End-to-end Multiple Instance Learning for Whole-Slide Cytopathology of Urothelial Carcinoma

Joshua Butke, Tatjana Frick, Florian Roghmann, Samir El-Mashtoly, Klaus Gerwert, Axel Mosig

Keywords: multiple instance learning, urine sediment, cytopathology

TL;DR: Using a multiple instance learning approach that combines attention and hard negative mining to identifiy (cancerous) cells and facilitate reliable assessment of the urothelial cancer status at the level of large whole-slide samples of urine sediment

Abstract:

As a non-invasive approach, cytopathology of urine sediment is a highly promising approach to diagnosing urothelial carcinoma. However, computational assessment of the cytopathological status of a sample raises the challenge of identifying few cancerous cells among thousands of cells in a microscopic whole-slide image.
To address this challenge, we propose an end-to-end trainable multiple instance learning approach that combines the attention mechanism and hard negative mining to classify hematoxylin and eosin stained patient-level whole-slide images of urine sediment cells.
The singular cells are extracted by a simple foreground detection algorithm. With feature embeddings computed for each image patch in a bag by a convolutional neural network, the attention mechanism serves as the pooling operator, enabling a bag-level prediction while still giving an interpretable score for each image patch.
This enables the identification of key instances and potential regions of interest that trigger a patient-level decision. Our results show that the proposed system can differentiate between normal and cancerous urothelial cells, thus enabling the non-invasive diagnosis of urothelial carcinoma in patients using urine sediment analysis.