Analysis of the Human Protein Atlas Image Classification competition.
Wei OuyangCasper F WinsnesMartin HjelmareAnthony J CesnikLovisa ÅkessonHao XuDevin P SullivanShubin DaiJun LanPark JinmoShaikat M GalibChristof HenkelKevin HwangDmytro PoplavskiyBojan TunguzRussel D WolfingerYinzheng GuChuanpeng LiJinbin XieDmitry BuslovSergei FironovAlexander KiselevDmytro PanchenkoXuan CaoRunmin WeiYuanhao WuXun ZhuKuan-Lun TsengZhifeng GaoCheng JuXiaohan YiHongdong ZhengConstantin KappelEmma LundbergPublished in: Nature methods (2019)
Pinpointing subcellular protein localizations from microscopy images is easy to the trained eye, but challenging to automate. Based on the Human Protein Atlas image collection, we held a competition to identify deep learning solutions to solve this task. Challenges included training on highly imbalanced classes and predicting multiple labels per image. Over 3 months, 2,172 teams participated. Despite convergence on popular networks and training techniques, there was considerable variety among the solutions. Participants applied strategies for modifying neural networks and loss functions, augmenting data and using pretrained networks. The winning models far outperformed our previous effort at multi-label classification of protein localization patterns by ~20%. These models can be used as classifiers to annotate new images, feature extractors to measure pattern similarity or pretrained networks for a wide range of biological applications.
Keyphrases
- deep learning
- convolutional neural network
- artificial intelligence
- machine learning
- endothelial cells
- protein protein
- neural network
- binding protein
- high resolution
- single cell
- optical coherence tomography
- induced pluripotent stem cells
- small molecule
- body composition
- virtual reality
- resistance training
- high speed
- pluripotent stem cells
- electronic health record