Evaluating Generic Auto-ML Tools for Computational Pathology

Image analysis tasks in computational pathology are commonly solved using convolutional neural networks (CNNs). The selection of a suitable CNN architecture and hyperparameters is usually done through exploratory iterative optimization, which is computationally expensive and requires substantial manual work. The goal of this article is to evaluate how generic tools for neural network architecture search and hyperparameter optimization perform for common use cases in computational pathology. For this purpose, we evaluated one on-premises and one cloudbased tool for three di erent classi cation tasks for histological images: tissue classi cation, mutation prediction, and grading. We found that the default CNN architectures and parameterizations of the evaluated AutoML tools already yielded classi cation performance on par with the original publications. Hyperparameter optimization for these tasks did not substantially improve performance, despite the additional computational e ort. However, performance varied substantially between classi ers obtained from individual AutoML runs due to non-deterministic e ects. Generic CNN architectures and AutoML tools could thus be a viable alternative to manually optimizing CNN architectures and parametrizations. This would allow developers of software solutions for computational pathology to focus e orts on harder-to-automate tasks such as data curation.