Analysis of Key Factors for Radiomic Feature Extraction Stability and Robustness on 4DCT Image
X. Li,Q. Deng,S. Ma,L. Zhang,Y. Kuang
DOI: https://doi.org/10.1016/j.ijrobp.2020.07.246
2020-01-01
Abstract:The precise radiomics analysis on thoracic 4DCT data is easily compromised by the respiratory motion and CT scan parameter setting, thus leading to potentially overfit and/or misinterpret the data in AI-enabled therapeutic model building. In this study, we investigated the optimal respiratory amplitudes and frequencies and CT scan pitch within the thoracic 4DCT scan for robust radiomics feature selection. A QUSAR lung tumor phantom was used to simulate different respiratory amplitudes and frequencies and CT scan pitches. A static CT images of the phantom acquired were used as a reference. The range of breathing amplitudes was mimicked at 3mm at left and right direction and 3mm - 15 mm at the craniocaudal direction with an interval of 2 mm. The respiratory frequencies were set at 10, 11, 12, 13, 14, 15 and 20 beats per minute (BPMs), respectively. The CT scan pitches were selected based on a procedure described in Med. Phys. 30(1):88-97. The pairwise Concordance Correlation Coefficient (CCC) was used to determine the agreement in feature values between 1766 radiomics features extracted from the images acquired under different combinations of respiratory amplitudes and frequencies and CT scan pitches of 4DCT and those extracted from the static CT images. (1) When the breathing amplitudes were at 3, 5, 7, 9, 12 and 15mm, the maximum CCC index could be achieved at the reconstructed 4DCT phases of 60%, 70%, 30%, 20%, 60%∼70% and 10%, respectively. Under these six amplitudes, the maximum intensity projection (MIP) and average intensity projection (AIP) images reconstructed show mean CCC values of 0.7781 and 0.6695, respectively, in pairwise radiomics feature extraction comparison between 4DCT and static CT. (2) When the breathing amplitude was set at 12 mm, the maximum CCC index could be consistently achieved at the reconstructed 4DCT phase of 90% for the seven respiratory frequencies of 10, 11, 12, 13, 14, 15 and 20 BPMs. Under these respiratory states, the MIP and AIP images reconstructed show mean CCC values of 0.5624 and 0.8278, respectively. (3) When the breathing amplitude was set at 12 mm and the respiratory frequency was set at 10 BPM, the maximum CCC index could be obtained at the reconstructed 4DCT phase of 90% for all scan pitches used except the 0% phase which was obtained at 0.048 pitch setting. Under these CT scan pitch settings, the MIP and AIP images reconstructed show mean CCC values of 0.5575 and 0.7815, respectively. (4) The total number of robust features were 50, 34 and 35 with different respiratory amplitudes and phases and CT scanning pitches used (CCC values ≥ 0.99). In 4DCT, respiratory amplitude and frequency and CT scan pitch are three limiting factors that greatly affect the robustness of features. The reconstructed 4DCT phases with better robustness along with optimal respiratory amplitude and CT scan pitch determined could be used to guide the breathing training for patients with lung cancer for RT to improve the robust feature selection process.