CellSegmentation

Logo

MATLAB GUI tool for image segmentation of autophagy events detected with Rosella pH-sensitive biosensor in high-throughput high-content microscopy images.

View the Project on GitHub BeatrizGSC/CellSegmentation

CellSegmentation GUI for MATLAB

Welcome to the instructions to install CellSegmentation GUI for MATLAB

Cite as

Paper coming soon :-)

Preprint available at Research Square.

Beatriz García Santa Cruz, Jan Sölter, Gemma Gomez Giro et al. Generalising from Conventional Pipelines: A Case Study in Deep Learning-Based for High-Throughput Screening, 29 October 2021, PREPRINT (Version 1) available at Research Square. https://doi.org/10.21203/rs.3.rs-991404/v1

Description

The study of complex diseases relies on large amounts of data to build models toward precision medicine. Such data acquisition is feasible in the context of high-throughput screening, in which the quality of the results relies on the accuracy of the image analysis. Although state-of-the-art solutions for image segmentation employ deep learning approaches, the high cost of manually generating ground truth labels for model training hampers the day-to-day application in experimental laboratories.

Alternatively, traditional computer vision-based solutions do not need expensive labels for their implementation. Our work combines both approaches by training a deep learning network using weak training labels automatically generated with conventional computer vision methods. Our network surpasses the conventional segmentation quality by generalising beyond noisy labels, providing a 25% increase of mean intersection over union, and simultaneously reducing the development and inference times.

Our solution was embedded into an easy-to-use graphical user interface that allows researchers to assess the predictions and correct potential inaccuracies with minimal human input. To demonstrate the feasibility of training a deep learning solution on a large dataset of noisy labels automatically generated by a conventional pipeline, we compared our solution against the common approach of training a model from a small manually curated dataset by several experts. Our work suggests that humans perform better in context interpretation, such as error assessment, while computers outperform in pixel-by-pixel fine segmentation. Such pipelines are illustrated with a case study on image segmentation for autophagy events.

This work aims for better translation of new technologies to real-world settings in microscopy-image analysis.

Schema

Panel A: General workflow of the proposed pipeline, part 1: (In blue, CIP-based DL: CDL). [1.1] First a weakly labelled dataset is created using conventional imaging processing (CIP). [1.2] After that, a U-net like architecture is trained and [1.3] the accuracy of the evaluated. Including an integration of the trained network in an intuitive tool for biologists that allows easy correction.

Proposed pipeline, part 2: (In green, for Manually based DL: MDL). [2.1] Manual corrected masks are easily generated using the GUI, which is employed to train a U-net from scratch [2.2].

Schema

Panel B: The biological process of autophagy and its detection with Rosella biosensor. The four main phases are: The initial state - Phagophore, intermediate states~- Autophagosome and early autolysosome, and final state - Autolysosome. The fusion with the lysosome during the autophagy process yields a pH decrease which induces a change of colour in the fluorescent microscopy image.

Samples

Instructions

To install and use the GUI in MATLAB:

Video demo

The following video depicts how to add the algorithm to the image labeler GUI in MATLAB and use it for Cell Segmentation.


Acknowledgments