Fine-Tuning and training of densenet for histopathology image representation using TCGA diagnostic slides

TLDR

Pre‑trained deep neural networks provide dominant image representations, and their performance can be improved by fine‑tuning, though training from scratch on pathology data is limited by scarce labels and computational cost. This study introduces KimiaNet, a DenseNet‑based architecture fine‑tuned and trained on histopathology images in various configurations. Using over 240,000 1000×1000 patches generated by a high‑cellularity mosaic from 7,126 TCGA whole‑slide images, KimiaNet was evaluated on TCGA, endometrial, and colorectal datasets for search and classification, alongside several convolutional batch‑normalized ReLU networks. KimiaNet outperformed the original DenseNet and smaller CBR networks as a feature extractor for histopathology image representation.

Abstract

Feature vectors provided by pre-trained deep artificial neural networks have become a dominant source for image representation in recent literature. Their contribution to the performance of image analysis can be improved through fine-tuning. As an ultimate solution, one might even train a deep network from scratch with the domain-relevant images, a highly desirable option which is generally impeded in pathology by lack of labeled images and the computational expense. In this study, we propose a new network, namely KimiaNet, that employs the topology of the DenseNet with four dense blocks, fine-tuned and trained with histopathology images in different configurations. We used more than 240,000 image patches with 1000×1000 pixels acquired at 20× magnification through our proposed "high-cellularity mosaic" approach to enable the usage of weak labels of 7126 whole slide images of formalin-fixed paraffin-embedded human pathology samples publicly available through The Cancer Genome Atlas (TCGA) repository. We tested KimiaNet using three public datasets, namely TCGA, endometrial cancer images, and colorectal cancer images by evaluating the performance of search and classification when corresponding features of different networks are used for image representation. As well, we designed and trained multiple convolutional batch-normalized ReLU (CBR) networks. The results show that KimiaNet provides superior results compared to the original DenseNet and smaller CBR networks when used as feature extractor to represent histopathology images.