Classification of histological subtypes of non-small cell lung cancer using computerized tomography texture analysis

Yıl 2024, , 168 - 172, 11.10.2024

Tümay Bekci , Merve Nur Tasdemir , Esma Çınar , Demet Sengul , Eylem Karaçay , Sevval Arslan , Sena Nur Cure

https://doi.org/10.70058/cjm.1535113

Öz

Objective: This study aimed to differentiate between the two main histological subtypes of non-small cell lung cancer using a non-invasive technique, computerized tomography texture analysis.

Method: We included 53 patients. All patients were histopathologically proven non-small cell lung cancer cases. All patients underwent thorax CT scans. In CT images, the differences present in the texture features of adenocarcinoma and squamous cell carcinoma, which are the two main histological subtypes of non-small cell lung cancer, were determined by the consensus of two radiologists for computerized tomography-based texture analysis.

Results: A total of 44 texture features were extracted, including 12 first-order features and 32 second-order features derived from gray-level co-occurrence matrix (GLCM), gray-level run-length matrix (GLRLM), neighborhood gray-level different matrix (NGLDM), and gray-level zone length matrix (GLZLM) features in 51 CT images. None of the evaluated texture parameters were statistically significant. However, in patients with squamous cell lung cancer, the values of Intensity Histogram, NGTDM Complexity, and Intensity Based Robust Mean Absolute Deviation higher from adenocarcinoma patients and had the highest area under the curve in roc analyses (0.727, 0.664, 0.666 respectively)

Conclusion: Intensity Histogram, NGTDM Complexity, and Intensity Based Robust Mean Absolute Deviation features can be used to differentiate between the subtypes of non-small cell lung cancer, adenocarcinoma and squamous cell carcinoma. These features were highly associated with the high intratumoral heterogeneity of squamous cell lung cancer.

Anahtar Kelimeler

Artificial intelligence, CT, Non-small cell lung cancer, Radiomics, Texture Analysis

Etik Beyan

This retrospective study obtained approval from the institutional review board of our hospital, and written informed consent was waived [GEAH/KAEK-216].

Kaynakça

• 1. Megyesfalvi Z, Gay CM, Popper H, et al. Clinical insights into small cell lung cancer: Tumor heterogeneity, diagnosis, therapy, and future directions. CA Cancer J Clin. 2023;73(6):620-652.
• 2. Chen Z, Fillmore CM, Hammerman PS, Kim CF, Wong KK. Non-small-cell lung cancers: a heterogeneous set of diseases. Nat Rev Cancer. 2015 Apr;15(4):247].
• 3. de Sousa VML, Carvalho L. Heterogeneity in Lung Cancer. Pathobiology. 2018;85(1-2):96-107.
• 4. Wu F, Fan J, He Y, et al. Single-cell profiling of tumor heterogeneity and the microenvironment in advanced non-small cell lung cancer. Nat Commun. 2021;12(1):2540.
• 5. Fang B, Mehran RJ, Heymach JV, Swisher SG. Predictive biomarkers in precision medicine and drug development against lung cancer. Chin J Cancer. 2015;34(7):295-309.
• 6. Phillips I, Ajaz M, Ezhil V, et al. Clinical applications of textural analysis in non-small cell lung cancer. Br J Radiol. 2018;91(1081):20170267.
• 7. Suga M, Nishii R, Miwa K, et al. Differentiation between nonsmall cell lung cancer and radiation pneumonitis after carbonion radiotherapy by 18F-FDG PET/CT texture analysis. Sci Rep. 2021;11(1):11509.
• 8. Chen S, Harmon S, Perk T, et al. Using neighborhood gray tone difference matrix texture features on dual time point PET/CT images to differentiate malignant from benign FDG-avid solitary pulmonary nodules. Cancer Imaging. 2019;19(1):56.
• 9. Nioche C, Orlhac F, Boughdad S, et al. LIFEx: A Freeware for Radiomic Feature Calculation in Multimodality Imaging to Accelerate Advances in the Characterization of Tumor Heterogeneity. Cancer Res. 2018;78(16):4786-4789.
• 10. Bekci T, Cakir IM, Aslan S. Differentiation of affected and nonaffected ovaries in ovarian torsion with magnetic resonance imaging texture analysis. Rev Assoc Med Bras (1992). 2022;68(5):641-646.