Research progress of micronucleus visualization analysis and artificial intelligence detection strategy.
10.3760/cma.j.cn112150-20210408-00340
- Author:
Bao Min WANG
1
;
Geng HU
1
;
Li Hua HU
2
;
Dan CHEN
1
;
Yu AN
3
;
Cheng LI
4
;
Guang JIA
5
;
Gui Ping HU
4
Author Information
1. School of Biological Science and Medical Engineering, Beihang University, Beijing 100191, China School of Medical Science and Engineering, Beihang University, Beijing 100191, China.
2. Peking University First Hospital, Beijing 100034, China.
3. High-tech Innovation Center of Big Data Precision Medicine, Beihang University, Beijing 100191, China.
4. School of Medical Science and Engineering, Beihang University, Beijing 100191, China High-tech Innovation Center of Big Data Precision Medicine, Beihang University, Beijing 100191, China.
5. School of Public Health, Peking University, Beijing 100191, China.
- Publication Type:Review
- MeSH:
Artificial Intelligence;
Big Data;
Humans;
Machine Learning;
Precision Medicine
- From:
Chinese Journal of Preventive Medicine
2022;56(3):391-396
- CountryChina
- Language:Chinese
-
Abstract:
The micronucleomics test can comprehensively display a variety of harmful endpoints, such as DNA damage and repair, chromosome breakage or loss and cell growth inhibition, with fast, simple and economical feature. Micronucleomics is not only widely used in the comprehensive assessment of the types and modes of genetic action of exogenous chemicals (such as drugs, food additives, cosmetics, environmental pollutants, etc.), but also plays an important role in the screening and risk assessment of cancer population at high risk. However, the traditional micronucleomics image counting method has the characteristics of time-consuming, low accuracy, and high cost, which cannot meet the current analysis requirements of large-scale, multi-index, rapidity, high precision and visualization. In recent years, with the rapid development of the era of precision medicine based on big data, visualized analysis of new micronucleomics based on machine learning and detection strategies based on deep learning have shown a good application prospect. This review, based on the application value of micronucleomics, systematically compares the traditional and new artificial intelligence counting of micronucleus images, and discusses the future direction of micronucleus image detection.
