Computer Aided Diagnosis for Confocal Laser Endomicroscopy in Advanced Colorectal Adenocarcinoma

Abstract

IntroductionConfocal laser endomicroscopy (CLE) is becoming a popular method for optical biopsy of digestive mucosa for both diagnostic and therapeutic procedures. Computer aided diagnosis of CLE images, using image processing and fractal analysis can be used to quantify the histological structures in the CLE generated images. The aim of this study is to develop an automatic diagnosis algorithm of colorectal cancer (CRC), based on fractal analysis and neural network modeling of the CLE-generated colon mucosa images.Materials and MethodsWe retrospectively analyzed a series of 1035 artifact-free endomicroscopy images, obtained during CLE examinations from normal mucosa (356 images) and tumor regions (679 images). The images were processed using a computer aided diagnosis (CAD) medical imaging system in order to obtain an automatic diagnosis. The CAD application includes image reading and processing functions, a module for fractal analysis, grey-level co-occurrence matrix (GLCM) computation module, and a feature identification module based on the Marching Squares and linear interpolation methods. A two-layer neural network was trained to automatically interpret the imaging data and diagnose the pathological samples based on the fractal dimension and the characteristic features of the biological tissues.ResultsNormal colon mucosa is characterized by regular polyhedral crypt structures whereas malignant colon mucosa is characterized by irregular and interrupted crypts, which can be diagnosed by CAD. For this purpose, seven geometric parameters were defined for each image: fractal dimension, lacunarity, contrast correlation, energy, homogeneity, and feature number. Of the seven parameters only contrast, homogeneity and feature number were significantly different between normal and cancer samples. Next, a two-layer feed forward neural network was used to train and automatically diagnose the malignant samples, based on the seven parameters tested. The neural network operations were cross-entropy with the results: training: 0.53, validation: 1.17, testing: 1.17, and percent error, resulting: training: 16.14, validation: 17.42, testing: 15.48. The diagnosis accuracy error was 15.5%.ConclusionsComputed aided diagnosis via fractal analysis of glandular structures can complement the traditional histological and minimally invasive imaging methods. A larger dataset from colorectal and other pathologies should be used to further validate the diagnostic power of the method.

Artificial intelligence and colonoscopy: Current status and future perspectives

“…Two were focused on automated pathological prediction and two exclusively on quantitative image quality control steps that aided the physician's interpretation of confocal endomicroscopic images . The former two studies were conducted by Andre et al ., who reported 89.6% accuracy in the differentiation of adenoma from non‐neoplastic polyps, and Stefanescu et al ., who showed 84.5% accuracy in distinguishing advanced CRC from normal colon mucosa . Both diagnostic algorithms were based on k‐nearest neighbor classification and neural network analysis.…”

“…51,52 The former two studies were conducted by Andre et al, who reported 89.6% accuracy in the differentiation of adenoma from nonneoplastic polyps, 49 and Stefanescu et al, who showed 84.5% accuracy in distinguishing advanced CRC from normal colon mucosa. 50 Both diagnostic algorithms were based on k-nearest neighbor classification and neural network analysis. Unfortunately, they were evaluated in experimental settings only, with no follow-up clinical trials.…”

Quality assurance of computer-aided detection and diagnosis in colonoscopy

“…Compared with CADe in which white-light endoscopy is used as the target of the image analysis, several optical technologies can be used for CADx: white light endoscopy, 25,26 magnifying narrow-band imaging (NBI), [27][28][29][30][31][32] magnifying chromoendoscopy, 33 endocytoscopy, 13,[34][35][36][37] confocal laser endomicroscopy, 38,39 spectroscopy, 40,41 and autofluorescence endoscopy. 42,43,44 Among these, the most extensively studied has been magnifying NBI; 29 probably because it may have better diagnostic performance than nonmagnified NBI and does not require staining like dye-based chromoendoscopy that can be time consuming in routine clinical use.…”

“…After several pilot studies, 34,35,37,45,46 this research group conducted a large-scale prospective study using CADx, demonstrating 91.4% sensitivity, 91.7% specificity, 88.9% PPV, and 93.7% NPV in the classification of diminutive rectosigmoid adenomas. CADx has also been explored for other modalities such as confocal laser endomicroscopy 38,39 and autofluorescence endoscopy. 42,43 However, the number of publications and performance of the developed models are limited when compared with the aforementioned modalities.…”

Optical biopsy in gastroenterology: Focus on confocal laser endomicroscopy

“…It is likely that fully automated computer-aided diagnosis will reduce the dependence on optical diagnosis obtained in real time through CLE. In addition, it may reduce the interobserver variability associated with qualitative CLE analysis of images [33,34].…”