Computer-aided diagnosis for preoperative invasion depth of gastric cancer with dual-energy spectral CT imaging.

摘要: Rationale and Objectives This study evaluates the accuracy of dual-energy spectral computed tomography (DEsCT) imaging with aid computer-aided diagnosis (CAD) system in assessing serosal invasion patients gastric cancer. Materials Methods Thirty cancer were enrolled this study. Two types features (information) collected use DEsCT imaging: conventional including patient's clinical information (eg, age, gender) descriptive characteristics on CT images location lesion, wall thickness at cardia) additional extracted from monochromatic 60 keV) material-decomposition iodine- water-density images). The classification results CAD compared to pathologic findings. Important can be found out using support vector machine method combination feature-selection technique thereby helping radiologists diagnose better. Results Statistical analysis showed that for cases, feature “long axis” was significantly different between group A (serosa negative) B positive) ( P Conclusions designed machine-learning algorithms may used improve identification assessment provide some indicators which useful predicting prognosis.