Medical procedures such as RFA and cryosurgery require needle insertion, which is difficult because it can easily result in organs being deformed and displaced. In addition, Because deflection occurs more easily with thin needles, needle deflection must be considered. We developed an intelligent robot for needle insertion, incorporating visual feedback, force control, and organ-model-based control. Two experiments were evaluating hepatic properties for organ-model-based robot control. And a dynamic viscoelastic test was done to show dynamic hepatic properties as a differential equation. Their nonlinearity was supported by a creep test. And, this paper shows the deflection correction with (a) the force sensor only, (b) liver model only, (c) both force sensor and liver model is done to control the position of the needle tip. The experimental result shows that using (c) gives optimal effectiveness among the proposed approaches.