基于Bayes理论的计算机辅助诊断系统在孤立性肺结节CT诊断中的应用.docVIP

基于Bayes理论的计算机辅助诊断系统在孤立性肺结节CT诊断中的应用 陈伟 刘进康 李文政 熊曾 龙学颖 周晖 （中南大学湘雅医院放射科，湖南长沙410008） 摘要： 目的 初步探讨基于Bayes理论的计算机辅助诊断（computer-aided diagnosis, CAD）系统在孤立性肺结节（solitary pulmonary nodule, SPN）CT诊断中的价值。方法 依据Bayes理论先从352例SPN训练集（恶性135例，良性217例）中求出恶性SPN的验前比及各临床和CT表现的似然比，再建立基于Bayes理论的CAD系统，用它计算每个SPN的恶性概率。并前瞻性地检验该系统在132例SPN测试集（恶性61例，良性71例）中的诊断效能，与2位高年资和2位低年资放射科医生常规阅片的表现作比较。结果 成功构建基于Bayes理论的CAD系统，它诊断训练集SPN的敏感度、特异度和符合率分别为88.9％、93.1％、91.5％，诊断测试集SPN的敏感度、特异度、符合率、阳性预测值及阴性预测值分别为88.5％、85.9％、87.1％、84.4％、89.7％，其诊断符合率与高年资甲、乙医生比较无统计学差异（P均 ＞0.05），但高于低年资丙、丁医生（P均 ＜0.05）。 结论 基于Bayes理论的CAD系统将是一个有效的诊断辅助工具，可以提高医生尤其是低年资医生对SPN良恶性质的正确判断，并在指导SPN的临床决策中有一定的参考作用。 关键词： 硬币病变，肺；体层摄影术，X线计算机；诊断，计算机辅助；Bayes理论 Application of Bayesian theory based computer-aided diagnosis system in the CT diagnosis of solitary pulmonary nodules CHEN Wei, LIU Jin-kang, LI Wen-zheng, XIONG Zeng, LONG Xue-ying, ZHOU Hui. (Department of Radiology, Xiangya Hospital, Central South University, Changsha 410008, China) Abstract Objective To preliminarily explore the value of computer-aided diagnosis (CAD)system based on Bayesian theory in the diagnosis of solitary pulmonary nodules (SPNs) with CT. Methods Three hundred and fifty-two consecutive SPN cases (malignancy n=135, benignity n=217) were collected retrospectively to form the training set. According to Bayesian theory, the prior odds of malignant SPNs and the likelihood ratios of clinical and CT findings were derived from the training set firstly, then a Bayesian theory based CAD system was constructed and the probability of malignancy in each SPN was obtained from this system. SPs with ≥50% calculated probability were judged as malignancy and those with 50% calculated probability was judged as benignity. On the test set (malignancy n=61, benignity n=71), the Bayesian theory based CAD system was tested prospectively for its diagnostic validation, compared with the performance of the two chest radiologists and two radiologic residents using routine diagnostic method. Results Th