Title | ||
---|---|---|
SimSearch: A Human-in-The-Loop Learning Framework for Fast Detection of Regions of Interest in Microscopy Images |
Abstract | ||
---|---|---|
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Objective:</i>
Large-scale microscopy-based experiments often result in images with rich but sparse information content. An experienced microscopist can visually identify regions of interest (ROIs), but this becomes a cumbersome task with large datasets. Here we present SimSearch, a framework for quick and easy user-guided training of a deep neural model aimed at fast detection of ROIs in large-scale microscopy experiments.
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Methods:</i>
The user manually selects a small number of patches representing different classes of ROIs. This is followed by feature extraction using a pre-trained deep-learning model, and interactive patch selection pruning, resulting in a smaller set of
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">clean</i>
(user approved) and larger set of
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">noisy</i>
(unapproved) training patches of ROIs and background. The pre-trained deep-learning model is thereafter first trained on the large set of
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">noisy</i>
patches, followed by refined training using the
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">clean</i>
patches.
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Results:</i>
The framework is evaluated on fluorescence microscopy images from a large-scale drug screening experiment, brightfield images of immunohistochemistry-stained patient tissue samples, and malaria-infected human blood smears, as well as transmission electron microscopy images of cell sections. Compared to state-of-the-art and manual/visual assessment, the results show similar performance with maximal flexibility and minimal a priori information and user interaction.
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Conclusions:</i>
SimSearch quickly adapts to different data sets, which demonstrates the potential to speed up many microscopy-based experiments based on a small amount of user interaction.
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Significance:</i>
SimSearch can help biologists quickly extract informative regions and perform analyses on large datasets helping increase the throughput in a microscopy experiment. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1109/JBHI.2022.3177602 | IEEE Journal of Biomedical and Health Informatics |
Keywords | DocType | Volume |
Humans,Image Processing, Computer-Assisted,Microscopy, Fluorescence | Journal | 26 |
Issue | ISSN | Citations |
8 | 2168-2194 | 0 |
PageRank | References | Authors |
0.34 | 5 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ankit Gupta | 1 | 0 | 0.34 |
Alan Sabirsh | 2 | 0 | 0.34 |
Carolina Wählby | 3 | 27 | 6.30 |
Ida-Maria Sintorn | 4 | 114 | 13.85 |