Deep Learning Model for Cell Nuclei Segmentation and Lymphocyte Identification in Whole Slide Histology Images

Abstract

Anti-cancer immunotherapy dramatically changes the clinical management of many types of tumours towards less harmful and more personalized treatment plans than conventional chemotherapy or radiation. Precise analysis of the spatial distribution of immune cells in the tumourous tissue is necessary to select patients that would best respond to the treatment. Here, we introduce a deep learning-based workflow for cell nuclei segmentation and subsequent immune cell identification in routine diagnostic images. We applied our workflow on a set of hematoxylin and eosin (H&E) stained breast cancer and colorectal cancer tissue images to detect tumour-infiltrating lymphocytes. Firstly, to segment all nuclei in the tissue, we applied the multiple-image input layer architecture (Micro-Net, Dice coefficient (DC) 0.79 ± 0.02). We supplemented the Micro-Net with an introduced texture block to increase segmentation accuracy (DC = 0.80 ± 0.02). We preserved the shallow architecture of the segmentation network with only 280 K trainable parameters (e.g. U-net with ∼1900 K parameters, DC = 0.78 ± 0.03). Subsequently, we added an active contour layer to the ground truth images to further increase the performance (DC = 0.81 ± 0.02). Secondly, to discriminate lymphocytes from the set of all segmented nuclei, we explored multilayer perceptron and achieved a 0.70 classification f-score. Remarkably, the binary classification of segmented nuclei was significantly improved (f-score = 0.80) by colour normalization. To inspect model generalization, we have evaluated trained models on a public dataset that was not put to use during training. We conclude that the proposed workflow achieved promising results and, with little effort, can be employed in multi-class nuclei segmentation and identification tasks.

Regional Network Education Information Collection Platform for Smart Classrooms based on Big Data Technology

“…The deep integration of the Internet and education provides new ideas and new ways to meet people's needs and solve various problems. The research on big data of online education can provide better education services for educational activities, drive the reform of teaching mode and optimize the teaching methods of education [27,28]. Therefore, this article has developed a regional network education information collection platform based on big data technology.…”

Deep Learning-Based Mapping of Tumor Infiltrating Lymphocytes in Whole Slide Images of 23 Types of Cancer

“…Methods were also developed to study TILs in Hematoxylin and Eosin (H&E) stained tissue images. Budginaite et al (23) developed a deep learning workflow based on the Micro-Net architecture (24) and multi-layer perceptrons to identify lymphocytes in tissue images from breast and colorectal cancer cases. Corredor et al (25) investigated the spatial patterns of TILs in early stage nonsmall cell lung cancer cases with the goal of predicting cancer recurrence.…”

“…Budginaite et al. ( 23 ) developed a deep learning workflow based on the Micro-Net architecture ( 24 ) and multi-layer perceptrons to identify lymphocytes in tissue images from breast and colorectal cancer cases. Corredor et al.…”

ICOSeg: Real-Time ICOS Protein Expression Segmentation from Immunohistochemistry Slides Using a Lightweight Conv-Transformer Network

“…Recently, there has been a lot of research conducted on the advantages of computer-assisted quantitative cell count studies [ 12 ], cell segmentation tasks [ 13 ], and the effect of the automatic versus manual estimation [ 14 ]. Current artificial intelligence (AI) guided CAD systems show promising outcomes using deep learning algorithms in digital pathology, for example in predicting colon cancer metastases [ 15 ], nuclei segmentation [ 16 ], lymphocyte identification [ 17 ], in assessing tumor grade [ 18 ], and predicting molecular signatures [ 19 ]. Most of the developed CAD systems focus on identifying tiles in whole-slide images.…”