Automatic Detection Method for Cancer Cell Nucleus Image Based on Deep-Learning Analysis and Color Layer Signature Analysis Algorithm

Abstract

Exploring strategies to treat cancer has always been an aim of medical researchers. One of the available strategies is to use targeted therapy drugs to make the chromosomes in cancer cells unstable such that cell death can be induced, and the elimination of highly proliferative cancer cells can be achieved. Studies have reported that the mitotic defects and micronuclei in cancer cells can be used as biomarkers to evaluate the instability of the chromosomes. Researchers use these two biomarkers to assess the effects of drugs on eliminating cancer cells. However, manual work is required to count the number of cells exhibiting mitotic defects and micronuclei either directly from the viewing window of a microscope or from an image, which is tedious and creates errors. Therefore, this study aims to detect cells with mitotic defects and micronuclei by applying an approach that can automatically count the targets. This approach integrates the application of a convolutional neural network for normal cell identification and the proposed color layer signature analysis (CLSA) to spot cells with mitotic defects and micronuclei. This approach provides a method for researchers to detect colon cancer cells in an accurate and time-efficient manner, thereby decreasing errors and the processing time. The following sections will illustrate the methodology and workflow design of this study, as well as explain the practicality of the experimental comparisons and the results that were used to validate the practicality of this algorithm.

Organ Contouring for Lung Cancer Patients with a Seed Generation Scheme and Random Walks

“…The most frequent way to undertake a comparison of performance is by using the dataset in-house and the gold standards, made manually by physicians in-house [26]. A similar situation was described in the author's research [27]. These authors had a dataset in-house (cancer cell nucleus images), and they compared these images with manual counting.…”

A deep learning workflow for quantification of micronuclei in DNA damage studies in cultured cancer cell lines: A proof of principle investigation

“…Efficient methods to study DNA damage as well as integration of ML has the potential to yield reliable results and decrease turnaround times in low-resource academic research and hospital settings. Proof of concept genotoxic studies [16][17][18][19][20][21][22][23][24], have tried to capitalize on several statistical models to detect nuclei; these studies have been of limited use due to lack of differentiation between cell types for detections [19]. Studies have attempted to use off-the-shelf ML algorithms and tools like Cell Profiler to perform segmentation of cell nuclei in microscopic cell images [20].…”

Automatic recognition of micronucleus by combining attention mechanism and AlexNet

“…In recent years, computer vision methods have succeeded in medical image analysis. It has advantages such as stability, standardization, long-term operation, and consistency [ 4 ]. Methods for diagnosis using computer vision are generally divided into traditional and deep learning techniques.…”

“…Wu et al [ 12 ] proposed the covid-al framework, which can consider both data diversity and data uncertainty, improving the efficiency of active learning methods. In addition, Work in [ 4 ] fused neural networks and traditional methods, introducing the YOLO algorithm into cell micronucleus image detection, and achieved great performance.

Fig.

…”