Welcome to Pictologics

Pictologics is a pure python, IBSI 1 and 2 compliant library for radiomic feature extraction from medical images.

See also the NOTICE file for attribution and third-party library information.

Why Pictologics?

🚀 High Performance: Uses numba for Just In Time (JIT) compilation, achieving significant speedups over other libraries (speedups between 15-300x compared to pyradiomics, see Benchmarks page for details).
✅ IBSI Compliant: Implements standard algorithms verified against the IBSI digital and CT phantoms, and clinical datasets:
- IBSI 1: Feature extraction (compliance report)
- IBSI 2: Image filters (Phase 1 filter compliance | Phase 2 filtered features | Phase 3 reproducibility)
🔧 Versatile: Provides utilities for DICOM parsing and common scientific image processing tasks. Natively supports common image formats (NIfTI, DICOM, DICOM-SEG, DICOM-SR).
✨ User-Friendly: Pure Python implementation with a simple installation process and user-friendly pipeline module supporting easy feature extraction and analysis, ensuring a smooth experience from setup to analysis.
🛠️ Actively Maintained: Continuously maintained and developed with the intention to provide robust latent radiomic features that can reliably describe morphological characteristics of diseases on radiological images.

Loaders: Support for NIfTI and DICOM image, segmentation (DICOM-SEG), and report (DICOM-SR) formats.
Preprocessing: Resampling, resegmentation, outlier filtering, discretisation and others.
Features:
- Morphology: Volume, Surface Area, Compactness, etc.
- Intensity: Mean, Median, Skewness, Kurtosis, etc.
- Texture: GLCM, GLRLM, GLSZM, GLDZM, NGTDM, NGLDM.
Filters: IBSI 2-compliant convolutional filters including Mean, LoG, Laws, Gabor, Wavelets, Simoncelli and others.
Intelligent Deduplication: Automatically avoids redundant computation when multiple configurations share preprocessing but differ only in discretization. Feature families like morphology and intensity are computed once and reused.
Configuration Management: Export/import pipeline configurations in YAML/JSON. Share reproducible workflows, version control settings, and ensure consistent multi-site processing.
Utilities: Built-in DICOM database parsing, organization and viewing tools.