Siebler M, Rose G, Sitzer M, Bender A, Steinmetz H
Department of Neurology, Heinrich-Heine University, Düsseldorf, Germany.
Radiology. 1994 Sep;192(3):739-42. doi: 10.1148/radiology.192.3.7914706.
Abnormal transcranial Doppler ultrasonographic (US) signals indicating cerebral microembolism have characteristic but complex features. The authors wanted to assess the agreement among human observers and test the feasibility of an automated detection system.
Automated on-line detection of cerebral microemboli was accomplished by employing real-time overlapping Fourier transform and artificial neural network technology. By using long-term transcranial Doppler US recordings of the middle cerebral artery in consecutive cerebrovascular and cardiac patients, the method was evaluated in a clinical setting.
The proportion of specific agreement (ps) among four experienced investigators identifying cerebral microemboli was high (mean ps, 0.91). Agreement among the neural network and the human observers was only slightly less (mean ps, 0.77).
The technique allows highly reliable on-line evaluation of transcranial Doppler US recordings across multiple centers. It obviates time-consuming analyses by human observers.
提示脑微栓塞的异常经颅多普勒超声（US）信号具有特征性但复杂的特点。作者想要评估人类观察者之间的一致性，并测试自动检测系统的可行性。
采用实时重叠傅里叶变换和人工神经网络技术实现脑微栓子的自动在线检测。通过对连续的脑血管和心脏疾病患者的大脑中动脉进行长期经颅多普勒超声记录，在临床环境中对该方法进行评估。
四位经验丰富的研究人员识别脑微栓子的特定一致性比例（ps）很高（平均ps，0.91）。神经网络与人类观察者之间的一致性略低（平均ps，0.77）。
该技术可对多个中心的经颅多普勒超声记录进行高度可靠的在线评估。它避免了人类观察者耗时的分析。
