Iranian Journal of Nuclear Medicine

Iranian Journal of Nuclear Medicine

Stability of textural PET radiomics features across SUVmax-based VOI segmentation thresholds: A clinical PET/CT study

Articles in Press

Document Type : Original Article

Authors
Peyman Sheikhzadeh 1, 2
Saeed Farzanehfar 1
Younus soleymani 3
Abolfazl Asadollahi 4
Aria Forouz 5
Zahra Valibeigloo 6
Elaheh Saiyadi 7
Davood Khezerloo 8, 9
1 Department of Nuclear Medicine, Imam Khomeini Hospital Complex, Tehran University of Medical Sciences, Tehran, Iran
2 Research Center for Nuclear Medicine, Tehran University of Medical Sciences, Tehran, Iran
3 Department of Neuroscience and Addiction Studies, School of Advanced Technologies in Medicine, Tehran University of Medical Sciences, Tehran, Iran
4 Department of Radiotherapy, Faculty of Medicine, Madani Hospital, Tabriz University of Medical Sciences, Tabriz, Iran
5 Department of Nuclear Medicine, Emsey Hospital, Istanbul, Turkey
6 Department of Medical Physics, Faculty of Medicine, Tabriz University of Medical Sciences, Tabriz, Iran
7 Department of Nuclear Medicine, Omid PET/CT Center, Tabriz, Iran
8 Medical Radiation Sciences Research Center, Tabriz University of Medical Sciences, Tabriz, Iran
9 Department of Radiology, Faculty of Allied Medical Sciences, Tabriz University of Medical Sciences, Tabriz, Iran
10.22034/irjnm.2026.130602.1772
Abstract
Introduction: The stability of quantitative PET radiomics features is an important consideration for reproducible radiomics analysis. This study investigated the stability of textural PET radiomics features across different SUVmax-based VOI segmentation.
Methods: PET/CT images from 24 patients with primary colon adenocarcinoma and 24 patients with hepatic metastases were retrospectively analyzed. Lesion VOIs were automatically generated using four SUVmax-based segmentation thresholds (10%, 20%, 30%, and 40%). A total of 32 textural features were extracted for each VOI. Feature stability was assessed using the intraclass correlation coefficient (ICC), with 95% confidence intervals.
Results: The stability of textural radiomics features varied across individual features and SUVmax-based segmentation thresholds. In the liver metastasis cohort, 28.1% of features demonstrated good to excellent stability, compared with 53.1% in the colon adenocarcinoma cohort. The feature with the highest stability was GLZLM_HGZE in the liver metastasis cohort (ICC = 0.923) and GLZLM_SZHGE in the colon adenocarcinoma cohort (ICC = 0.947). Several of the features with the highest ICC values belonged to the GLZLM and GLRLM families, whereas GLCM features generally showed lower stability, particularly in the liver metastasis cohort.
Conclusion: Textural PET radiomics features exhibit feature-dependent stability to changes in SUVmax-based VOI segmentation thresholds. The identification of features that remain stable across segmentation conditions may help improve the reproducibility of PET radiomics analyses. Differences in the observed stability patterns between colon adenocarcinoma and hepatic metastases should be interpreted cautiously because lesion size, uptake characteristics, lesion-to-background contrast, and partial-volume effects may also contribute to these differences.
Keywords
Radiomics
Positron emission tomography
Feature stability
Subjects
  1. Kirienko M, Gelardi F, Fiz F, Bauckneht M, Ninatti G, Pini C, Briganti A, Falconi M, Oyen WJG, van der Graaf WTA, Sollini M. Personalised PET imaging in oncology: an umbrella review of meta-analyses to guide the appropriate radiopharmaceutical choice and indication. Eur J Nucl Med Mol Imaging. 2024 Dec;52(1):208-24.
  2. Lew LZW, Mac Curtain BM, Siew T, Ng ZQ. Predictive role of FDG PET-CT in localised rectal carcinoma: a systematic review and meta-analysis. J Med Imaging Radiat Oncol. 2025 Apr;69(3):352-72.
  3. Li S, Bai R, Wang H, Sun Q, Wang G, Jiang S, Dong Y. Prognostic significance of 18F-FDG PET/CT parameters in soft tissue sarcoma: a systematic review and meta-analysis. Cancer Imaging. 2025 Jul 29;25(1):94.
  4. Qiao T, Cheng Z, Duan Y. Innovative applications and future trends of multiparametric PET in the assessment of immunotherapy efficacy. Front Oncol. 2025 Jan 20;14:1530507.
  5. da Cunha Tavares SC, Bessa Dos Santos Oliveiros Paiva BC, das Neves Teixeira de Lemos LP, Amado Lapa PA. Metabolic tumour volume as an independent predictor of chemoradiotherapy response in oesophageal cancer. Clin Transl Imaging. 2025 Oct;13(5):545-55.
  6. Huang M, Law HKW, Tam SY. Use of radiomics in characterizing tumor hypoxia. Int J Mol Sci. 2025 Jul 11;26(14):6679.
  7. Soleymani Y, Jahanshahi AR, Pourfarshid A, Khezerloo D. Reproducibility assessment of radiomics features in various ultrasound scan settings and different scanner vendors. J Med Imaging Radiat Sci. 2022 Dec;53(4):664-71.
