A Marked Poisson Process Driven Latent Shape Model for 3D Segmentation of Reflectance Confocal Microscopy Image Stacks of Human Skin

Abstract

Segmenting objects of interest from 3D datasets is a common problem encountered in biological data. Small field of view and intrinsic biological variability combined with optically subtle changes of intensity, resolution and low contrast in images make the task of segmentation difficult, especially for microscopy of unstained living or freshly excised thick tissues. Incorporating shape information in addition to the appearance of the object of interest can often help improve segmentation performance. However, shapes of objects in tissue can be highly variable and design of a flexible shape model that encompasses these variations is challenging. To address such complex segmentation problems, we propose a unified probabilistic framework that can incorporate the uncertainty associated with complex shapes, variable appearance and unknown locations. The driving application which inspired the development of this framework is a biologically important segmentation problem: the task of automatically detecting and segmenting the dermal-epidermal junction (DEJ) in 3D reflectance confocal microscopy (RCM) images of human skin. RCM imaging allows noninvasive observation of cellular, nuclear and morphological detail. The DEJ is an important morphological feature as it is where disorder, disease and cancer usually start. Detecting the DEJ is challenging because it is a 2D surface in a 3D volume which has strong but highly variable number of irregularly spaced and variably shaped “peaks and valleys”. In addition, RCM imaging resolution, contrast and intensity vary with depth. Thus a prior model needs to incorporate the intrinsic structure while allowing variability in essentially all its parameters. We propose a model which can incorporate objects of interest with complex shapes and variable appearance in an unsupervised setting by utilizing domain knowledge to build appropriate priors of the model. Our novel strategy to model this structure combines a spatial Poisson process with shape priors and performs inference using Gibbs sampling. Experimental results show that the proposed unsupervised model is able to automatically detect the DEJ with physiologically relevant accuracy in the range 10 – 20µm.

Skin strata delineation in reflectance confocal microscopy images using recurrent convolutional networks with attention

“…Finally, we emphasize that all classifications were performed slice-wise, that is, an entire slice is classified as belonging to a single layer. This is in contrast to some of our group’s earlier work 20 , 22 which belongs to the first set of methods that find a continuous 3D boundary of the skin layers. We adopted the current approach because it is more relevant to the driving clinical need to select a depth at which to subsequently acquire mosaics.…”

Three-dimensional conditional random field for the dermal–epidermal junction segmentation

“…They also propose a second approach, which incorporates a mathematical shape model for the DEJ using a Bayesian model. 10 The DEJ shape is modeled using a marked Poisson process. Their model can account for uncertainty in number, location, shape, and appearance of the dermal papillae.…”

Utilizing Machine Learning for Image Quality Assessment for Reflectance Confocal Microscopy

“…Quantitative and objective analysis of RCM images via ML has been investigated in several studies (Bozkurt et al, 2017a(Bozkurt et al, ,b, 2018Ghanta et al, 2017;Hames et al, 2016;Kose et al, 2020;Kurugol et al, 2015). As most of the diagnostic information within the RCM images lies in the architectural and morphological patterns (Bozkurt et al, 2018;Gill et al, 2014), texture analysisebased methods have been our focus.…”

“…As most of the diagnostic information within the RCM images lies in the architectural and morphological patterns (Bozkurt et al, 2018;Gill et al, 2014), texture analysisebased methods have been our focus. Initial efforts concentrated on automated delineation of skin layers within RCM stacks (Bozkurt et al, 2017b;Ghanta et al, 2017;Hames et al, 2016;Kurugol et al, 2015) using hand-crafted mathematical relations between the pixels (intensity profile [Kurugol et al, 2015], textons [Julesz, 1981], log-Gabor [Field, 1985], or wavelet transform features [Laine and Fan, 1993;Randen and Husoy, 1999]). Similar textural analysis methods were also used by Koller et al (2011) to detect malignancy in RCM images.…”