Examining the effects of slice thickness on the reproducibility of CT radiomics for patients with colorectal liver metastases Conference Paper
| Authors: | Peoples, J. J.; Hamghalam, M.; James, I.; Wasim, M.; Gangai, N.; Kang, H. C.; Rong, X. J.; Chun, Y. S.; Do, R. K. G.; Simpson, A. L. |
| Title: | Examining the effects of slice thickness on the reproducibility of CT radiomics for patients with colorectal liver metastases |
| Conference Title: | 5th Workshop on Uncertainty for Safe Utilization of Machine Learning in Medical Imaging (UNSURE 2023) |
| Abstract: | We present an analysis of 81 patients with colorectal liver metastases from two major cancer centers prospectively enrolled in an imaging trial to assess reproducibility of radiomic features in contrastenhanced CT. All scans were reconstructed with different slice thicknesses and levels of iterative reconstruction. Radiomic features were extracted from the liver parenchyma and largest metastasis from each reconstruction, using different levels of resampling and methods of feature aggregation. The prognostic value of reproducible features was tested using Cox proportional hazards to model overall survival in an independent, public data set of 197 hepatic resection patients with colorectal liver metastases. Our results show that larger differences in slice thickness reduced the concordance of features (p < 10-6). Extracting features with 2.5D aggregation and no axial resampling produced the most robust features, and the best test-set performance in the survival model on the independent data set (C-index = 0.65). Across all feature extraction methods, restricting the survival models to use reproduciblefeatures had no statistically significant effect on the test set performance (p = 0.98). In conclusion, our results show that feature extraction settings can positively impact the robustness of radiomics features to variations in slice thickness, without negatively effecting prognostic performance. |
| Keywords: | prospective studies; reproducibility; biomarkers; imaging; computed tomography; colorectal liver metastases; features; concordance correlation-coefficient; radiomics; feature stability |
| Journal Title | Lecture Notes in Computer Science |
| Volume: | 14291 |
| Conference Dates: | 2023 Oct 12 |
| Conference Location: | Vancouver, Canada |
| ISBN: | 0302-9743 |
| Publisher: | Springer |
| Date Published: | 2023-01-01 |
| Start Page: | 42 |
| End Page: | 52 |
| Language: | English |
| ACCESSION: | WOS:001115884500005 |
| DOI: | 10.1007/978-3-031-44336-7_5 |
| PROVIDER: | wos |
| Notes: | This paper was published in proceedings book ISBN: 978-3-031-44335-0) -- Source: Wos |
Altmetric
Citation Impact
BMJ Impact Analytics
Related MSK Work