Abstract: The method to diagnose intelligently abnormal information from heart rate by BCG signal ( ballistocardiogram signal) based on cloud model was studied. The qualitative expert diagnosis system and quantitative computer aided diagnosis system were combined by using the cloud model， and the expert diagnosis process was simulated. A cloud model for abnormal heart rate intelligent diagnosis was established. Then， distribution curve diagram of JJ interval in BCG signal was constructed， and the model parameters were adjusted automatically to establish the intelligent diagnosis mechanism. The signal acquisition system in our laboratory was used to extract 2000 groups of BCG signal as the sample object， the feasibility of the proposed method was verified by comparing with the traditional standard threshold method. The experiment results showed that the automatic clustering accuracy of the proposed method could reach 90.2%， which was 2 percentage points higher than that of the traditional method.

Key words: BCG signal, intelligent diagnosis system, cloud model, automatic clustering, abnormal heart rate