Radiomics analysis of dual-layer spectral-detector CT-derived iodine maps for predicting tumor deposits in colorectal cancer.
Saved in:
| Title: | Radiomics analysis of dual-layer spectral-detector CT-derived iodine maps for predicting tumor deposits in colorectal cancer. |
|---|---|
| Authors: | Feng, Fei-Wen1 (AUTHOR), Jiang, Fei-Yu2 (AUTHOR), Liu, Yuan-Qing1,3 (AUTHOR), Sun, Qi4 (AUTHOR), Hong, Rong1 (AUTHOR), Hu, Chun-Hong1,3 (AUTHOR) sdhuchunhong@sina.com, Hu, Su1,3 (AUTHOR) husu@suda.edu.cn |
| Source: | European Radiology. Jan2025, Vol. 35 Issue 1, p105-116. 12p. |
| Subjects: | Receiver operating characteristic curves, Computed tomography, Radiomics, Support vector machines, Decision making |
| Abstract: | Objective: To investigate the value of radiomics analysis of dual-layer spectral-detector computed tomography (DLSCT)-derived iodine maps for predicting tumor deposits (TDs) preoperatively in patients with colorectal cancer (CRC). Materials and methods: A total of 264 pathologically confirmed CRC patients (TDs + (n = 80); TDs − (n = 184)) who underwent preoperative DLSCT from two hospitals were retrospectively enrolled, and divided into training (n = 124), testing (n = 54), and external validation cohort (n = 86). Conventional CT features and iodine concentration (IC) were analyzed and measured. Radiomics features were derived from venous phase iodine maps from DLSCT. The least absolute shrinkage and selection operator (LASSO) was performed for feature selection. Finally, a support vector machine (SVM) algorithm was employed to develop clinical, radiomics, and combined models based on the most valuable clinical parameters and radiomics features. Area under receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis were used to evaluate the model's efficacy. Results: The combined model incorporating the valuable clinical parameters and radiomics features demonstrated excellent performance in predicting TDs in CRC (AUCs of 0.926, 0.881, and 0.887 in the training, testing, and external validation cohorts, respectively), which outperformed the clinical model in the training cohort and external validation cohorts (AUC: 0.839 and 0.695; p: 0.003 and 0.014) and the radiomics model in two cohorts (AUC: 0.922 and 0.792; p: 0.014 and 0.035). Conclusion: Radiomics analysis of DLSCT-derived iodine maps showed excellent predictive efficiency for preoperatively diagnosing TDs in CRC, and could guide clinicians in making individualized treatment strategies. Clinical relevance statement: The radiomics model based on DLSCT iodine maps has the potential to aid in the accurate preoperative prediction of TDs in CRC patients, offering valuable guidance for clinical decision-making. Key Points: Accurately predicting TDs in CRC patients preoperatively based on conventional CT features poses a challenge. The Radiomics model based on DLSCT iodine maps outperformed conventional CT in predicting TDs. The model combing DLSCT iodine maps radiomics features and conventional CT features performed excellently in predicting TDs. [ABSTRACT FROM AUTHOR] |
| Copyright of European Radiology is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Objective: To investigate the value of radiomics analysis of dual-layer spectral-detector computed tomography (DLSCT)-derived iodine maps for predicting tumor deposits (TDs) preoperatively in patients with colorectal cancer (CRC). Materials and methods: A total of 264 pathologically confirmed CRC patients (TDs + (n = 80); TDs − (n = 184)) who underwent preoperative DLSCT from two hospitals were retrospectively enrolled, and divided into training (n = 124), testing (n = 54), and external validation cohort (n = 86). Conventional CT features and iodine concentration (IC) were analyzed and measured. Radiomics features were derived from venous phase iodine maps from DLSCT. The least absolute shrinkage and selection operator (LASSO) was performed for feature selection. Finally, a support vector machine (SVM) algorithm was employed to develop clinical, radiomics, and combined models based on the most valuable clinical parameters and radiomics features. Area under receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis were used to evaluate the model's efficacy. Results: The combined model incorporating the valuable clinical parameters and radiomics features demonstrated excellent performance in predicting TDs in CRC (AUCs of 0.926, 0.881, and 0.887 in the training, testing, and external validation cohorts, respectively), which outperformed the clinical model in the training cohort and external validation cohorts (AUC: 0.839 and 0.695; p: 0.003 and 0.014) and the radiomics model in two cohorts (AUC: 0.922 and 0.792; p: 0.014 and 0.035). Conclusion: Radiomics analysis of DLSCT-derived iodine maps showed excellent predictive efficiency for preoperatively diagnosing TDs in CRC, and could guide clinicians in making individualized treatment strategies. Clinical relevance statement: The radiomics model based on DLSCT iodine maps has the potential to aid in the accurate preoperative prediction of TDs in CRC patients, offering valuable guidance for clinical decision-making. Key Points: Accurately predicting TDs in CRC patients preoperatively based on conventional CT features poses a challenge. The Radiomics model based on DLSCT iodine maps outperformed conventional CT in predicting TDs. The model combing DLSCT iodine maps radiomics features and conventional CT features performed excellently in predicting TDs. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 09387994 |
| DOI: | 10.1007/s00330-024-10918-x |
