Radiomic feature reproducibility in contrast-enhanced CT of the pancreas is affected by variabilities in scan parameters and manual segmentation
- 72 Downloads
This study aims to measure the reproducibility of radiomic features in pancreatic parenchyma and ductal adenocarcinomas (PDAC) in patients who underwent consecutive contrast-enhanced computed tomography (CECT) scans.
In this IRB-approved and HIPAA-compliant retrospective study, 37 pairs of scans from 37 unique patients who underwent CECTs within a 2-week interval were included in the analysis of the reproducibility of features derived from pancreatic parenchyma, and a subset of 18 pairs of scans were further analyzed for the reproducibility of features derived from PDAC. In each patient, pancreatic parenchyma and pancreatic tumor (when present) were manually segmented by two radiologists independently. A total of 266 radiomic features were extracted from the pancreatic parenchyma and tumor region and also the volume and diameter of the tumor. The concordance correlation coefficient (CCC) was calculated to assess feature reproducibility for each patient in three scenarios: (1) different radiologists, same CECT; (2) same radiologist, different CECTs; and (3) different radiologists, different CECTs.
Among pancreatic parenchyma-derived features, using a threshold of CCC > 0.90, 58/266 (21.8%) and 48/266 (18.1%) features met the threshold for scenario 1, 14/266 (5.3%) and 15/266 (5.6%) for scenario 2, and 14/266 (5.3%) and 10/266 (3.8%) for scenario 3. Among pancreatic tumor-derived features, 11/268 (4.1%) and 17/268 (6.3%) features met the threshold for scenario 1, 1/268 (0.4%) and 5/268 (1.9%) features met the threshold for scenario 2, and no features for scenario 3 met the threshold, respectively.
Variations between CECT scans affected radiomic feature reproducibility to a greater extent than variation in segmentation. A smaller number of pancreatic tumor-derived radiomic features were reproducible compared with pancreatic parenchyma-derived radiomic features under the same conditions.
• For pancreatic-derived radiomic features from contrast-enhanced CT (CECT), fewer than 25% are reproducible (with a threshold of CCC < 0.9) in a clinical heterogeneous dataset.
• Variations between CECT scans affected the number of reproducible radiomic features to a greater extent than variations in radiologist segmentation.
• A smaller number of pancreatic tumor-derived radiomic features were reproducible compared with pancreatic parenchyma-derived radiomic features under the same conditions.
KeywordsReproducibility of results Pancreatic ductal carcinoma X-ray computed tomography Radiomics Texture analysis
Abbreviations and acronyms
Angle co-occurrence matrix
Concordance correlation coefficient
Contrast-enhanced computed tomography
Digital Imaging and Communications in Medicine
Gray-level co-occurrence matrix
Intraclass correlation coefficient
Local binary pattern
Pancreatic ductal adenocarcinoma
Run length matrix
We thank Luz Adriana Escobar Hoyos and Juliana Brooke Schilsky for their support in figure creation and Joanne Chin for editorial assistance.
This study was funded in part through the 2016 Society of Abdominal Radiology Wylie J. Dodds Research Award, the National Institutes of Health/National Cancer Institute Cancer Center Support Grant P30 CA008748, and the Japan Society Promotion of Science Overseas Research Fellowships JSPS/OT/290125.
Compliance with ethical standards
The scientific guarantor of this publication is Richard K. G. Do, MD, PhD.
Conflict of interest
The authors declare that they have no conflict of interest.
Statistics and biometry
Two of the authors (Joanne Chou and Mithat Gonen, PhD) have significant statistical expertise.
Written informed consent was waived by the Institutional Review Board.
Institutional Review Board approval was obtained.
• cross-sectional study
• performed at one institution
- 6.Sandrasegaran K, Lin Y, Asare-Sawiri M et al (2019) CT texture analysis of pancreatic cancer. Eur Radiol 29(3):1067–1073. https://doi.org/10.1007/s00330-018-5662-1
- 9.Kim BR, Kim JH, Ahn SJ et al (2019) CT prediction of resectability and prognosis in patients with pancreatic ductal adenocarcinoma after neoadjuvant treatment using image findings and texture analysis. Eur Radiol 29(1):362–372. https://doi.org/10.1007/s00330-018-5574-0
- 12.Berenguer R, Pastor-Juan MDR, Canales-Vázquez J et al (2018) Radiomics of CT features may be nonreproducible and redundant: influence of CT acquisition parameters. Radiology 288(2):407–415. https://doi.org/10.1148/radiol.2018172361
- 13.Perrin T, Midya A, Yamashita R et al (2018) Short-term reproducibility of radiomic features in liver parenchyma and liver malignancies on contrast-enhanced CT imaging. Abdom Radiol (NY) 43(12):3271–3278. https://doi.org/10.1007/s00261-018-1600-6
- 19.Pavic M, Bogowicz M, Würms X et al (2018) Influence of inter-observer delineation variability on radiomics stability in different tumor sites. Acta Oncol 57(8):1070–1074. https://doi.org/10.1080/0284186X.2018.1445283
- 20.Mingqiang Y, Kidiyo K, Joseph R (2008) A survey of shape feature extraction techniques. Peng-Yeng Yin. Pattern Recognition, IN-TECH, pp 43–90. https://doi.org/10.5772/6237
- 27.Mehta R, Egiazarian K (2013) Rotated local binary pattern (RLBP): rotation invariant texture descriptor. In 2nd International Conference on Pattern Recognition Applications and Methods, ICPRAM 2013, Barcelona, Spain, 15.-18.2.2013, pp 497-502. (International Conference on Pattern Recognition Applications and Methods). Institute of Electrical and Electronics Engineers IEEEGoogle Scholar
- 29.Costa AF, Humpire-Mamani G, Traina AJM (2012) An efficient algorithm for fractal analysis of textures. 25th SIBGRAPI Conference on Graphics, Patterns and Images, Ouro Preto, pp 39–46. https://doi.org/10.1109/SIBGRAPI.2012.15
- 31.Chakraborty J, Rangayyan RM, Banik S, et al (2012) Detection of architectural distortion in prior mammograms using statistical measures of orientation of texture. Proc. SPIE 8315, Medical Imaging 2012: Computer-Aided Diagnosis, 831521 (23 February 2012). https://doi.org/10.1117/12.910937
- 32.Chakraborty J, Midya A, Mukhopadhyay S, Sadhu A (2013) Automatic characterization of masses in mammograms. 6th International Conference on Biomedical Engineering and Informatics, Hangzhou, pp 111–115. https://doi.org/10.1109/BMEI.2013.6746917
- 34.ISO 5725-2:1994 - Accuracy (trueness and precision) of measurement methods and results—part 2: basic method for the determination of repeatability and reproducibility of a standard measurement method. https://www.iso.org/standard/11834.html. Accessed 14 Dec 2018
- 39.Zwanenburg A, Leger S, Vallières M, et al (2016) Image biomarker standardisation initiative. arXiv:1612.07003v7 [cs.CV]Google Scholar