API Reference

Welcome to the Nellie API documentation. The Nellie package provides a complete pipeline for automated organelle segmentation, tracking, and hierarchical feature extraction in 2D/3D live-cell microscopy.

Main Components

• Main Interface - Entry point for running the Nellie pipeline
• Segmentation - Image filtering, labeling, and network analysis
• Tracking - Object tracking across timepoints
• Feature Extraction - Hierarchical feature computation
• Image Info - File handling and metadata management
• Utilities - Helper functions for logging and GPU operations

Quick Start

from nellie import run
from nellie.im_info import FileInfo

# Load your microscopy image
file_info = FileInfo("path/to/image.tif")

# Run the complete Nellie pipeline
im_info = run(file_info)

Pipeline Overview

The Nellie pipeline consists of several stages:

1. Filtering: Multi-scale Frangi filtering for vesselness detection
2. Segmentation: Threshold-based instance segmentation
3. Network Analysis: Skeletonization and topological analysis
4. Marker Detection: Motion capture marker generation
5. Tracking: Temporal tracking using Hu moments and flow interpolation
6. Voxel Reassignment: Frame-to-frame voxel tracking
7. Feature Extraction: Multi-level hierarchical features