Waibel, D. | Search bioRxiv

bioRxiv2021

SHAPR - An AI approach to predict 3D cell shapes from 2D microscopic images

Reconstruction of shapes and sizes of three-dimensional (3D) objects from two-dimensional (2D) information is an intensely studied subject in computer vision. We here consider the level of single cells and nuclei and present a neural network-based SHApe PRediction autoencoder. For proof-of-concept, SHAPR reconstructs 3D shapes of red blood cells from single view 2D confocal microscopy images more accurately than naive stereological models and significantly increases the feature-based prediction of red blood cell types from F1 = 79.0% to F1 = 87.4%. Applied to 2D images containing spheroidal aggregates of densely grown human induced pluripotent stem cells, we find that SHAPR learns fundamental shape properties of cell nuclei and allows for prediction-based morphometry. Reducing imaging time and data storage, SHAPR will help to optimize and up-scale image-based high-throughput applications for biomedicine.

bioinformatics↗

bioRxiv2020

Annotation-efficient classification combining active learning, pre-training and semi-supervised learning for biomedical images

Deep learning based classification of biomedical images requires manual annotation by experts, which is time-consuming and expensive. Incomplete-supervision approaches including active learning, pre-training and semi-supervised learning address this issue and aim to increase classification performance with a limited number of annotated images. Up to now, these approaches have been mostly benchmarked on natural image datasets, where image complexity and class balance typically differ considerably from biomedical classification tasks. In addition, it is not clear how to combine them to improve classification performance on biomedical image data. We thus performed an extensive grid search combining seven active learning algorithms, three pre-training methods and two training strategies as well as respective baselines (random sampling, random initialization, and supervised learning). For four biomedical datasets, we started training with 1% of labeled data and increased it by 5% iteratively, using 4-fold cross-validation in each cycle. We found that the contribution of pre-training and semi-supervised learning can reach up to 25% macro F1-score in each cycle. In contrast, the state-of-the-art active learning algorithms contribute less than 5% to macro F1-score in each cycle. Based on performance, implementation ease and computation requirements, we recommend the combination of BADGE active learning, ImageNet-weights pre-training, and pseudo-labeling as training strategy, which reached over 90% of fully supervised results with only 25% of annotated data for three out of four datasets. We believe that our study is an important step towards annotation and resource efficient model training for biomedical classification challenges.

bioinformatics↗