| BAI Hua,LU Changhao,MA Ming,YAN Shulin,ZHANG Jianzhong,HAN Zhibo.An improved U-Net for cell confluence estimation[J].Optoelectronics Letters,2022,(6):378-384 |
| An improved U-Net for cell confluence estimation |
| Author Name | Affiliation | | BAI Hua | Tianjin Key Laboratory of Optoelectronic Detection Technology and Systems, School of Electronic and Information Engineering, Tiangong University, Tianjin 300387, China | | LU Changhao | Tianjin Key Laboratory of Optoelectronic Detection Technology and Systems, School of Electronic and Information Engineering, Tiangong University, Tianjin 300387, China | | MA Ming | Tianjin Key Laboratory of Optoelectronic Detection Technology and Systems, School of Electronic and Information Engineering, Tiangong University, Tianjin 300387, China | | YAN Shulin | Tianjin Key Laboratory of Engineering Technologies for Cell Phamaceutical, Tianjin 300457, China
National Engineering Research Center of Cell Products/AmCellGene Co., Ltd., Tianjin 300457, China | | ZHANG Jianzhong | Tianjin Key Laboratory of Engineering Technologies for Cell Phamaceutical, Tianjin 300457, China
National Engineering Research Center of Cell Products/AmCellGene Co., Ltd., Tianjin 300457, China | | HAN Zhibo | Tianjin Key Laboratory of Engineering Technologies for Cell Phamaceutical, Tianjin 300457, China
National Engineering Research Center of Cell Products/AmCellGene Co., Ltd., Tianjin 300457, China
State Key Lab of Experimental Hematology, Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin 300020, China |
|
| Abstract: |
| Cell confluence is an important metric to determine the growth and the best harvest time of adherent cells. At present, the evaluation of cell confluence mainly relies on experienced labor, and thus it is not conducive to the automated cell culture. In this paper, we proposed an improved U-Net algorithm (called DU-Net) for the segmentation of adherent cells. First, the general convolution was replaced by the dilated convolution to expand the receptive fields for feature extraction. Then, the convolutional layers were combined with the batch normalization layers to reduce the dependence of the network on initialization. As a result, the segmentation accuracy and F1-score of the proposed DU-Net for adherent cells with low confluence (<50%) reached 96.94% and 93.87%, respectively, and for those with high confluence (≥50%), they reached 98.63% and 98.98%, respectively. Further, the paired t-test results showed that the proposed DU-Net was statistically superior to the traditional U-Net algorithm. |
| Hits: 311 |
| Download times: 1 |
|
| View Full Text Download reader |
|
|
|
