Complex ‘Big Data’ questions that involve machine learning require large datasets for training. This is particularlyproblematic for Deep Learning methods in the biomedical imaging domain and specifically Digital Pathology. TransferLearning has been shown to be a promising method for training classifiers on smaller sized datasets. In this work weinvestigate the effectiveness of aggregated Transfer Learning using VGG19 trained on ImageNet and then fine-tuningparameters with tissue histopathological patches from breast cancer metastatic tissue patches to classify soft tissuesarcoma patches. We compare results with and without transfer learning, and fine tuning applied to different layers.From the results, it is apparent that fine-tuning earlier VGG19 convolutional blocks with breast cancer patches andapplying bottleneck feature extraction to soft tissue sarcoma can have an adverse effect on accuracy and otherperformance measures. Nevertheless, the aggregated approach is a promising method for digital pathology and requiresmuch more investigation.
展开▼
