Neural Image Compression

Overview of the neural image compression pipeline.
Overview of the neural image compression pipeline

View on Github

We propose Neural Image Compression (NIC), a two-step method to build convolutional neural networks (CNNs) for gigapixel image analysis solely using weak image-level labels. First, gigapixel images are compressed using a neural network trained in an unsupervised fashion, retaining high-level information while suppressing pixel-level noise. Second, a CNN is trained on these compressed image representations to predict image-level labels, avoiding the need for fine-grained manual annotations. A full description of our method can be found in the journal paper Neural Image Compression for Gigapixel Histopathology Image Analysis and Arxiv link.

People

David Tellez

David Tellez

PhD Candidate

Geert Litjens

Geert Litjens

Assistant Professor

Jeroen van der Laak

Jeroen van der Laak

Associate Professor

Francesco Ciompi

Francesco Ciompi

Assistant Professor

Publications

  • D. Tellez, G. Litjens, J. van der Laak and F. Ciompi, "Neural Image Compression for Gigapixel Histopathology Image Analysis", IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019;58:101544. Abstract/PDF DOI PMID Cited by ~7