscopy: a preliminary study

Classification of the Confocal Microscopy Images of Colorectal Tumor and Inflammatory Colitis Mucosa Tissue Using Deep Learning

Confocal microscopy image analysis is a useful method for neoplasm diagnosis. Many ambiguous cases are difficult to distinguish with the naked eye, thus leading to high inter-observer variability and significant time investments for learning this method. We aimed to develop a deep learning-based neoplasm classification model that classifies confocal microscopy images of 10× magnified colon tissues into three classes: neoplasm, inflammation, and normal tissue. ResNet50 with data augmentation and transfer learning approaches was used to efficiently train the model with limited training data. A class activation map was generated by using global average pooling to confirm which areas had a major effect on the classification. The proposed method achieved an accuracy of 81%, which was 14.05% more accurate than three machine learning-based methods and 22.6% better than the predictions made by four endoscopists. ResNet50 with data augmentation and transfer learning can be utilized to effectively identify neoplasm, inflammation, and normal tissue in confocal microscopy images. The proposed method outperformed three machine learning-based methods and identified the area that had a major influence on the results. Inter-observer variability and the time required for learning can be reduced if the proposed model is used with confocal microscopy image analysis for diagnosis.

The Status of Advanced Imaging Techniques for Optical Biopsy of Colonic Polyps

The progressive miniaturization of photonic components presents the opportunity to obtain unprecedented microscopic images of colonic polyps in real time during endoscopy. This information has the potential to act as "optical biopsy" to aid clinical decision-making, including the possibility of adopting new paradigms such as a "resect and discard" approach for low-risk lesions. The technologies discussed in this review include confocal laser endomicroscopy, optical coherence tomography, multiphoton microscopy, Raman spectroscopy, and hyperspectral imaging. These are in different stages of development and clinical readiness, but all show the potential to produce reliable in vivo discrimination of different tissue types. A structured literature search of the imaging techniques for colorectal polyps has been conducted. The significant developments in endoscopic imaging were identified for each modality, and the status of current development was discussed. Of the advanced imaging techniques discussed, confocal laser endomicroscopy is in clinical use and, under optimal conditions with an experienced operator, can provide accurate histological assessment of tissue. The remaining techniques show potential for incorporation into endoscopic equipment and practice, although further component development is needed, followed by robust prospective validation of accuracy. Optical coherence tomography illustrates tissue "texture" well and gives good assessment of mucosal thickness and layers. Multiphoton microscopy produces high-resolution images at a subcellular resolution. Raman spectroscopy and hyperspectral imaging are less developed endoscopically but provide a tissue "fingerprint" which can distinguish between tissue types. Molecular imaging may become a powerful adjunct to other techniques, with its ability to precisely label specific molecules within tissue and thereby enhance imaging.

Validation of Probe-based Confocal Laser Endomicroscopy (pCLE) Criteria for Diagnosing Colon Polyp Histology

BACKGROUND

Validated probe-based confocal endomicroscopy (pCLE) criteria for distinguishing hyperplastic polyps (HPs) and tubular adenomas (TA) have not yet been developed.

AIM

To develop pCLE criteria for distinguishing HP from TA and evaluate its performance characteristics among experts.

METHODS

pCLE criteria for colon polyp histology were developed and tested in 2 phases prospectively. Phase I: 8 preliminary criteria were developed and tested internally. Criteria achieving an accuracy of >75% (epithelial surface: regular vs. irregular; goblet cells: increased vs. decreased; gland axis: horizontal vs. vertical; gland shape: slit/stellate vs. villiform; image scale: gray vs. dark) were evaluated in Phase II of study wherein external assessors evaluated these criteria in a separate set of pCLE videos. Accuracy and interobserver agreement (95% confidence intervals) were determined for colon histology prediction.

