Cardiovascular system is of great importance to human body, the corresponding cardiovascular diseases (CVDs) are highly threatening people’s health and life. CVD is one of the leading causes of death in the world nowadays. Consequently, helpful to learn the health status of cardiovascular system, the corresponding monitoring schemes, which are based on the noninvasive and invasive comprehensive intelligent cardiovascular monitoring, are of the urgent need to depress the intimidation from CVD. Adapting the noninvasive and invasive comprehensive intelligent CVDs diagnoses in e-home healthcare can guarantee high-precise diagnostic accuracy. The proposed intelligent diagnosis system includes three entities, namely upstream (hospital server), midstream (local computers), and downstream (embedded-link devices, such as smartphone, iPad, etc.), which can satisfy users with diversified backgrounds and requirements. To achieve high-precise diagnostic accuracy, the integration of noninvasive e-home healthcare with invasive clinical diagnosis must be established. The clinical diagnosis actually maps the sampled cardiovascular parameters (CPs) into the relevant CVDs through using medical knowledge and depending on pathological characters of the disease. The CPs include noninvasive physical and invasive chemical information. At the aid of pervasive body sensor network, multi noninvasive signals such as electrocardiography (ECG), sphygmography (SPG), and heart sound (HS) waveforms can be obtained. Through signal pre- and post-processing, the sampled waveforms are converted into noninvasive parameters (NiP). With the aid of hospital clinical biochemical detection, multi invasive blood test parameters (BTPs) such as total cholesterol (TCH) and aspartate (AST) etc., which are a-part-of invasive parameters (InP), can be obtained. Noninvasive or invasive diagnoses are actually to infer the CVDs based on patients’ NiP or InP individually. The unit, scale, value-range, describing method, and storage method of these two types of parameters are quite different; consequently the knowledge representations and relevant inference machines are quite different. Tackle the bottleneck problem of designing a knowledge hierarchical representation (KHR) scheme, and expressing such complex and diversified data, many aspects must be considered systematically and scientifically. For instance, the confirmation of some diseases may refer to other new CPs such as creatine kinase (CK), or other relevant parameters such as brain wave parameters (BWPs), ultrasound parameters, lung-heart parameters, etc. Moreover, CVDs are divided into different categories with the allowance of producing new variants. For example, in the coronary heart disease (CHD) category it may produce a new disease called as latent coronary heart disease (LCH). Finally a frame-based KHR with re-usable shell structure is proposed which is expandable and adaptable for representing such complex and diversified data. With such an organic combination of noninvasive and invasive diagnosis, it establishes not only a solid mainstay for realizing more accurately the functionality of entire system, but also provides people with a reliable health monitor and management.