  8. Jahanshahi A, Soleymani Y, Fazel Ghaziani M, Khezerloo D. Radiomics reproducibility challenge in computed tomography imaging as a nuisance to clinical generalization: a mini-review. Egyptian J Radiol Nucl Med. 2023 May 9;54(1):83.
  9. Dwivedi P, Barage S, Singh R, Jha A, Choudhury S, Agrawal A, Rangarajan V. PET-based radiomic analysis in multicentre lung cancer study and impact of feature domain harmonization. Phys Eng Sci Med. 2025 Dec;48(4):1841-51.
  10. Soleymani Y, Sheikhzadeh P, Mohammadzadeh M, Khezerloo D. Simplifying radiomics workflow for predicting grade of glioma: an approach for rapid and reproducible radiomics. J Biomed Phys Eng. 2025 Feb 1;15(1):27-36.
  11. He W, Huang W, Zhang L, Wu X, Zhang S, Zhang B. Radiogenomics: bridging the gap between imaging and genomics for precision oncology. MedComm (2020). 2024 Sep 9;5(9):e722.
  12. Cepero A, Yang Y, Young L, Huang J, Ji X, Yang F. Longitudinal FDG-PET radiomics for early prediction of treatment response to chemoradiation in locally advanced cervical cancer: a pilot study. Cancers (Basel). 2024 Nov 13;16(22):3813.
  13. Valibeiglou Z, Islamian JP, Soleymani Y, Farzanehfar S, Aghahosseini F, Gilani N, Rahmim A, Sheikhzadeh P. Enhancing radiomics robustness using bayesian penalized likelihood PET reconstruction: application to phantom and non-small cell lung cancer patient studies. BMC Med Imaging. 2025 Jul 1;25(1):223.
  14. Liu Y, Wang G, Yan X, Wang H, Li T, Yang S, Liu Z, Zhang S, Su X. Effect of reconstruction settings on radiomics feature robustness in positron emission tomography images: a clinical study. Quant Imaging Med Surg. 2025 Oct 1;15(10):9515-33.
  15. Fooladi M, Soleymani Y, Rahmim A, Farzanefar S, Aghahosseini F, Seyyedi N, Sh Zadeh P. Impact of different reconstruction algorithms and setting parameters on radiomics features of PSMA PET images: A preliminary study. Eur J Radiol. 2024 Mar;172:111349.
  16. Gallivanone F, Interlenghi M, D'Ambrosio D, Trifirò G, Castiglioni I. Parameters influencing PET imaging features: a phantom study with irregular and heterogeneous synthetic lesions. Contrast Media Mol Imaging. 2018 Sep 10;2018:5324517.
  17. Pasini G, Russo G, Mantarro C, Bini F, Richiusa S, Morgante L, Comelli A, Russo GI, Sabini MG, Cosentino S, Marinozzi F, Ippolito M, Stefano A. A critical analysis of the robustness of radiomics to variations in segmentation methods in 18F-PSMA-1007 PET images of patients affected by prostate cancer. Diagnostics (Basel). 2023 Dec 11;13(24):3640.
  18. Fan R, Long X, Chen X, Wang Y, Chen D, Zhou R. The value of machine learning-based radiomics model characterized by PET Imaging with 68Ga-FAPI in assessing microvascular invasion of hepatocellular carcinoma. Acad Radiol. 2025 Apr;32(4):2233-46.
  19. Stern NM, Mikalsen LT, Dueland S, Schulz A, Line PD, Stokke C, Grut H. The prognostic value of [18F] FDG PET/CT texture analysis prior to transplantation for unresectable colorectal liver metastases. Clin Physiol Funct Imaging. 2025 Jan;45(1):e12908.
  20. Azimi MS, Rahmim A, Arabi H, Sanaat A, Zeraatkar N, Bouchareb Y, Liu C, Alavi A, King M, Boellaard R, Zaidi H. Critical review of partial volume correction methods in PET and SPECT imaging: benefits, pitfalls, challenges, and future outlook. Eur J Nucl Med Mol Imaging. 2026 Mar;53(4):2830-61.
  21. Azimi MS, Cheraghi M, MahdiMaleki F, MahdiMaleki F, Sanaat A, Høilund-Carlsen PF, Alavi A, Zaidi H. Toward standardization: Assessing the reproducibility of radiomics features in partial volume-corrected brain PET images. Neuroimage. 2025 Sep;318:121398.
  22. Flouris K, Jimenez-Del-Toro O, Aberle C, Bach M, Schaer R, Obmann MM, Stieltjes B, Müller H, Depeursinge A, Konukoglu E. Assessing radiomics feature stability with simulated CT acquisitions. Sci Rep. 2022 Mar 18;12(1):4732.
  23. Luo Y, Li Y, Yang Z, Zhang Y, Yu H, Zhao Z, Yu K, Guo Y, Wang X, Yang N, Zhang Y, Zheng T, Zhou J. A multi-view prognostic model for diffuse large B-cell lymphoma based on kernel canonical correlation analysis and support vector machine. BMC Cancer. 2024 Dec 5;24(1):1495.
  24. Watanabe M, Ashida R, Miyakoshi C, Arizono S, Suga T, Kanao S, Kitamura K, Ogawa T, Ishikura R. Prognostic analysis of curatively resected pancreatic cancer using harmonized positron emission tomography radiomic features. Eur J Hybrid Imaging. 2023 Mar 6;7(1):5.
  25. Liu J, Shen C, Jiang S, Wu Y, Tu J, Bao Y, Li H, Wang N, Liu Y, Xiong J, Liu X. Deep learning-based tumor segmentation and radiogenomic model for predicting EGFR amplification and assessing intratumoural heterogeneity in glioblastoma. J Big Data. 2025 Jul 11;12(1):168.
Iranian Journal of Nuclear Medicine

Articles in Press, Accepted Manuscript
Available Online from 02 November 2026