RESULTS

Phase I (criteria development/internal testing): 8 criteria were assessed by 4 pCLE experts using 28 videos (14 HP/14 TA). Five of 8 pCLE criteria met selection for phase II (accuracy >75%). Phase II (external validation): 36 pCLE colon polyp videos (HP 16/TA 20) were evaluated by 8 external assessors. Overall accuracy in diagnosis of colon polyp histology was 84.9% (95% confidence interval, 81.7-87.7). Of predictions made with high confidence (75%), histology was predicted with an accuracy of 91%, sensitivity 83%, specificity 100%, negative predictive value 87% and positive predictive value 98%. Interobserver agreement was substantial (κ=0.73).

CONCLUSIONS

We demonstrate the development and validation of pCLE criteria for prediction of colon polyp histology. Using these criteria, overall accuracy in differentiating TA from HP was high with substantial interobserver agreement.

Confocal Laser Endomicroscopy in Gastrointestinal and Pancreatobiliary Diseases: A Systematic Review and Meta-Analysis

Confocal laser endomicroscopy (CLE) is an endoscopic-assisted technique developed to obtain histopathological diagnoses of gastrointestinal and pancreatobiliary diseases in real time. The objective of this systematic review is to analyze the current literature on CLE and to evaluate the applicability and diagnostic yield of CLE in patients with gastrointestinal and pancreatobiliary diseases. A literature search was performed on MEDLINE, EMBASE, Scopus, and Cochrane Oral Health Group Specialized Register, using pertinent keywords without time limitations. Both prospective and retrospective clinical studies that evaluated the sensitivity, specificity, or accuracy of CLE were eligible for inclusion. Of 662 articles identified, 102 studies were included in the systematic review. The studies were conducted between 2004 and 2015 in 16 different countries. CLE demonstrated high sensitivity and specificity in the detection of dysplasia in Barrett's esophagus, gastric neoplasms and polyps, colorectal cancers in inflammatory bowel disease, malignant pancreatobiliary strictures, and pancreatic cysts. Although CLE has several promising applications, its use has been limited by its low availability, high cost, and need of specific operator training. Further clinical trials with a particular focus on cost-effectiveness and medicoeconomic analyses, as well as standardized institutional training, are advocated to implement CLE in routine clinical practice.

Progress of research and clinical application of probe-based confocal laser endomicroscopy

Probe based confocal laser endomicroscopy (pCLE) is a new endoscopic diagnostic method which integrates micro-imaging technology into endoscopy and can identify the morphological structure of particular tissue surface, so that the ongoing procedure of endoscopy can make the histopathological diagnosis. pCLE has the "optical biopsy" function, avoids shortcomings in physical biopsies such as being time-consuming, cumbersome, and prone to sampling error, and thus improves the diagnostic accuracy and efficiency. The field of research of pCLE as a new endoscopic imaging technology has been expanded from the digestive cavity structure to the respiratory system and urinary system. This article reviews the progress of research and clinical application of pCLE.

Application and Efficacy of Super-Magnifying Endoscopy for the Lower Intestinal Tract

Endoscopy plays a significant role in the diagnosis, management, and surveillance of colorectal cancer (CRC) and inflammatory bowel diseases (IBDs). Moreover, magnifying endoscopy and image-enhanced endoscopy has a crucial role in the clinical setting. Recently, a super-magnifying endoscope has been developed, and two devices, confocal laser endomicroscopy (CLE) and an endocytoscopy system (ECS), which allow in vivo microscopic inspection of the microstructural mucosal features of the gastrointestinal tract, are currently available. Studies on the use of ECS in CRC were reported by a Japanese group. Additionally, a few studies on the use of ECS in IBD have been reported. CLE has been shown to be reliable in assessing the activity of the disease in IBDs in both ulcerative colitis and Crohn's disease. Various published studies evaluated the use of CLE during colonoscopy to distinguish colorectal polyp pathology and neoplasia. However, these studies are heterogeneous, and further evidence is necessary to confirm the efficacy of CLE.

In-vivo microscopy in the diagnosis of intestinal neoplasia and inflammatory conditions

Confocal laser endomicroscopy (CLE) is a rapidly emerging tool in endoscopic imaging allowing in-vivo microscopy of examined gastrointestinal mucosa. This review will discuss the most recent advances of confocal laser endomicroscopy in the diagnosis of intestinal neoplasia and inflammatory conditions.