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: 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.

Optimal slice thickness for improved accuracy of quantitative analysis of fluorescent cell and microsphere distribution in cryo-images

“…In cryo-imaging, the tissue sample is alternatively sliced and imaged with a fixed section thickness 1 , 2 , 4 , 10 , 15 , 18 . Diagrams in Fig.…”

Preparing Data for Artificial Intelligence in Pathology with Clinical-Grade Performance

“…For example, the current shortage of pathologists in China is up to 100,000; in addition, the number of pathologists in the United States also decreased by 17.53% from 2007 to 2017 [5]. In the past few years, a large number of AIP systems have emerged, focusing on tasks such as classification, grading, outcome prediction, prognosis determination [6,7] and the diagnosis of various cancers such as gastric cancer [8,9], prostate cancer [10][11][12][13], bowel cancer [14], breast cancer [15][16][17][18][19], and cervical cancer [20,21] among others. AIP predominantly relies on deep learning, utilizing datasets consisting of hundreds to tens of thousands of WSIs for training and testing.…”

Deep learning multi-organ segmentation for whole mouse cryo-images including a comparison of 2D and 3D deep networks

“…High-resolution cryo-imaging of a whole-mouse can produce as much as 120 GB of data, which makes manual analysis a daunting task. We previously created a method for automatic segmentation of fluorescent protein-labeled metastases 7 and fluorescent-labeled stem cells 5 , which enables the quantification of cells labeled with dyes, quantum dots, and fluorescent proteins. Automatic organ segmentation is required to enable further analysis and quantification of organ distributions in such applications.…”