Automated neurite extraction using dynamic programming for high-throughput screening of neuron-based assays.

Area of Science:

• Neuroscience
• Biotechnology
• Image Analysis

Background:

• High-throughput screening (HTS) is crucial for drug discovery.
• Analyzing microscopy neuron images in HTS presents significant challenges.
• Complex neuronal structures and poor image quality hinder accurate analysis.

Purpose of the Study:

• To present a novel algorithm for extracting and quantifying neurite segments from HTS neuron images.
• To develop a robust method for analyzing complex neuronal structures and low-quality images.
• To enhance the efficiency and accuracy of neuron-based assays in drug discovery.

Main Methods:

• A novel algorithm utilizing seed point detection on grid lines and iterative centerline tracing.
• Application of the live-wire method with dynamic programming for linking seed and end points.
• Accurate extraction of neurite segment centerlines, robust to noise and artifacts.

Main Results:

• The algorithm successfully detects and links neurites in complex structures and poor imaging conditions.
• Accurate and robust extraction of neurite segment centerlines is achieved.
• Preservation of thin neurites and low-intensity contrast segments is demonstrated.

Conclusions:

• The proposed algorithm offers a robust and accurate solution for neurite extraction in HTS neuron images.
• Its ability to handle complex structures and image artifacts makes it valuable for drug discovery.
• The algorithm enhances the utility of neuron-based assays in high-throughput screening.