An Overlapping Cell Image Synthesis Method for Imbalance Data
DNA ploidy analysis of cells is an automation technique applied in pathological diagnosis. It is important for this technique to classify various nuclei images accurately. However, the lack of overlapping nuclei images in training data (imbalanced training data) results in low recognition rates of o...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2018-01-01
|
| Series: | Analytical Cellular Pathology |
| Online Access: | http://dx.doi.org/10.1155/2018/7919503 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832560796859629568 |
|---|---|
| author | Yi Ning Xie Lian Yu Guo Hui Guan Yong Jun He |
| author_facet | Yi Ning Xie Lian Yu Guo Hui Guan Yong Jun He |
| author_sort | Yi Ning Xie |
| collection | DOAJ |
| description | DNA ploidy analysis of cells is an automation technique applied in pathological diagnosis. It is important for this technique to classify various nuclei images accurately. However, the lack of overlapping nuclei images in training data (imbalanced training data) results in low recognition rates of overlapping nuclei images. To solve this problem, a new method which synthesizes overlapping nuclei images with single-nuclei images is proposed. Firstly, sample selection is employed to make the synthesized samples representative. Secondly, random functions are used to control the rotation angles of the nucleus and the distance between the centroids of the nucleus, increasing the sample diversity. Then, the Lambert-Beer law is applied to reassign the pixels of overlapping parts, thus making the synthesized samples quite close to the real ones. Finally, all synthesized samples are added to the training sets for classifier training. The experimental results show that images synthesized by this method can solve the data set imbalance problem and improve the recognition rate of DNA ploidy analysis systems. |
| format | Article |
| id | doaj-art-304d0d681fc641ee85d13a21b21babe2 |
| institution | Kabale University |
| issn | 2210-7177 2210-7185 |
| language | English |
| publishDate | 2018-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Analytical Cellular Pathology |
| spelling | doaj-art-304d0d681fc641ee85d13a21b21babe22025-02-03T01:26:43ZengWileyAnalytical Cellular Pathology2210-71772210-71852018-01-01201810.1155/2018/79195037919503An Overlapping Cell Image Synthesis Method for Imbalance DataYi Ning Xie0Lian Yu1Guo Hui Guan2Yong Jun He3Harbin University of Science and Technology, Harbin 150080, ChinaHarbin University of Science and Technology, Harbin 150080, ChinaHarbin Dongan Mitsubishi Automotive Engine Manufacturing Company, Harbin 150060, ChinaHarbin University of Science and Technology, Harbin 150080, ChinaDNA ploidy analysis of cells is an automation technique applied in pathological diagnosis. It is important for this technique to classify various nuclei images accurately. However, the lack of overlapping nuclei images in training data (imbalanced training data) results in low recognition rates of overlapping nuclei images. To solve this problem, a new method which synthesizes overlapping nuclei images with single-nuclei images is proposed. Firstly, sample selection is employed to make the synthesized samples representative. Secondly, random functions are used to control the rotation angles of the nucleus and the distance between the centroids of the nucleus, increasing the sample diversity. Then, the Lambert-Beer law is applied to reassign the pixels of overlapping parts, thus making the synthesized samples quite close to the real ones. Finally, all synthesized samples are added to the training sets for classifier training. The experimental results show that images synthesized by this method can solve the data set imbalance problem and improve the recognition rate of DNA ploidy analysis systems.http://dx.doi.org/10.1155/2018/7919503 |
| spellingShingle | Yi Ning Xie Lian Yu Guo Hui Guan Yong Jun He An Overlapping Cell Image Synthesis Method for Imbalance Data Analytical Cellular Pathology |
| title | An Overlapping Cell Image Synthesis Method for Imbalance Data |
| title_full | An Overlapping Cell Image Synthesis Method for Imbalance Data |
| title_fullStr | An Overlapping Cell Image Synthesis Method for Imbalance Data |
| title_full_unstemmed | An Overlapping Cell Image Synthesis Method for Imbalance Data |
| title_short | An Overlapping Cell Image Synthesis Method for Imbalance Data |
| title_sort | overlapping cell image synthesis method for imbalance data |
| url | http://dx.doi.org/10.1155/2018/7919503 |
| work_keys_str_mv | AT yiningxie anoverlappingcellimagesynthesismethodforimbalancedata AT lianyu anoverlappingcellimagesynthesismethodforimbalancedata AT guohuiguan anoverlappingcellimagesynthesismethodforimbalancedata AT yongjunhe anoverlappingcellimagesynthesismethodforimbalancedata AT yiningxie overlappingcellimagesynthesismethodforimbalancedata AT lianyu overlappingcellimagesynthesismethodforimbalancedata AT guohuiguan overlappingcellimagesynthesismethodforimbalancedata AT yongjunhe overlappingcellimagesynthesismethodforimbalancedata |