Quantitative analysis of metastatic breast cancer in mice using deep learning on cryo-image data
Abstract Cryo-imaging sections and images a whole mouse and provides ~ 120-GBytes of microscopic 3D color anatomy and fluorescence images, making fully manual analysis of metastases an onerous task. A convolutional neural network (CNN)-based metastases segmentation algorithm included three steps: ca...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2021-09-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-021-96838-y |
| _version_ | 1818422120308277248 |
|---|---|
| author | Yiqiao Liu Madhusudhana Gargesha Mohammed Qutaish Zhuxian Zhou Peter Qiao Zheng-Rong Lu David L. Wilson |
| author_facet | Yiqiao Liu Madhusudhana Gargesha Mohammed Qutaish Zhuxian Zhou Peter Qiao Zheng-Rong Lu David L. Wilson |
| author_sort | Yiqiao Liu |
| collection | DOAJ |
| description | Abstract Cryo-imaging sections and images a whole mouse and provides ~ 120-GBytes of microscopic 3D color anatomy and fluorescence images, making fully manual analysis of metastases an onerous task. A convolutional neural network (CNN)-based metastases segmentation algorithm included three steps: candidate segmentation, candidate classification, and semi-automatic correction of the classification result. The candidate segmentation generated > 5000 candidates in each of the breast cancer-bearing mice. Random forest classifier with multi-scale CNN features and hand-crafted intensity and morphology features achieved 0.8645 ± 0.0858, 0.9738 ± 0.0074, and 0.9709 ± 0.0182 sensitivity, specificity, and area under the curve (AUC) of the receiver operating characteristic (ROC), with fourfold cross validation. Classification results guided manual correction by an expert with our in-house MATLAB software. Finally, 225, 148, 165, and 344 metastases were identified in the four cancer mice. With CNN-based segmentation, the human intervention time was reduced from > 12 to ~ 2 h. We demonstrated that 4T1 breast cancer metastases spread to the lung, liver, bone, and brain. Assessing the size and distribution of metastases proves the usefulness and robustness of cryo-imaging and our software for evaluating new cancer imaging and therapeutics technologies. Application of the method with only minor modification to a pancreatic metastatic cancer model demonstrated generalizability to other tumor models. |
| first_indexed | 2024-12-14T13:21:12Z |
| format | Article |
| id | doaj.art-53b589638ab74fa5a3ffbad5db734d76 |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-12-14T13:21:12Z |
| publishDate | 2021-09-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-53b589638ab74fa5a3ffbad5db734d762022-12-21T22:59:56ZengNature PortfolioScientific Reports2045-23222021-09-0111111510.1038/s41598-021-96838-yQuantitative analysis of metastatic breast cancer in mice using deep learning on cryo-image dataYiqiao Liu0Madhusudhana Gargesha1Mohammed Qutaish2Zhuxian Zhou3Peter Qiao4Zheng-Rong Lu5David L. Wilson6Department of Biomedical Engineering, Case Western Reserve UniversityBioInVision IncDepartment of Biomedical Engineering, Case Western Reserve UniversityDepartment of Biomedical Engineering, Case Western Reserve UniversityDepartment of Biomedical Engineering, Case Western Reserve UniversityDepartment of Biomedical Engineering, Case Western Reserve UniversityDepartment of Biomedical Engineering, Case Western Reserve UniversityAbstract Cryo-imaging sections and images a whole mouse and provides ~ 120-GBytes of microscopic 3D color anatomy and fluorescence images, making fully manual analysis of metastases an onerous task. A convolutional neural network (CNN)-based metastases segmentation algorithm included three steps: candidate segmentation, candidate classification, and semi-automatic correction of the classification result. The candidate segmentation generated > 5000 candidates in each of the breast cancer-bearing mice. Random forest classifier with multi-scale CNN features and hand-crafted intensity and morphology features achieved 0.8645 ± 0.0858, 0.9738 ± 0.0074, and 0.9709 ± 0.0182 sensitivity, specificity, and area under the curve (AUC) of the receiver operating characteristic (ROC), with fourfold cross validation. Classification results guided manual correction by an expert with our in-house MATLAB software. Finally, 225, 148, 165, and 344 metastases were identified in the four cancer mice. With CNN-based segmentation, the human intervention time was reduced from > 12 to ~ 2 h. We demonstrated that 4T1 breast cancer metastases spread to the lung, liver, bone, and brain. Assessing the size and distribution of metastases proves the usefulness and robustness of cryo-imaging and our software for evaluating new cancer imaging and therapeutics technologies. Application of the method with only minor modification to a pancreatic metastatic cancer model demonstrated generalizability to other tumor models.https://doi.org/10.1038/s41598-021-96838-y |
| spellingShingle | Yiqiao Liu Madhusudhana Gargesha Mohammed Qutaish Zhuxian Zhou Peter Qiao Zheng-Rong Lu David L. Wilson Quantitative analysis of metastatic breast cancer in mice using deep learning on cryo-image data Scientific Reports |
| title | Quantitative analysis of metastatic breast cancer in mice using deep learning on cryo-image data |
| title_full | Quantitative analysis of metastatic breast cancer in mice using deep learning on cryo-image data |
| title_fullStr | Quantitative analysis of metastatic breast cancer in mice using deep learning on cryo-image data |
| title_full_unstemmed | Quantitative analysis of metastatic breast cancer in mice using deep learning on cryo-image data |
| title_short | Quantitative analysis of metastatic breast cancer in mice using deep learning on cryo-image data |
| title_sort | quantitative analysis of metastatic breast cancer in mice using deep learning on cryo image data |
| url | https://doi.org/10.1038/s41598-021-96838-y |
| work_keys_str_mv | AT yiqiaoliu quantitativeanalysisofmetastaticbreastcancerinmiceusingdeeplearningoncryoimagedata AT madhusudhanagargesha quantitativeanalysisofmetastaticbreastcancerinmiceusingdeeplearningoncryoimagedata AT mohammedqutaish quantitativeanalysisofmetastaticbreastcancerinmiceusingdeeplearningoncryoimagedata AT zhuxianzhou quantitativeanalysisofmetastaticbreastcancerinmiceusingdeeplearningoncryoimagedata AT peterqiao quantitativeanalysisofmetastaticbreastcancerinmiceusingdeeplearningoncryoimagedata AT zhengronglu quantitativeanalysisofmetastaticbreastcancerinmiceusingdeeplearningoncryoimagedata AT davidlwilson quantitativeanalysisofmetastaticbreastcancerinmiceusingdeeplearningoncryoimagedata |