Mimicry Embedding Facilitates Advanced Neural Network Training for Image-Based Pathogen Detection

Abstract

The use of deep neural networks (DNNs) for analysis of complex biomedical images shows great promise but is hampered by a lack of large verified data sets for rapid network evolution. Here, we present a novel strategy, termed “mimicry embedding,” for rapid application of neural network architecture-based analysis of pathogen imaging data sets. Embedding of a novel host-pathogen data set, such that it mimics a verified data set, enables efficient deep learning using high expressive capacity architectures and seamless architecture switching. We applied this strategy across various microbiological phenotypes, from superresolved viruses to in vitro and in vivo parasitic infections. We demonstrate that mimicry embedding enables efficient and accurate analysis of two- and three-dimensional microscopy data sets. The results suggest that transfer learning from pretrained network data may be a powerful general strategy for analysis of heterogeneous pathogen fluorescence imaging data sets. IMPORTANCE In biology, the use of deep neural networks (DNNs) for analysis of pathogen infection is hampered by a lack of large verified data sets needed for rapid network evolution. Artificial neural networks detect handwritten digits with high precision thanks to large data sets, such as MNIST, that allow nearly unlimited training. Here, we developed a novel strategy we call mimicry embedding, which allows artificial intelligence (AI)-based analysis of variable pathogen-host data sets. We show that deep learning can be used to detect and classify single pathogens based on small differences.

Machine Learning and Artificial Intelligence for the Prediction of Host–Pathogen Interactions: A Viral Case

“…In case of image-based data, the advent of resources like Bio-Image Archive, 75 data-dedicated journals and social coding platforms fostered deposition of specialized image data sets, including those focused on HPI. 46 , 76 The influence of CORD-19 65 on the adoption of NLP techniques in HPI research was unequivocally self-evident. Perhaps, it is the access to open research data sets akin to CORD-19 together with image-based data repositories that will become the missing piece for AI and ML to become a viral case in HPI research.…”

A weak-labelling and deep learning approach for in-focus object segmentation in 3D widefield microscopy

“…To develop an approach for in-focus region detection, we have employed a published dataset of Danio rerio (zebra sh) in vivo wide eld microscopy 13 . In this dataset, the fraction of the head of the zebra sh is located in the middle of the eld of view.…”

“…The program code used in this work is available for use and re-use under an open-source license and can be accessed via GitHub (https://github.com/casus/deepfocus). The Dataset of D. Rerio focal stacks was previously published in 13 and raw or additional data is available upon request from the corresponding authors 13 . As stated in 13 , the animal experiments were performed according to the Animals (Scienti c Procedures) Act of 1986 and approved by the Home O ce (project licenses PPL P84A89400 and P4E664E3C).…”

“…The wide eld microscopy image dataset we employed contains in vivo transmission light focal stack micrographs of Danio rerio (zebra sh) larva's head 13,14 . To obtain the GT we investigated 8 algorithms commonly used in autofocusing tasks including Brenner 15 , Variance 16 , Tenengrad 5 , etc.…”

Labels in a haystack: Approaches beyond supervised learning in biomedical applications

“…Another group 48 also used ImageNet pre-trained ResNet-50 for a classification task in micrographs of virus-infected cells. While other examples 49 include transfer learning from models trained on, for example, an MNIST dataset, 105 overall transfer learning from ImageNet-pre-trained models remains the most-favored approach. Proteomics and genomics benefit from transfer learning as well.…”