Cell Counting by Regression Using Convolutional Neural Network
The ability to accurately quantitate specific populations of cells is important for precision diagnostics in laboratory medicine. For example, the quantization of positive tumor cells can be used clinically to determine the need for chemotherapy in a canc
- PDF / 8,152,067 Bytes
- 17 Pages / 439.37 x 666.142 pts Page_size
- 61 Downloads / 213 Views
2
Department of Computing Science, University of Alberta, Edmonton, Canada [email protected] Department of Laboratory Medicine, University of Alberta, Edmonton, Canada
Abstract. The ability to accurately quantitate specific populations of cells is important for precision diagnostics in laboratory medicine. For example, the quantization of positive tumor cells can be used clinically to determine the need for chemotherapy in a cancer patient. In this paper, we describe a supervised learning framework with Convolutional Neural Network (CNN) and cast the cell counting task as a regression problem, where the global cell count is taken as the annotation to supervise training, instead of following the classification or detection framework. To further decrease the prediction error of counting, we tune several cutting-edge CNN architectures (e.g. Deep Residual Network) into the regression model. As the final output, not only the cell count is estimated for an image, but also its spatial density map is provided. The proposed method is evaluated with three state-of-the-art approaches on three cell image datasets and obtain superior performance. Keywords: Cell counting · Convolution neural network · Deep residual net · Detection · Classification
1 1.1
Introduction Problem Definition
Automatic cell counting is to obtain the number of certain types of cells in a medical image like microscopic images. It is of great interest to a wide range of medical scenarios [2,4,23]. An example is the diagnosis and treatment of breast cancer, which is one of the most common female diseases leading to death worldwide. The number of proliferating (e.g. Ki67 positive) tumor cells is an important index associated with the severity of disease clinically. One available method of quantization involves counting the nuclei of proliferating cells using traditional image analysis techniques on a microscopic image. However, it has been proven to be challenging because of inability to distinguish tumor cells from surrounding normal tissue like vessels, fat and fibrous tissue [11], especially in reality the resolution of input medical image could be very high, at the same time the target cells could easily be extremely dense. Consequently, it is quite difficult to manually count target cells one by one. This is the principal motivation of automatic cell counting. c Springer International Publishing Switzerland 2016 G. Hua and H. J´ egou (Eds.): ECCV 2016 Workshops, Part I, LNCS 9913, pp. 274–290, 2016. DOI: 10.1007/978-3-319-46604-0 20
Cell Counting by Regression Using Convolutional Neural Network
1.2
275
Background
From the perspective of computer vision community, automatic cell counting is a branch task of the object counting problem. Many methods have chosen to fulfill object counting task following the detection pipeline [17,20,21,27,30,32]. In this case, an object detection framework is designed to localize each object (e.g. cell, head or vehicle) one by one, after that a counter naturally takes all the detected objects and produces the final coun
Data Loading...
