CT和MR的形状和纹理特征可以鉴别胸膜炎的良恶性吗(英文)

2017-04-20 10:08:03 admin 5

Can CT and MR Shape and Textural Features Differentiate Benign Versus Malignant PleuralLesions?

Acad Radiol. 2017 Apr 20

Pena E1, Ojiaku M2, Inacio JR2, Gupta A2, Macdonald DB2, Shabana W2, Seely JM2, Rybicki FJ2, Dennie C2, Thornhill RE2.

Author information

    1.Department of Medical Imaging, Ottawa Hospital Research Institute, The Ottawa Hospital, 1053 Carling Avenue, Ottawa, ON K1Y E49, Canada; Department of Radiology, University of Ottawa, Ottawa, Ontario, Canada. 

    2.Department of Medical Imaging, Ottawa Hospital Research Institute, The Ottawa Hospital, 1053 Carling Avenue, Ottawa, ON K1Y E49, Canada; Department of Radiology, University of Ottawa, Ottawa, Ontario, Canada.

Abstract

RATIONALE AND OBJECTIVES


The study aimed to identify a radiomic approach based on CT and or magnetic resonance (MR) features(shape and texture) that may help differentiate benign versus malignant pleural lesions, and to assess if the radiomic model may improve confidence and accuracy of radiologists with different subspecialty backgrounds.

MATERIALS AND METHODS


Twenty-nine patients with pleural lesions studied on both contrast-enhanced CT and MR imaging were reviewed retrospectively. Three texture and three shape features were extracted. Combinations of features were used to generate logistic regression models using histopathology as outcome. Two thoracic and two abdominal radiologists evaluated their degree of confidence in malignancy. Diagnostic accuracy of radiologists was determined using contingency tables. Cohen's kappa coefficient was used to assess inter-reader agreement. Using optimal threshold criteria, sensitivity, specificity, and accuracy of each feature and combination of featureswere obtained and compared to the accuracy and confidence of radiologists.

RESULTS


The CT model that best discriminated malignant from benign lesions revealed an AUCCT = 0.92 ± 0.05 (P < 0.0001). The most discriminative MR model showed an AUCMR = 0.87 ± 0.09 (P < 0.0001). The CT model was compared to the diagnostic confidence of all radiologists and the model outperformed both abdominal radiologists (P < 0.002), whereas the top discriminative MR model outperformed one of the abdominal radiologists (P = 0.02). The most discriminative MR model was more accurate than one abdominal (P = 0.04) and one thoracic radiologist (P = 0.02).

CONCLUSION


Quantitative textural and shape analysis may help distinguish malignant from benign lesions. A radiomics-based approach may increase diagnostic confidence of abdominal radiologists on CT and MR and may potentially improve radiologists' accuracy in the assessment of pleural lesions characterized by MR.

KEYWORDS


Pleural disease; computed tomography; magnetic resonance; mesothelioma; radiomics


慧影医疗科技(北京)有限公司

地点:北京市海淀区中关村东升科技园B2-C103

电话:400-890-9020

邮箱:radcloud@huiyihuiying.com

关闭
图片
图片
  • 人工智能诊断云平台