Abstract
The premise of radiomics involves extracting high-dimensional quantitative features from medical images to aid clinical decision-making. While radiomics has shown promise in predicting disease characteristics, concerns regarding confounders, reproducibility, and interpretability limit its clinical adoption. In this study, we assessed the ability of radiomic features extracted from contoured CT images to classify two distinct tumour models, CT26 colorectal cancer (CRC) and 4T1 breast cancer (BC), in a highly controlled murine setting. We aimed to provide compelling data for the role of radiomics as a reliable cancer biomarker. We benchmarked radiomics-based classification against previously established blood-based biomarkers, including leukocyte populations and plasma proteins. Feature filtering reduced the original 1409 radiomic features to 18 non-redundant, high-importance predictors, primarily texture-based transformations. Unsupervised clustering via UMAP revealed that radiomics-based features did not segregate tumour types as effectively as blood biomarkers, suggesting potential confounding factors. Supervised machine learning using Random Forest showed that radiomic features achieved a classification accuracy of 0.87, lower than the 0.96 and 0.99 accuracies obtained with cell and plasma biomarkers, respectively. Furthermore, integrating radiomics with blood biomarkers did not enhance classification performance, and feature importance analysis using SHAP identified blood-based markers as the dominant predictors. These findings suggest that while radiomics retains some predictive value, it is less effective than blood biomarkers in this classification task and does not significantly contribute to multimodal tumour classification models. Our study underscores the need for further standardization and validation of radiomics before its clinical implementation.
Data availability
The initial (before filtering) radiomics features are available in Supplementary File (2) The raw sample data for the blood biomarkers are available in the original paper[19] and Supplementary File (1) The R code for the analysis is in Supplementary File (3) The tumour conformal contours and underlying DICOMS images are available in the Australian National University (ANU) DATA COMMONS repository at DOI: https://doi.org/10.25911/zkcm-ab43. Sphere contours and underlying DICOMS images are available in the Zenodo repository at DOI: https://doi.org/10.5281/zenodo.15070060.
References
van Timmeren, J. E., Cester, D., Tanadini-Lang, S., Alkadhi, H. & Baessler, B. Radiomics in medical imaging—‘how-to’ guide and critical reflection. Insights into Imaging. 11 (1), 91. https://doi.org/10.1186/s13244-020-00887-2 (2020).
Lambin, P. et al. Radiomics: Extracting more information from medical images using advanced feature analysis. Eur. J. Cancer 48(4), 441–446. https://doi.org/10.1016/j.ejca.2011.11.036 (2012).
van Griethuysen, J. J. M. et al. Computational radiomics system to Decode the radiographic phenotype. Cancer Res. 77 (21), e104–e107. https://doi.org/10.1158/0008-5472.CAN-17-0339 (2017).
Welcome to. pyradiomics documentation! — pyradiomics v3.1.0rc2.post5 + g6a761c4 documentation. Accessed: Feb. 06, 2025. [Online]. Available: https://pyradiomics.readthedocs.io/en/latest/.
Li, C. et al. Multi-View mammographic density classification by dilated and Attention-Guided residual learning. IEEE/ACM Trans. Comput. Biol. Bioinf. 18 (3), 1003–1013. https://doi.org/10.1109/TCBB.2020.2970713 (2021).
Kolossváry, M. et al. Contribution of risk factors to the development of coronary atherosclerosis as confirmed via coronary CT angiography: A longitudinal Radiomics-based study. Radiology 299 (1), 97–106. https://doi.org/10.1148/radiol.2021203179 (2021).
Refaee, T. et al. The emerging role of radiomics in COPD and lung cancer. Respiration 99 (2), 99–107. https://doi.org/10.1159/000505429 (2020).
Aghakhanyan, G. et al. Radiomics insight into the neurodegenerative ‘hot’ brain: A narrative review from the nuclear medicine perspective. Front. Nucl. Med. 3, 1143256. https://doi.org/10.3389/fnume.2023.1143256 (2023).
Virtual Biopsy. A Safer Future for Cancer Management, Diagnostic Imaging. Accessed: Feb. 06, 2025. [Online]. Available: https://www.diagnosticimaging.com/view/virtual-biopsy-a-safer-future-for-cancer-management.
