Diffusion-weighted Magnetic Resonance Imaging and Receiver Operating Characteristic (ROC) Analysis for Lung Cancer Diagnosis
- 10.2991/itms-15.2015.163How to use a DOI?
- Lung cancer; magnetic resonance imaging; diffusion-weighted imaging; receiver operating characteristic analysis; diagnostic accuracy assessment.
Diffusion-weighted magnetic resonance (DWI-MR) technique is a promising imaging method for lung cancer diagnosis. The purpose of this study is to investigate whether DWI-MR can differentiate lung cancer from infectious nodule accurately. Approved by institutional review board, this study included DWI-MR images of 38 patients who had solitary lung nodules in chest CT examination. Three different non-zero b-values, 600, 800, and 1000 s/mm2, were used for DWI-MR imaging and apparent diffusion coefficient (ADC) was calculated for lung cancer detection. Histopathological analysis of biopsy or anti-inflammation treatment results were used as the gold standard. Area under the receiver operating characteristic curve (AUC) was used as the figure of merit for diagnostic accuracy assessment. Results indicated that ADC for lung cancer was statistically significantly smaller than that for infectious nodules. The AUC values of ADC were 0.92±0.07, 0.82±0.09, and 0.79±0.10, for b-values of 600, 800, and 1000 s/mm2, respectively. We conclude that the DWI-MR imaging can differentiate lung cancer from infectious nodules accurately.
- © 2015, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Yahui Peng AU - Yongxia Lei AU - Xinchun Li AU - Houjin Chen PY - 2015/11 DA - 2015/11 TI - Diffusion-weighted Magnetic Resonance Imaging and Receiver Operating Characteristic (ROC) Analysis for Lung Cancer Diagnosis BT - Proceedings of the 2015 International Conference on Industrial Technology and Management Science PB - Atlantis Press SP - 691 EP - 694 SN - 2352-538X UR - https://doi.org/10.2991/itms-15.2015.163 DO - 10.2991/itms-15.2015.163 ID - Peng2015/11 ER -