Inflammatory bowel disease (IBD) is a chronic inflammation of the gastrointestinal tract. IBD is classified into two main types: Crohn's disease (CD) and ulcerative colitis (UC). The prevalence of CD and UC is the highest in Europe, at 322 and 505 per 100,000 people, respectively (Molodecky et al., 2012). Conventionally, the severity of IBD is diagnosed using histopathological examination performed by a trained pathologist. Morphological changes such as crypt distortion, presence of infiltrates in the lamina propria, and erosion of the epithelial layer are used as inflammatory markers to predict disease stage and plan clinical therapy. Say no to plagiarism. Get a tailor-made essay on “Why Violent Video Games Should Not Be Banned”? Get an original essay Over the past decade, label-free multiphoton microscopy (MPM) has been recognized as a real-time in vivo imaging technique for IBD. Its increased penetration depth, high spatial resolution and molecular specificity have accelerated the diagnosis of IBD. MPM techniques such as two-photon excited fluorescence (TPEF) and second harmonic generation (SHG) as well as coherent anti-Stokes Raman scattering (CARS) can be used to visualize molecular changes associated with IBD (Schürmann et al ., 2013).Chernavskaia et al. used intensity-related properties of CARS/TPEF/SHG and crypt morphology to assign the histological index to a tissue section from an IBD patient. In their study, the mucosal and crypt regions were annotated by a trained pathologist, which is a time-consuming and laborious task (Chernavskaia et al., 2016). Therefore, automatic segmentation of the crypt region and mucosa using a multimodal image is a prerequisite for estimating the histological index associated with different stages of IBD. Nevertheless, automatic segmentation of the crypt and mucosa region is a very difficult task due to several reasons. First, crypt morphology changes among patients with different pathological activity. The crypt structure is distorted for patients with higher stage MII. Second, the crypts are located in the mucosal region and therefore the two regions overlap, making classification even more difficult. Third, it is difficult to identify clear boundaries of the crypt structure because the crypts are very close to each other. Finally, the availability of annotated medical data that captures various tissue structures of an IBD patient is limited. Therefore, segmentation of these regions by image processing and conventional machine learning techniques is inefficient. Semantic segmentation using a deep convolutional neural network (DCNN) has achieved positive results in the past. Deep neural networks like U-Net and SegNet have been used for biomedical image segmentation and are the gold standard for pixel-wise segmentation. In this paper, we propose automatic segmentation of multimodal images into four regions using a DCNN. Furthermore, we compare the segmentation results obtained by DCNN with the classical machine learning approach. Keep in mind: this is just a sample. Get a personalized article from our expert writers now. Get a Custom Essay The article is organized as follows, in section (2) we present previous works related to gland segmentation using histological images, in section (5).