Bera, K., Braman, N., Gupta, A., Velcheti, V. & Madabhushi, A. Predicting cancer outcomes with radiomics and artificial intelligence in radiology. Nat. Rev. Clin. Oncol. 19 (2), 132–146. https://doi.org/10.1038/s41571-021-00560-7 (2022).
Jiang, W., Wu, R., Yang, T., Yu, S. & Xing, W. Profiling regulatory T lymphocytes within the tumor microenvironment of breast cancer via radiomics. Cancer Med. 12 (24), 21861–21872. https://doi.org/10.1002/cam4.6757 (2023).
Gao, X. et al. A CT-based radiomics signature for evaluating tumor infiltrating Treg cells and outcome prediction of gastric cancer. Ann. Transl Med. 8 (7), 469. https://doi.org/10.21037/atm.2020.03.114 (2020).
Katsoulakis, E. et al. Radiomic analysis identifies tumor subtypes associated with distinct molecular and microenvironmental factors in head and neck squamous cell carcinoma. Oral Oncol. 110, 104877. https://doi.org/10.1016/j.oraloncology.2020.104877 (2020).
Lu, L. et al. Uncontrolled confounders May lead to false or overvalued radiomics signature: A proof of concept using survival analysis in a multicenter cohort of kidney cancer. Front. Oncol. 11, 638185. https://doi.org/10.3389/fonc.2021.638185 (2021).
Zwanenburg, A. et al. The image biomarker standardization initiative: standardized quantitative radiomics for High-Throughput image-based phenotyping. Radiology 295 (2), 328–338. https://doi.org/10.1148/radiol.2020191145 (2020).
Mali, S. A. et al. Making radiomics more reproducible across scanner and imaging protocol variations: a review of harmonization methods. J. Pers. Med. 11(9), Art. no. 9. https://doi.org/10.3390/jpm11090842 (2021).
Wang, X. et al. A pathology foundation model for cancer diagnosis and prognosis prediction. Nature 634 (8035), 970–978. https://doi.org/10.1038/s41586-024-07894-z (2024).
Simon Davis, D. A. et al. Identifying cancer-associated leukocyte profiles using high-resolution flow cytometry screening and machine learning. Front. Immunol. 14, 1211064. https://doi.org/10.3389/fimmu.2023.1211064 (2023).
Simon Davis, D. A. et al. Machine learning predicts cancer subtypes and progression from blood immune signatures. PLoS One. 17, e0264631. https://doi.org/10.1371/journal.pone.0264631 (2022). no. 2.
Kursa, M. B. & Rudnicki, W. R. Feature selection with the Boruta package. J. Stat. Softw. 36, 1–13. https://doi.org/10.18637/jss.v036.i11 (2010).
Breiman, L. Random Forests Mach. Learn., 45, 1, 5–32, doi: https://doi.org/10.1023/A:1010933404324. (2001).
Oksanen, J. et al. Jan. 29,., vegan: Community Ecology Package. Accessed: Feb. 06, 2025. [Online]. (2025). Available: https://cran.r-project.org/web/packages/vegan/index.html.
Anderson, M. J. A new method for non-parametric multivariate analysis of variance. Austral Ecol. 26 (1), 32–46. https://doi.org/10.1111/j.1442-9993.2001.01070.pp.x (2001).
Clarke, K. R. Non-parametric multivariate analyses of changes in community structure. Aust. J. Ecol. 18 (1), 117–143. https://doi.org/10.1111/j.1442-9993.1993.tb00438.x (1993).
McInnes, L., Healy, J., Saul, N. & Großberger, L. Uniform manifold approximation and projection. J. Open. Source Softw. 3 (29), 861. https://doi.org/10.21105/joss.00861 (2018).
Levine, J. H. et al. Data-Driven phenotypic dissection of AML reveals Progenitor-like cells that correlate with prognosis. Cell 162 (1), 184–197. https://doi.org/10.1016/j.cell.2015.05.047 (2015).
