Self-Supervised Learning of Phenotypic Representations from Cell Images with Weak LabelsDownload PDF

09 Oct 2022 (modified: 26 Mar 2024)LMRL 2022 PaperReaders: Everyone
Keywords: Deep Learning, Representation Learning, Computer Vision, Self-Supervised Learning, Drug Discovery, Microscopy Imaging
TL;DR: WS-DINO: a novel framework to use weak label information in a self-supervised setting to learn phenotypic representations from high-content fluorescent images of cells.
Abstract: We propose WS-DINO as a novel framework to use weak label information in learning phenotypic representations from high-content fluorescent images of cells. Our model is based on a knowledge distillation approach with a vision transformer backbone (DINO), and we use this as a benchmark model for our study. Using WS-DINO, we fine-tuned with weak label information available in high-content microscopy screens (treatment and compound) and achieve state-of-the-art performance in not-same-compound mechanism of action prediction on the BBBC021 dataset (98%), and not-same-compound-and-batch performance (96%) using the compound as the weak label. Our method bypasses single cell cropping as a pre-processing step, and using self-attention maps we show that the model learns structurally meaningful phenotypic profiles.
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 1 code implementation](https://www.catalyzex.com/paper/arxiv:2209.07819/code)
0 Replies

Loading