Related Experiment Video
Updated: May 12, 2026

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
Published on: November 11, 2022
Deep-Learning-Based Segmentation of Cells and Analysis (DL-SCAN)
Alok Bhattarai1, Jan Meyer2, Laura Petersilie2
1Department of Physics, University of South Florida, Tampa, FL 33647, USA.
DL-SCAN, a deep learning tool, automates fluorescence microscopy image analysis for cell segmentation and ion dynamics studies. It offers consistent, unbiased results, overcoming limitations of traditional methods for analyzing low signal-to-noise ratio data.
Area of Science:
- Cellular biology
- Neuroscience
- Biophysics
Background:
- Fluorescence microscopy is crucial for studying cellular processes, but traditional image analysis is manual, time-consuming, and prone to user bias.
- Low signal-to-noise ratio (SNR) in fluorophores complicates analysis, often requiring extensive training and limiting automation.
- Existing software lacks automation for diverse datasets and struggles with low-brightness fluorophores.
Purpose of the Study:
- To develop an automated tool, DL-SCAN, for segmenting and analyzing fluorescence microscopy images using deep learning.
- To overcome limitations of manual analysis, including user bias and time inefficiency, especially for low SNR data.
- To enable objective, efficient, and reproducible analysis of cellular ion dynamics in neuroscience research.
Main Methods:
- Developed DL-SCAN, a deep learning-based software for automated segmentation of regions of interest in fluorescence microscopy images.
- Validated DL-SCAN using synthetic image stacks with varying SNR to assess cell identification accuracy.
- Applied DL-SCAN to analyze experimental Na+ and Ca2+ imaging data from mouse brain tissue slices under chemical ischemia.
Main Results:
- DL-SCAN successfully automates cell identification and segmentation in fluorescence microscopy images.
- The tool demonstrated consistent and reproducible analysis of cellular ion dynamics (Na+, Ca2+) in neuronal and astrocyte imaging data.
- Results showed DL-SCAN is free from user bias, enabling rapid and objective data analysis.
Conclusions:
- DL-SCAN provides an efficient, automated, and unbiased solution for analyzing fluorescence microscopy data, particularly for low SNR conditions.
- The open-source nature of DL-SCAN allows for customization and extension to diverse cell types and ion dynamics studies.
- This tool significantly enhances the speed and reliability of cellular imaging analysis in neuroscience and related fields.
Related Concept Videos
Restorative Care
Continuing Care
Nursing Process for Patient and Caregiver Teaching I: Assessment and Diagnosis
It is critical to determine the patient's learning needs during the assessment. Determination of learning needs compounds data from the...
Documentation in Long-Term and Home Healthcare Setting
Long-Term Care Facilities
Alzheimer's Disease: Overview
The clinical diagnosis of AD hinges on the presence of memory and other cognitive impairments. Biomarkers, such as changes in Aβ and tau...
Dementia
The progression of dementia is generally gradual.