Fawagreh, K., Gaber, M. M. & Elyan, E. Random forests: from early developments to recent advancements, Systems Science & Control Engineering: An Open Access Journal, Dec. Accessed: Sep. 30, 2025. [Online]. Available: https://www.tandfonline.com/doi/abs/ (2014). https://doi.org/10.1080/21642583.2014.956265.
Hadley Wickham, M. K. Tidymodels: a collection of packages for modeling and machine learning using tidyverse principles. Accessed: Feb. 06, 2025. [Online]. Available: https://www.tidymodels.org/.
van Buuren, S. & Groothuis-Oudshoorn, K. Mice: multivariate imputation by chained equations in R. J. Stat. Softw. 45, 1–67. https://doi.org/10.18637/jss.v045.i03 (2011).
Lundberg, S. & Lee, S. I. A Unified Approach to Interpreting Model Predictions, Nov. 25, arXiv: arXiv:1705.07874. (2017). https://doi.org/10.48550/arXiv.1705.07874.
Kothari, G. et al. The impact of inter-observer variation in delineation on robustness of radiomics features in non-small cell lung cancer. Sci. Rep. 12 (1), 12822. https://doi.org/10.1038/s41598-022-16520-9 (2022).
Traverso, A., Wee, L., Dekker, A. & Gillies, R. Repeatability and reproducibility of radiomic features: A systematic review. Int. J. Radiat. Oncol. Biol. Phys. 102 (4), 1143–1158. https://doi.org/10.1016/j.ijrobp.2018.05.053 (2018).
The Technome - A Predictive Internal Calibration Approach for Quantitative Imaging Biomarker Research | Scientific Reports. Accessed: Mar. 07, 2025. [Online]. Available: https://www.nature.com/articles/s41598-019-57325-7.
Permuth, J. B. et al. Comparison of radiomic features in a diverse cohort of patients with pancreatic ductal adenocarcinomas. Front. Oncol. 11 https://doi.org/10.3389/fonc.2021.712950 (2021).
Kavandi, H. et al. Radiomics-Based prediction of patient demographic characteristics on chest radiographs: looking beyond deep learning for risk of bias. Am. J. Roentgenol. 224 (2), e2431963. https://doi.org/10.2214/AJR.24.31963 (2025).
Whitney, H. M. et al. Mar., Effect of diversity of patient population and acquisition systems on the use of radiomics and machine learning for classification of 2,397 breast lesions. In Medical Imaging 2019: Computer-Aided Diagnosis, SPIE, 323–329. https://doi.org/10.1117/12.2512507 (2019).
Hagiwara, A., Fujita, S., Ohno, Y. & Aoki, S. Variability and standardization of quantitative imaging. Invest. Radiol. 55 (9), 601–616. https://doi.org/10.1097/RLI.0000000000000666 (2020).
Ferro, M. et al. Radiomics in prostate cancer: an up-to-date review. Ther. Adv. Urol. 14, 17562872221109020. https://doi.org/10.1177/17562872221109020 (2022).
Qi, Y. J. et al. Radiomics in breast cancer: current advances and future directions. Cell. Rep. Med. 5 (9), 101719. https://doi.org/10.1016/j.xcrm.2024.101719 (2024).
Staal, F. C. R. et al. Radiomics for the prediction of treatment outcome and survival in patients with colorectal cancer: A systematic review. Clin. Colorectal Cancer. 20 (1), 52–71. https://doi.org/10.1016/j.clcc.2020.11.001 (2021).
Traverso, A. et al. Machine learning helps identifying volume-confounding effects in radiomics. Physica Med. 71, 24–30. https://doi.org/10.1016/j.ejmp.2020.02.010 (2020).
Liu, Z. et al. The applications of radiomics in precision diagnosis and treatment of oncology: opportunities and challenges. Theranostics 9 (5), 1303–1322. https://doi.org/10.7150/thno.30309 (2019).
Einen, C. et al. Impact of the tumor microenvironment on delivery of nanomedicine in tumors treated with ultrasound and microbubbles. J. Controlled Release. 378, 656–670. https://doi.org/10.1016/j.jconrel.2024.12.037 (2025).
