Related Experiment Video
Updated: Nov 25, 2025
07:17
Author Spotlight: Studying Spatial Protein Expression Using Agarose Embedded Lung Tissue Sections
Published on: October 6, 2023
7.1K
Correction: Deep-LUMEN assay - human lung epithelial spheroid classification from brightfield images using deep
Lyan Abdul1, Shravanthi Rajasekar, Dawn S Y Lin
1School of Biomedical Engineering, McMaster University, 1280 Main Street West, Hamilton, ON L8S 4L8, Canada.
Lab on a Chip
|December 17, 2020
Summary
This study corrects a previous publication on classifying human lung epithelial spheroids using deep learning. The Deep-LUMEN assay provides a robust method for analyzing these complex biological structures.
Area of Science:
- Biomedical Engineering
- Cell Biology
- Machine Learning
Background:
- Accurate classification of human lung epithelial spheroids is crucial for disease modeling.
- Deep learning offers potential for automating spheroid analysis.
- Previous work using the Deep-LUMEN assay required correction.
Purpose of the Study:
- To provide a correction to the previously published 'Deep-LUMEN assay' study.
- To ensure accurate classification of human lung epithelial spheroids.
- To validate the use of deep learning in analyzing brightfield images of spheroids.
Main Methods:
- Re-analysis of brightfield images of human lung epithelial spheroids.
- Application of deep learning algorithms for classification.
- Correction of parameters and results from the original study.
Main Results:
- Identified and corrected errors in the original deep learning model's classification.
- Validated the improved accuracy of the Deep-LUMEN assay for spheroid analysis.
- Confirmed the utility of brightfield imaging with deep learning for lung spheroid characterization.
Conclusions:
- The corrected Deep-LUMEN assay provides a reliable method for classifying lung spheroids.
- Deep learning is a powerful tool for quantitative analysis in lung epithelial biology.
- Accurate computational methods are essential for advancing respiratory research.
