CytoImageNet: A large-scale pretraining dataset for bioimage transfer learning
This addresses the problem of efficient bioimage analysis for researchers, but it is incremental as it adapts an existing paradigm (ImageNet) to a specific domain.
The authors tackled the need for automated analysis of high-throughput microscopy images by curating CytoImageNet, a large-scale dataset of 890K weakly-labeled images across 894 classes, and showed that pretraining on it yields features competitive with ImageNet for downstream microscopy tasks, capturing additional information.
Motivation: In recent years, image-based biological assays have steadily become high-throughput, sparking a need for fast automated methods to extract biologically-meaningful information from hundreds of thousands of images. Taking inspiration from the success of ImageNet, we curate CytoImageNet, a large-scale dataset of openly-sourced and weakly-labeled microscopy images (890K images, 894 classes). Pretraining on CytoImageNet yields features that are competitive to ImageNet features on downstream microscopy classification tasks. We show evidence that CytoImageNet features capture information not available in ImageNet-trained features. The dataset is made available at https://www.kaggle.com/stanleyhua/cytoimagenet.