Deeply-supervised density regression for automatic cell counting in microscopy images

Abstract

Accurately counting the number of cells in microscopy images is required in many medical diagnosis and biological studies. This task is tedious, timeconsuming, and prone to subjective errors. However, designing automatic counting methods remains challenging due to low image contrast, complex background, large variance in cell shapes and counts, and significant cell occlusions in two-dimensional microscopy images. In this study, we proposed a new density regression-based method for automatically counting cells in microscopy images. The proposed method processes two innovations compared to other state-ofthe-art density regression-based methods. First, the density regression model (DRM) is designed as a concatenated fully convolutional regression network (C-FCRN) to employ multi-scale image features for the estimation of cell density maps from given images. Second, auxiliary convolutional neural networks (AuxCNNs) are employed to assist in the training of intermediate layers of the designed C-FCRN to improve the DRM performance on unseen datasets. Experimental studies evaluated on four datasets demonstrate the superior performance of the proposed method.

Automatic cell counting from stimulated Raman imaging using deep learning

“…In another research, a FCN-based framework was proposed, which consists of a primary FCN and a set of auxiliary FCNs that provide extra learning features from intermediate layers for the primary FCN. In addition, shortcut connections were integrated into the primary FCN, which can enhance the granularity of the features and density map estimation [1,55]. Morelli et al (2021) follow the similar concept that employs the FCN with short connections between convolution blocks to segment cells [56].…”

Automatic cell counting for phase‐contrast microscopic images based on a combination of Otsu and watershed segmentation method

“…In the study of the antitoxic drugs which inoculated on virus-infected cells, whether the selected drug is lethal to the virus can be determined by counting the number of living cells and infected cells; in this way, the effective treatment drugs can be finally screened. And the cell counting researches also contribute to different tumor types (Coates et al, 2015), understand the cellular and molecular genetic mechanisms (Solnica-Krezel, 2005), and provide helpful information to many other applications (Forero, Kato, & Hidalgo, 2012;He, Minn, Solnica-Krezel, Anastasioe, & Lie, 2021;Vayrynen et al, 2012).…”

Object detection for automatic cancer cell counting in zebrafish xenografts

“…However, the cell-counting process is a bottleneck. Due to the cells’ morphology, simple pre-processing techniques do not allow for an accurate counting of the cells; instead, the process requires manual counting, which is a laborious, time-consuming, prone to human error [ 8 , 9 ]. Furthermore, manual cell counting is a subjective process due to its considerable inter-observer and intra-observer variability [ 10 ].…”

“…With the advent of deep learning (DL) and convolutional neural networks (CNNs), several authors proposed various techniques that surpassed most traditional approaches. Automatic cell counting can be divided into two major categories: detection-based counting, which demands prior detection or segmentation, and another method based on density estimation or regression without the need for prior labeling [ 8 , 11 ].…”

“…Another well-explored category is cell counting using regression networks [ 8 , 11 , 18 , 19 ], in which the global cell count is taken as an annotation to supervise training, instead of the counting following a classification or detection framework. In this case, there is no need for prior object detection or segmentation as the ground truth.…”