Unleashing the potential of digital pathology data by training computer-aided diagnosis models without human annotations

• Published in: NPJ digital medicine Vol. 5; no. 1; p. 102
• Main Authors: Marini, Niccolò, Marchesin, Stefano, Otálora, Sebastian, Wodzinski, Marek, Caputo, Alessandro, van Rijthoven, Mart, Aswolinskiy, Witali, Bokhorst, John-Melle, Podareanu, Damian, Petters, Edyta, Boytcheva, Svetla, Buttafuoco, Genziana, Vatrano, Simona, Fraggetta, Filippo, van der Laak, Jeroen, Agosti, Maristella, Ciompi, Francesco, Silvello, Gianmaria, Muller, Henning, Atzori, Manfredo
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 22-07-2022; Nature Publishing Group; Nature Portfolio
• Subjects: 692/699/67/2321; 692/700/139/422; 692/700/459/1748; Biomedicine; Biotechnology; Cancer; Digital technology; Medical diagnosis; Medical imaging; Medicine; Medicine & Public Health; Neural networks
• ISSN: 2398-6352; 2398-6352
• DOI: 10.1038/s41746-022-00635-4

The digitalization of clinical workflows and the increasing performance of deep learning algorithms are paving the way towards new methods for tackling cancer diagnosis. However, the availability of medical specialists to annotate digitized images and free-text diagnostic reports does not scale with the need for large datasets required to train robust computer-aided diagnosis methods that can target the high variability of clinical cases and data produced. This work proposes and evaluates an approach to eliminate the need for manual annotations to train computer-aided diagnosis tools in digital pathology. The approach includes two components, to automatically extract semantically meaningful concepts from diagnostic reports and use them as weak labels to train convolutional neural networks (CNNs) for histopathology diagnosis. The approach is trained (through 10-fold cross-validation) on 3’769 clinical images and reports, provided by two hospitals and tested on over 11’000 images from private and publicly available datasets. The CNN, trained with automatically generated labels, is compared with the same architecture trained with manual labels. Results show that combining text analysis and end-to-end deep neural networks allows building computer-aided diagnosis tools that reach solid performance (micro-accuracy = 0.908 at image-level) based only on existing clinical data without the need for manual annotations.