HSI-based Vector Graphics Algorithm for Enhanced Detection and 3D Visualization of Pulmonary Lesions on CT
Open Access
- 1 June 2026
- journal article
- Published by Bentham Science Publishers Ltd. in Current Medical Imaging
- Vol. 22, 1-8
Abstract
Introduction/Objective: Multidetector computed tomography (MDCT) enables high-resolution multiplanar and 3D reconstructions in thoracic imaging. However, conventional 3D reconstruction methods are computationally expensive, time-consuming, and require specialized expertise. This short communication introduces a novel algorithm (HGM) based on vector graphics and RGB-to-HSI (hue, saturation, intensity) color-space transformation, designed to provide a low-cost, high-precision tool for visualization and classification of intrathoracic lesions. Methods: Chest CT DICOM datasets were processed using the HGM algorithm, developed in collaboration with visual-effects engineers. The pipeline incorporated RGB-to-HSI conversion, vector-graphics–based modeling, and scripted segmentation for automated isolation of thoracic organs and lesions. Reconstructions were rendered into an interactive format compatible with Adobe Animate®. Results: Pilot testing in a 41-year-old woman with an intrapulmonary teratoma demonstrated accurate 3D reconstruction for preoperative planning. Tumor dimensions were 14×12×10.3 cm, closely matching radiologic measurements (14×12×10 cm). The mean processing time was 15 minutes per dataset using modest computational resources (~US $1,600 workstation) and minimal training (~4 hours). Discussion: The HGM algorithm enabled rapid anatomical segmentation, real-time 3D manipulation, and dynamic visualization from multiple perspectives, offering an alternative to conventional 3D CT reconstruction methods. Conclusion: RGB-to-HSI transformation provides low-cost, high-precision 3D images of intrathoracic lesions. Future directions will focus on validating the algorithm by testing its performance in different tumors, post-surgical follow-up, radiotherapy planning, and other applications.Keywords
This publication has 18 references indexed in Scilit:
- Missed injuries in trauma care: An analysis of mechanisms and prevention of one of the surgeon's worst nightmaresInjury, 2025
- Unsupervised non‐small cell lung cancer tumor segmentation using cycled generative adversarial network with similarity‐based discriminatorJournal of Applied Clinical Medical Physics, 2025
- Multisensory Extended Reality Applications Offer Benefits for Volumetric Biomedical Image Analysis in Research and MedicineJournal of Imaging Informatics in Medicine, 2024
- A dichotomy color quantization algorithm for the HSI color spaceScientific Reports, 2023
- Advances in Imaging to Aid Segmentectomy for Lung CancerSurgical Oncology Clinics of North America, 2022
- Three-dimensional virtual planning for nodule resection in solid organs: A systematic review and meta-analysisSurgical Oncology, 2021
- Computed Tomography–Derived 3D Modeling to Guide Sizing and Planning of Transcatheter Mitral Valve InterventionsJACC: Cardiovascular Imaging, 2021
- Potential Pitfalls in Imaging of the MediastinumRadiologic Clinics of North America, 2021
- Color image quantization with peak-picking and color spaceMultimedia Systems, 2020
- Safety margin after radiofrequency ablation of hepatocellular carcinoma: precise assessment with a three-dimensional reconstruction technique using CT imagingInternational Journal of Hyperthermia, 2018