Statistical Karyotype Analysis Using CNN and Geometric Optimization

Abstract

Karyotype analysis is one of the main techniques of cytogenetics through medical image processing, which has an important role in modern medical treatment and diagnosis. The process of human karyotype analysis contains two key components: Firstly, chromosomes are segmented from metaphase chromosome digital images taken under a microscope. Chromosomes then are analyzed, compared, ordered and classified one by one carefully. Under this procedure, the operation on segmentation and classification is cumbersomely time-consuming, where traditional geometric or statistical methods only have limited effect due to low accuracy. Thus, in most conditions, human effort is still heavily required to monitor the workflow and correct the errors. In this paper, we present an integrated workflow to segment out and classify chromosomes automatically using a combination of multiple input convolutional neural networks (CNN) and geometric optimization, called mCNN_GO. We investigate Mask R-CNN to segment out chromosomes from metaphase chromosome images and train the mCNN_GO to classify the sub-images. To improve the performance of the segmentation network, we adapt a new feature-based approach to synthesize images on the labeled data. Furthermore, we develop a geometric algorithm to straighten the chromosomes before classification to ensure the consistency of the data. Experimental results demonstrate that our approach significantly outperforms the state-of-the-art methods on automatic karyotype analysis. INDEX TERMS Biomedical image processing, image processing, machine learning.

Automated System for Chromosome Karyotyping to Recognize the Most Common Numerical Abnormalities Using Deep Learning

“…3 and as we notice we achieved better performance than [8] and [34]. Papers [4], [33], and [43] used different deep learning networks for feature extraction and classification on their own collected dataset whereas paper [39], [41], and [42] built their own CNN and used it for both feature extraction and classification on their own collected dataset. As obvious from Table 12, our classification method on the CEGMR dataset using scheme no.…”

“…The comparison is relative because the used techniques and datasets are different. We notice that three papers [22], [32], and [33] used U-Net semantic segmentation and paper [43] used another type of segmentation which is instance segmentation (Mask R-CNN).…”

“…Without preprocessing steps, they obtained 68.5% classification accuracy and by preprocessing steps, they improved the accuracy to 86.7%. Dataset used by Ning et al [43] for segmentation contains original and synthetic images. They improved the 'cut and paste' image synthesis method to synthesize the annotated images.…”

Intelligent Question Answering System by Deep Convolutional Neural Network in Finance and Economics Teaching

“…e most common way for students to acquire knowledge is to use search engines. However, traditional commercial search engines are not completely developed for students, and there are many drawbacks [12]. 1. e Internet search engines contain a lot of relevant information.…”

A Robust Framework of Chromosome Straightening with ViT-Patch GAN

“…(i) Printed images of bent chromosome are physically cut and rearranged for straightening [4], [5]. (ii) Geometric algorithms are extensively designed based on chromosome micrographs for automatic straightening, which consist of two main categories: extraction of the medial axis [6]- [9] and finding bending points [10]- [12]. However, these geometric methods may extract inaccurate medial axes when chromosomes have large widths.…”