Related Experiment Video
Updated: Jun 5, 2026
Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function
Published on: April 12, 2024
PySERA: Open-source standardized python library for automated, scalable, and reproducible handcrafted and deep
Mohammad R Salmanpour1, Amir Hossein Pouria2, Sirwan Barichin2
1Department of Basic and Translational Research, BC Cancer Research Institute, Vancouver, BC, Canada; Department of Radiology, University of British Columbia, Vancouver, BC, Canada; Technological Virtual Collaboration (TECVICO Corp.), Vancouver, BC, Canada.
PySERA is a new Python framework for radiomics analysis, improving reproducibility and integrating deep learning features. It offers a scalable and standardized solution for AI-ready precision imaging research.
Area of Science:
- Medical Imaging
- Computational Biology
- Artificial Intelligence
Background:
- Radiomics analysis faces challenges in reproducibility and scalability due to fragmented implementations.
- Existing tools often lack integration with deep learning (DL) radiomics and support only partial standards.
Purpose of the Study:
- To develop PySERA, an open-source, Python-native, standardized radiomics framework.
- To enhance automation, reproducibility, and AI integration in radiomics analysis.
Main Methods:
- PySERA is a modular, object-oriented Python framework implementing standardized radiomics analysis.
- It computes 557 features, including 487 IBSI-compliant features and DL radiomics embeddings (ResNet50, DenseNet121, VGG16).
- Features standardized preprocessing, multi-format I/O, adaptive memory handling, and a parallel multi-core engine for scalable extraction, integrating with ML libraries.
Main Results:
- PySERA demonstrated >94% IBSI reproducibility, outperforming PyRadiomics.
- Achieved accuracies of 0.43-0.84 across 8 public datasets for outcome prediction.
- Efficient processing with deterministic outputs across platforms, handling both handcrafted and DL features.
Conclusions:
- PySERA provides a reproducible and extensible foundation for AI-ready precision imaging research.
- It unifies standardized handcrafted and DL radiomics in a scalable, transparent, Python-integrable framework.