Filatenkov, A. et al. Ablative tumor radiation can change the tumor immune cell microenvironment to induce durable complete remissions. Clin. Cancer Res. 21 (16), 3727–3739. https://doi.org/10.1158/1078-0432.CCR-14-2824 (2015).
De Henau, O. et al. Overcoming resistance to checkpoint Blockade therapy by targeting PI3Kγ in myeloid cells. Nature 539, 443–447. https://doi.org/10.1038/nature20554 (2016).
Rupp, T. et al. Anti-CTLA-4 and anti-PD-1 immunotherapies repress tumor progression in preclinical breast and colon model with independent regulatory T cells response. Translational Oncol. 20, 101405. https://doi.org/10.1016/j.tranon.2022.101405 (2022).
Sato, Y., Fu, Y., Liu, H., Lee, M. Y. & Shaw, M. H. Tumor-immune profiling of CT-26 and colon 26 syngeneic mouse models reveals mechanism of anti-PD-1 response. BMC Cancer. 21 (1), 1222. https://doi.org/10.1186/s12885-021-08974-3 (2021).
Mali, S. A. et al. Making radiomics more reproducible across scanner and imaging protocol variations: A review of harmonization methods. J. Pers. Med. 11 (9), 842. https://doi.org/10.3390/jpm11090842 (2021).
Yang, F. et al. Impact of contouring variability on oncological PET radiomics features in the lung. Sci. Rep. 10 (1), 369. https://doi.org/10.1038/s41598-019-57171-7 (2020).
Fernandez, J. L. et al. A comparative analysis of orthotopic and subcutaneous pancreatic tumour models: tumour microenvironment and drug delivery. Cancers 15 (22), 5415. https://doi.org/10.3390/cancers15225415 (2023).
Hiam-Galvez, K. J., Allen, B. M. & Spitzer, M. H. Systemic immunity in cancer. Nat. Rev. Cancer. 21 (6), 345–359. https://doi.org/10.1038/s41568-021-00347-z (2021).
Paradoxical roles of the immune system during cancer development | Nature Reviews Cancer. Accessed: Mar. 11, 2025. [Online]. Available: https://www.nature.com/articles/nrc1782.
Finn, O. J. Immuno-oncology: Understanding the function and dysfunction of the immune system in cancer. Ann. Oncol. 23 (9), viii6–viii. https://doi.org/10.1093/annonc/mds256 (2012).
Boivin, G. et al. Durable and controlled depletion of neutrophils in mice. Nat. Commun. 11 (1), 2762. https://doi.org/10.1038/s41467-020-16596-9 (2020).
Brattain, M. G., Strobel-Stevens, J., Fine, D., Webb, M. & Sarrif, A. M. Establishment of mouse colonic carcinoma cell lines with different metastatic properties. Cancer Res. 40 (7), 2142–2146 (1980).
Aslakson, C. J. & Miller, F. R. Selective events in the metastatic process defined by analysis of the sequential dissemination of subpopulations of a mouse mammary tumor. Cancer Res. 52 (6), 1399–1405 (1992).
Fedorov, A. et al. 3D slicer as an image computing platform for the quantitative imaging network. Magn. Reson. Imaging. 30 (9), 1323–1341. https://doi.org/10.1016/j.mri.2012.05.001 (2012).
Riccardo, M. norok2/raster_geometryAug. 20,. Python. Accessed: Apr. 04, 2025. [Online]. (2023). Available: https://github.com/norok2/raster_geometry.
BQ_radiomics/BQ_radiomics/scripts/run_all_radiomics.py at master · dorkylever/BQ_radiomics, GitHub. Accessed: Feb. 18, 2025. [Online]. Available: https://github.com/dorkylever/BQ_radiomics/blob/master/BQ_radiomics/scripts/run_all_radiomics.py.
Li, X., Morgan, P. S., Ashburner, J., Smith, J. & Rorden, C. The first step for neuroimaging data analysis: DICOM to NIfTI conversion. J. Neurosci. Methods. 264, 47–56. https://doi.org/10.1016/j.jneumeth.2016.03.001 (2016).
