Total hip replacement is one of the standard procedures in orthopedic surgery. Due to various reasons revision surgery (RTHR) has to be performed. In case of the revision of a cemented prosthesis stem, the bone cement has to be removed from the femoral cavity. Conventionally the cement removal is done manually using a hammer, chisel or burr under X-ray control, causing a considerable radiation exposure for patient and the surgeon. Furthermore the risk of undesirable bone damage is high due to bad sight and access conditions, leading to complications and prolongation of the intervention. Different approaches addressing the mentioned problems were proposed, but did not achieve acceptance in clinical practice due to disadvantages concerning process controllability. Another possibility is to use a robot guided milling tool. However, to be able to control it typically a 3D reconstruction of the cement volume to be removed is necessary. Existing approaches use computed tomography based measurements combined with previously implanted markers, fluoroscopy or ultrasound based measurements, all requiring additional process steps prior to the surgery or to the actual cement removal. The ICOS project (Impedance Controlled Surgical Instrumentation, Chair of Medical Engineering, RWTH Aachen University) investigates the approach of electrical impedance controlled, robot assisted bone cement removal, based on real time cement detection during the removal process without radiation exposure or the necessity of prior imaging or marker implanting steps. Therefore the electrical impedance is measured between the milling head mounted on the surgical mini-robot MINARO and one or more electrodes attached to the skin of the patient's thigh. An impedance variation mainly results from decreasing thickness of bone cement near the milling head contact point due to material removal. Hence the proposed method does not generate a 3D volume allowing for a milling path generation prior to the process. It requires a strategy for real time path generation using only the limited local information. Up to now, only the differentiation between bone cement and bone, and thus the cement-bone interface breakthrough, is reliably detectable. To efficiently use this information for the tool path generation, generic a-priori knowledge of the bone cement shape after removal of the prosthesis stem is used. The concept for impedance controlled milling has been verified in first lab trials. For impedance measurements during the cement removal process the robots milling tool has been modified to achieve electrical insulation of the milling head. A strategy for online adaptive robot path planning has been implemented and tested in a Matlab/Simulink based process simulation. For all data sets a cement removal rate of about 90% with a bone removal of approximately 3% was achieved. These results confirm that it is generally possible to use only the limited local information for automated cement removal. Future work aims for a practical evaluation of the algorithm using real impedance measurement values. This work has been funded by the German Ministry for Education and Research (BMBF) in the framework of the ICOS project under grant No. BMBF 13EZ1005.
