Computational Radiomics System to Decode the Radiographic Phenotype

Radiomics aims to quantify phenotypic characteristics on medical imaging through automated algorithms—either engineered hard-coded methods or deep learning—to develop non-invasive imaging-based biomarkers. The paper notes that a lack of standardized algorithm definitions and image processing severely hampers the reproducibility and comparability of radiomic results. To address this, the authors developed PyRadiomics, a flexible open-source platform, implemented in Python and usable standalone or through 3D-Slicer, that can extract a large panel of engineered features from medical images.

The paper describes the workflow and architecture of PyRadiomics and demonstrates its application in characterizing lung lesions, with source code, documentation, and examples made publicly available. By providing a tested and maintained platform, the authors aim to establish a reference standard for radiomic analyses and to grow the community of radiomic developers addressing critical needs in cancer research.