Horner, N. R. et al. LAMA: automated image analysis for the developmental phenotyping of mouse embryos. Development 148, dev192955. https://doi.org/10.1242/dev.192955 (2021).
tidyverse 1.3.0. Accessed: Feb. 18, 2025. [Online]. Available: https://www.tidyverse.org/blog/2019/11/tidyverse-1-3-0/.
Create Elegant Data Visualisations Using the Grammar of Graphics. Accessed: Feb. 18. [Online]. (2025). Available: https://ggplot2.tidyverse.org/.
Preprocessing and Feature Engineering Steps for Modeling. Accessed: Feb. 18, 2025. [Online]. Available: https://recipes.tidymodels.org/index.html.
Goode, K. & Rey, K. ggResidpanel: Panels and Interactive Versions of Diagnostic Plots using ggplot2.May 31, Accessed: Feb. 18, 2025. [Online]. (2019). Available: https://cran.r-project.org/web/packages/ggResidpanel/index.html.
R. The R Stats Package. Accessed: Feb. 18, 2025. [Online]. Available: https://stat.ethz.ch/R-manual/R-devel/library/stats/html/00Index.html.
Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: A practical and powerful approach to multiple testing. J. Roy. Stat. Soc.: Ser. B (Methodol.). 57 (1), 289–300. https://doi.org/10.1111/j.2517-6161.1995.tb02031.x (1995).
McKight, P. E. & Najab, J. Kruskal-Wallis test, in The Corsini Encyclopedia of Psychology, John Wiley & Sons, Ltd, 1–1. doi: https://doi.org/10.1002/9780470479216.corpsy0491. (2010).
Dunn, O. J. Multiple Comparisons Using Rank Sums, Technometrics, Aug. Accessed: Sep. 30, 2025. [Online]. Available: https://www.tandfonline.com/doi/abs/ (1964). https://doi.org/10.1080/00401706.1964.10490181.
Bauer, D. F. Constructing confidence sets using rank statistics. J. Am. Stat. Assoc. 67 (339), 687–690. https://doi.org/10.1080/01621459.1972.10481279 (1972).
Kassambara, A. rstatix: Pipe-Friendly Framework for Basic Statistical TestsFeb. 01,. Accessed: Sep. 30, 2025. [Online]. (2023). Available: https://cran.r-project.org/web/packages/rstatix/index.html.
Torchiano, M. Effsize - a package for efficient effect size computation. Nov 13 https://doi.org/10.5281/zenodo.196082 (2016). Zenodo.
Konopka, T. umap: Uniform Manifold Approximation and Projection. (Feb. 01, 2023). Accessed: Feb. 19, 2025. [Online]. Available: https://cran.r-project.org/web/packages/umap/index.html.
Wright, M. N., Wager, S. & Probst, P. ranger: A Fast Implementation of Random Forests. (Nov. 08, 2024). Accessed: Feb. 19, 2025. [Online]. Available: https://cran.r-project.org/web/packages/ranger/index.html.
JinmiaoChenLab/Rphenograph. (Oct. 28, 2024). R. Jinmiao Chen’s Lab. Accessed: Mar. 06, 2025. [Online]. Available: https://github.com/JinmiaoChenLab/Rphenograph.
Figueroa, R. L., Zeng-Treitler, Q., Kandula, S. & Ngo, L. H. Predicting sample size required for classification performance. BMC Med. Inf. Decis. Mak. 12 (1), 8. https://doi.org/10.1186/1472-6947-12-8 (2012).
Perlich, C. et al. Tree Induction Vs. Logistic Regression: A Learning-Curve Analysis.
Funding
This work was supported by the Radiation Oncology Private Practice Trust Fund, Canberra Health Services and with assistance from the ACT Government’s Research and Innovation Fund.
Author information
Authors and Affiliations
Contributions
BJCQ, FMS, IA and HE conceived and designed the work; BJCQ, KD, DASD, JP, KG, NO, and KJ acquired the data; BJCQ, and KD analysed and interpreted the data; BJCQ drafted the manuscript; and KD, KG, KJ, and FMS revised the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Drover, K., Davis, D.A.S., Gosling, K. et al. Cancer classification with radiomics in controlled preclinical models. Sci Rep (2026). https://doi.org/10.1038/s41598-026-37757-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-37757-8