Department of Radiology & Functional and Molecular Imaging Key Lab of Shaanxi Province, Tangdu Hospital, Air Force Medical University, 569 Xinsi Road, Xi'an, 710038, Shaanxi, People's Republic of China.
Student Brigade, Air Force Medical University, Xi'an, 710032, Shaanxi, China.
BMC Neurol. 2020 Feb 7;20(1):48. doi: 10.1186/s12883-020-1613-y.
The medical imaging to differentiate World Health Organization (WHO) grade II (ODG2) from III (ODG3) oligodendrogliomas still remains a challenge. We investigated whether combination of machine leaning with radiomics from conventional T1 contrast-enhanced (T1 CE) and fluid attenuated inversion recovery (FLAIR) magnetic resonance imaging (MRI) offered superior efficacy.
Thirty-six patients with histologically confirmed ODGs underwent T1 CE and 33 of them underwent FLAIR MR examination before any intervention from January 2015 to July 2017 were retrospectively recruited in the current study. The volume of interest (VOI) covering the whole tumor enhancement were manually drawn on the T1 CE and FLAIR slice by slice using ITK-SNAP and a total of 1072 features were extracted from the VOI using 3-D slicer software. Random forest (RF) algorithm was applied to differentiate ODG2 from ODG3 and the efficacy was tested with 5-fold cross validation. The diagnostic efficacy of radiomics-based machine learning and radiologist's assessment were also compared.
Nineteen ODG2 and 17 ODG3 were included in this study and ODG3 tended to present with prominent necrosis and nodular/ring-like enhancement (P < 0.05). The AUC, ACC, sensitivity, and specificity of radiomics were 0.798, 0.735, 0.672, 0.789 for T1 CE, 0.774, 0.689, 0.700, 0.683 for FLAIR, as well as 0.861, 0.781, 0.778, 0.783 for the combination, respectively. The AUCs of radiologists 1, 2 and 3 were 0.700, 0.687, and 0.714, respectively. The efficacy of machine learning based on radiomics was superior to the radiologists' assessment.
Machine-learning based on radiomics of T1 CE and FLAIR offered superior efficacy to that of radiologists in differentiating ODG2 from ODG3.
区分世界卫生组织（WHO）分级 II（ODG2）和 III（ODG3）少突胶质细胞瘤的医学影像学仍然是一个挑战。我们研究了机器学习与常规 T1 对比增强（T1CE）和液体衰减反转恢复（FLAIR）磁共振成像（MRI）的放射组学相结合是否能提供更好的效果。
回顾性招募了 2015 年 1 月至 2017 年 7 月期间接受组织学证实的 ODG 治疗且有 T1CE 检查和 33 例有 FLAIR MRI 检查的 36 例患者。使用 ITK-SNAP 在 T1CE 和 FLAIR 切片上手动绘制覆盖整个肿瘤增强区域的感兴趣区域（VOI），并使用 3D 切片器软件从 VOI 中提取总共 1072 个特征。随机森林（RF）算法用于区分 ODG2 和 ODG3，并使用 5 折交叉验证进行验证。还比较了基于放射组学的机器学习和放射科医生评估的诊断效能。
本研究纳入了 19 例 ODG2 和 17 例 ODG3，ODG3 倾向于表现出明显的坏死和结节状/环状强化（P<0.05）。T1CE 的 AUC、ACC、敏感性和特异性分别为 0.798、0.735、0.672、0.789，FLAIR 为 0.774、0.689、0.700、0.683，联合为 0.861、0.781、0.778、0.783。放射科医生 1、2 和 3 的 AUC 分别为 0.700、0.687 和 0.714。基于放射组学的机器学习的效果优于放射科医生的评估。
T1CE 和 FLAIR 的基于放射组学的机器学习在区分 ODG2 和 ODG3 方面优于放射科医生的评估。
