2015/11/17 by Oren Kraus, Oren Z. Kraus, Lei Jimmy Ba +3 · 1 citation
Biochemistry, Genetics and Molecular Biology · Computer Science · Mathematics · #AI in cancer detection #Artificial intelligence #Cell Image Analysis Techniques #Computer science #Computer vision #Contextual image classification #Convolutional neural network #Deep learning #Digital Imaging for Blood Diseases #Feature (linguistics) #Image (mathematics) #Machine learning #Microscopy #Pattern recognition (psychology) #Pixel #Pooling #Segmentation #cs.CV #q-bio.SC #stat.ML
paper · pdf · doi:10.1093/bioinformatics/btw252
published as Bioinformatics (2016) 32 (12): i52-i59
arxiv created 2015/11/17 · openalex publication_date 2016/06/11 · arxiv updated 2017/05/25 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
MOTIVATION: High-content screening (HCS) technologies have enabled large scale imaging experiments for studying cell biology and for drug screening. These systems produce hundreds of thousands of microscopy images per day and their utility depends on automated image analysis. Recently, deep learning approaches that learn feature representations directly from pixel intensity values have dominated object recognition challenges. These tasks typically have a single centered object per image and existing models are not directly applicable to microscopy datasets. Here we develop an approach that combines deep convolutional neural networks (CNNs) with multiple instance learning (MIL) in order to classify and segment microscopy images using only whole image level annotations. RESULTS: We introduce a new neural network architecture that uses MIL to simultaneously classify and segment microscopy images with populations of cells. We base our approach on the similarity between the aggregation function used in MIL and pooling layers used in CNNs. To facilitate aggregating across large numbers of instances in CNN feature maps we present the Noisy-AND pooling function, a new MIL operator that is robust to outliers. Combining CNNs with MIL enables training CNNs using whole microscopy images with image level labels. We show that training end-to-end MIL CNNs outperforms several previous methods on both mammalian and yeast datasets without requiring any segmentation steps. AVAILABILITY AND IMPLEMENTATION: Torch7 implementation available upon request. CONTACT: [email protected].