Chromosome detection in metaphase cell images using morphological priors
Reliable chromosome detection in metaphase cell (MC) images can greatly alleviate the workload of cytogeneticists for karyotype analysis and the diagnosis of chromosomal disorders. However, it is still an extremely challenging task due to the complicated characteristics of chromosomes, e.g., dense d...
Saved in:
Main Authors: | , , , , , , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | English |
Published: |
2023
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/171790 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-171790 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1717902023-11-08T02:38:04Z Chromosome detection in metaphase cell images using morphological priors Wang, Jun Zhou, Chengfeng Chen, Songchang Hu, Jianwu Wu, Minghui Jiang, Xudong Xu, Chenming Qian, Dahong School of Electrical and Electronic Engineering Engineering::Electrical and electronic engineering Chromosome Rotated Object Detection Reliable chromosome detection in metaphase cell (MC) images can greatly alleviate the workload of cytogeneticists for karyotype analysis and the diagnosis of chromosomal disorders. However, it is still an extremely challenging task due to the complicated characteristics of chromosomes, e.g., dense distributions, arbitrary orientations, and various morphologies. In this article, we propose a novel rotated-anchor-based detection framework, named DeepCHM, for fast and accurate chromosome detection in MC images. Our framework has three main innovations: 1) A deep saliency map representing chromosomal morphological features is learned end-to-end with semantic features. This not only enhances the feature representations for anchor classification and regression but also guides the anchor setting to significantly reduce redundant anchors. This accelerates the detection and improves the performance; 2) A hardness-aware loss weights the contribution of positive anchors, which effectively reinforces the model to identify hard chromosomes; 3) A model-driven sampling strategy addresses the anchor imbalance issue by adaptively selecting hard negative anchors for model training. In addition, a large-scale benchmark dataset with a total of 624 images and 27,763 chromosome instances was built for chromosome detection and segmentation. Extensive experimental results demonstrate that our method outperforms most state-of-the-art (SOTA) approaches and successfully handles chromosome detection, with an AP score of 93.53%. This work was supported in part by the National Natural Science Foundation of China under Grants 81974276 and 62101318 and in part by the Key Research and Development Program of Jiangsu Province under Grant BE2020762. 2023-11-08T02:15:01Z 2023-11-08T02:15:01Z 2023 Journal Article Wang, J., Zhou, C., Chen, S., Hu, J., Wu, M., Jiang, X., Xu, C. & Qian, D. (2023). Chromosome detection in metaphase cell images using morphological priors. IEEE Journal of Biomedical and Health Informatics, 27(9), 4579-4590. https://dx.doi.org/10.1109/JBHI.2023.3286572 2168-2194 https://hdl.handle.net/10356/171790 10.1109/JBHI.2023.3286572 37318973 2-s2.0-85162615753 9 27 4579 4590 en IEEE Journal of Biomedical and Health Informatics © 2023 IEEE. All rights reserved. |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Engineering::Electrical and electronic engineering Chromosome Rotated Object Detection |
spellingShingle |
Engineering::Electrical and electronic engineering Chromosome Rotated Object Detection Wang, Jun Zhou, Chengfeng Chen, Songchang Hu, Jianwu Wu, Minghui Jiang, Xudong Xu, Chenming Qian, Dahong Chromosome detection in metaphase cell images using morphological priors |
description |
Reliable chromosome detection in metaphase cell (MC) images can greatly alleviate the workload of cytogeneticists for karyotype analysis and the diagnosis of chromosomal disorders. However, it is still an extremely challenging task due to the complicated characteristics of chromosomes, e.g., dense distributions, arbitrary orientations, and various morphologies. In this article, we propose a novel rotated-anchor-based detection framework, named DeepCHM, for fast and accurate chromosome detection in MC images. Our framework has three main innovations: 1) A deep saliency map representing chromosomal morphological features is learned end-to-end with semantic features. This not only enhances the feature representations for anchor classification and regression but also guides the anchor setting to significantly reduce redundant anchors. This accelerates the detection and improves the performance; 2) A hardness-aware loss weights the contribution of positive anchors, which effectively reinforces the model to identify hard chromosomes; 3) A model-driven sampling strategy addresses the anchor imbalance issue by adaptively selecting hard negative anchors for model training. In addition, a large-scale benchmark dataset with a total of 624 images and 27,763 chromosome instances was built for chromosome detection and segmentation. Extensive experimental results demonstrate that our method outperforms most state-of-the-art (SOTA) approaches and successfully handles chromosome detection, with an AP score of 93.53%. |
author2 |
School of Electrical and Electronic Engineering |
author_facet |
School of Electrical and Electronic Engineering Wang, Jun Zhou, Chengfeng Chen, Songchang Hu, Jianwu Wu, Minghui Jiang, Xudong Xu, Chenming Qian, Dahong |
format |
Article |
author |
Wang, Jun Zhou, Chengfeng Chen, Songchang Hu, Jianwu Wu, Minghui Jiang, Xudong Xu, Chenming Qian, Dahong |
author_sort |
Wang, Jun |
title |
Chromosome detection in metaphase cell images using morphological priors |
title_short |
Chromosome detection in metaphase cell images using morphological priors |
title_full |
Chromosome detection in metaphase cell images using morphological priors |
title_fullStr |
Chromosome detection in metaphase cell images using morphological priors |
title_full_unstemmed |
Chromosome detection in metaphase cell images using morphological priors |
title_sort |
chromosome detection in metaphase cell images using morphological priors |
publishDate |
2023 |
url |
https://hdl.handle.net/10356/171790 |
_version_ |
1783955518118690816 |