Accurate segmentation of prostate cancer histomorphometric features using a weakly supervised convolutional neural network

Abstract

Purpose: Prostate cancer primarily arises from the glandular epithelium. Histomophometric techniques have been used to assess the glandular epithelium in automated detection and classification pipelines; however, they are often rigid in their implementation, and their performance suffers on large datasets where variation in staining, imaging, and preparation is difficult to control. The purpose of this study is to quantify performance of a pixelwise segmentation algorithm that was trained using different combinations of weak and strong stroma, epithelium, and lumen labels in a prostate histology dataset. Approach: We have combined weakly labeled datasets generated using simple morphometric techniques and high-quality labeled datasets from human observers in prostate biopsy cores to train a convolutional neural network for use in whole mount prostate labeling pipelines. With trained networks, we characterize pixelwise segmentation of stromal, epithelium, and lumen (SEL) regions on both biopsy core and whole-mount H&E-stained tissue. Results: We provide evidence that by simply training a deep learning algorithm on weakly labeled data generated from rigid morphometric methods, we can improve the robustness of classification over the morphometric methods used to train the classifier. Conclusions: We show that not only does our approach of combining weak and strong labels for training the CNN improve qualitative SEL labeling within tissue but also the deep learning generated labels are superior for cancer classification in a higher-order algorithm over the morphometrically derived labels it was trained on.

Deep Learning Methodologies Applied to Digital Pathology in Prostate Cancer: A Systematic Review

“…Precisely annotating WSIs is time-consuming, and several approaches focused on strategies to reduce the number of annotations. In [ 30 ], the authors trained a model with rough annotations obtained with traditional image processing techniques. Then they fine-tuned this model with a few precise annotations made by pathologists (AUC gain of 0.04).…”

Dynamic Region of Interest Selection in Remote Photoplethysmography: Proof-of-Concept Study

“…To ensure that the improvement in SNR-F score represented a genuine improvement in signal quality, signals from the same videos were processed using the 2 different methods and assessed by an observer blinded to the prior signal processing (midface analysis vs T&A). Human observation is an accepted method for the task-based evaluation of medical images [ 25 ] and has been used to validate the machine-based assessment of, for example, atrial fibrillation [ 10 ], prostate cancer histology [ 34 ], and breast cancer diagnosis [ 35 ]. Independent annotator assessment has also been used to validate the signal quality of photoplethysmography signals recorded by mobile phones [ 36 ].…”

Whole slide imaging (WSI) scanner differences influence optical and computed properties of digitized prostate cancer histology

“…A color deconvolution algorithm was then applied to both raw and histogram-matched images to project color data in terms of relative stain intensities, resulting in an image with color channels that represent hematoxylin, eosin, and residual color information (HER). 23 , 24 These color deconvolved channels were likewise isolated to test whether a color deconvolved image could correct for any RGB heterogeneity between scanners ( Fig. 2 ).…”