Adiraju Rama Vasantha, Kalyani Kapula, Suryanarayana Gunnam, Zakariah Mohammed, Almazyad Abdulaziz S
Department of Electronics and Communication Engineering, Aditya University, Surampalem 533437, Andhra Pradesh, India.
Department of Electronics and Communication Engineering, Siddhartha Academy of Higher Education (Deemed to be a University), Vijayawada 520007, Andhra Pradesh, India.
Bioengineering (Basel). 2025 May 27;12(6):576. doi: 10.3390/bioengineering12060576.
Radiomics, an emerging discipline integrating imaging science, computational biology, and clinical oncology, enables the extraction of quantitative biomarkers from medical images for improved diagnosis and prognosis. However, variability in imaging protocols and insufficient validation studies hinder the clinical reliability of these biomarkers, limiting their integration into precision medicine. This study addresses these challenges by proposing an RW-ensemble method for extracting and validating radiomic features from segmented lung nodules. Using the Lung CT-Diagnosis dataset, which comprises CT images of 61 patients with segmentation annotations, nearly 38 radiomic features were extracted, incorporating texture-based features from the Grey-Level Co-occurrence Matrix (GLCM) and Grey-Level Run Length Matrix (GLRLM), as well as histogram-based features. The extracted features were validated against ground-truth data using Spearman's correlation coefficient (SCC), demonstrating moderate to strong correlations. These findings confirm the robustness of the RW-ensemble segmentation and reinforce the potential of radiomics in enhancing diagnostic accuracy and guiding therapeutic decisions in precision oncology. Establishing the reliability and reproducibility of these features is crucial for their seamless clinical integration, ultimately advancing the role of radiomics in the diagnosis and treatment of lung adenocarcinoma.
放射组学是一门融合了影像科学、计算生物学和临床肿瘤学的新兴学科，能够从医学影像中提取定量生物标志物，以改善诊断和预后。然而，成像协议的可变性和验证研究的不足阻碍了这些生物标志物的临床可靠性，限制了它们融入精准医学。本研究通过提出一种RW集成方法来解决这些挑战，该方法用于从分割的肺结节中提取和验证放射组学特征。使用包含61例患者带有分割注释的CT图像的Lung CT-Diagnosis数据集，提取了近38个放射组学特征，包括来自灰度共生矩阵（GLCM）和灰度游程长度矩阵（GLRLM）的基于纹理的特征以及基于直方图的特征。使用斯皮尔曼相关系数（SCC）根据真实数据对提取的特征进行验证，显示出中度到高度的相关性。这些发现证实了RW集成分割的稳健性，并加强了放射组学在提高精准肿瘤学诊断准确性和指导治疗决策方面的潜力。确定这些特征的可靠性和可重复性对于它们无缝融入临床至关重要，最终推动放射组学在肺腺癌诊断和治疗中的作用。
