AlveolEye: Rapid and precise lung morphometry guided by computer vision

Area of Science:

• Pulmonary research
• Histopathology
• Computational biology

Background:

• Accurate lung tissue evaluation is crucial for understanding development, injury, and drug effects.
• Traditional manual methods for lung morphometry (MLI, ASVD) are time-consuming and prone to inter-observer variability.
• Need for reproducible and high-throughput methods in lung research.

Purpose of the Study:

• To develop AlveolEye, an open-source, semi-automated tool for precise lung morphometry.
• To improve the reproducibility and efficiency of calculating mean linear intercept (MLI) and airspace volume density (ASVD).
• To validate AlveolEye's performance against manual measurements and across different species.

Main Methods:

• Development of AlveolEye, a computer vision-assisted software for analyzing H&E stained lung tissue.
• Semi-automated calculation of MLI and ASVD using AlveolEye.
• Comparison of AlveolEye-derived measurements with manual measurements in mouse models of lung injury.
• Analysis of human lung tissue to assess generalizability.

Main Results:

• AlveolEye-assisted MLI calculations closely matched manual measurements.
• Trends in measurements were preserved between control and injured lung samples.
• AlveolEye significantly reduced inter-observer variation, especially for inexperienced users.
• The tool demonstrated generalizability across species, including human lung tissue.

Conclusions:

• AlveolEye provides a rapid, reproducible, and accurate method for lung morphometry.
• The semi-automated design allows for investigator control and adaptability.
• AlveolEye enhances statistical power in preclinical studies by enabling larger sample sizes and improving measurement precision.