Histo-Fetch: On-the-Fly Processing of Gigapixel WSIs

In this work, we created a custom pipeline (histo-fetch) to efficiently extract random patches and labels from pathology whole slide images (WSIs) for input to a neural network on-the-fly. We prefetch these patches as needed during network training, avoiding the need for WSI preparation such as chopping/tiling. We demonstrate the utility of this pipeline to perform artificial stain transfer and image generation using the popular networks CycleGAN and ProGAN, respectively.

Histo-Fetch Processing Pipeline
Histo-Fetch Processing Pipeline. (a) The traditional method uses the CPU to chop whole slide images into patches which are saved to disk before convolutional neural network training. These patches are read and fed to the graphics processing unit for training. (b) Histo-fetch randomly selects indices containing tissue on the fly. These are processed on the CPU and supplied to the graphics processing unit. (c) Efficiency comparison of the two approaches using ProGAN, highlighting preprocessing time and additional disk space required using a dataset of 151 human biopsy whole slide images. The average training step time does not significantly change.

Select References

B. Lutnick, L. K. Murali, B. Ginley, A. Z. Rosenberg, and P. Sarder, “Histo-fetch – On-the-fly processing of gigapixel whole slide images simplifies and speeds neural network training,” Journal of Pathology Informatics – Elsevier, vol. 13, no. 7, pp. 1-5, Jan. 2022.

Underline indicates corresponding author.
* indicates equal contribution.
indicates Dr. Sarder’s faculty trainees.
§ indicates post doctoral associates.
indicates Dr. Sarder’s graduate students.
indicates Dr. Sarder’s undergraduate students.