An estimator that can estimate the blood flow rate through a rotary-type implantable ventricular assist device without invasive sensors was developed. This estimation algorithm uses the pump current and speed signals to approximate the blood flow rate based on the energy balance between the electric energy generated by the motor and the mechanical energy consumption, including the fluid load and mechanical losses. The corresponding estimator parameters were identified using experimental data in the laboratory. As described herein, the estimator performed well in both the mock circulatory loop experiments simulating different physiological conditions and the animal studies. For clinical applications, the estimated flow can be used to control the pump speed of the implantable device for optimizing the support of the device to the patient's physiological demand. In addition, it can be used to derive an index to monitor the patient's cardiac function continuously, considering the patient's heart might recover when it is offloaded by the device over time.