Objectives: This study aimed to evaluate the feasibility of ultrasound as a noninvasive method to characterize bone to implant alterations. Our goal was to show that ultrasonic sensors allow identification of gaps at a much smaller size than conventional radiological methods, thereby enabling early detection of loosened implanted endoprostheses. Materials and methods: First, two idealized three layer systems, a cylindrical aluminium-water-aluminium system and a planar bone-water-metal system, were examined. Finally, the same measurements were carried out with a realistic three layer bone-water-implant system. In all of these three layer system measurements a transducer (C 384-SU, Olympus) that generated ultrasonic signals controlled by a function generator (Agilent 33521A) was used. The transmitted signal was fed to an oscilloscope (Teledyne LeCroy WaveRunner 604Zi) and a multiplexer. The multiplexer, developed at the Institute of Sensor and Actuator Technology (ISAT), separated the transmit signal from the receive signal to ensure accurate voltage measurements. Finally, the signals were transmitted to a computer via a network interface for further processing. Results: The measurement methods of this experimental study were applied to idealized test setups as well as a realistic bone-implant system. Before performing tests on a bone-implant setup, the developed measurement procedure was secured by careful validation. Idealized test systems (a cylindrical three layer system composed of aluminium-water-aluminium and a planar three layer system composed of bone- water-metal) were used to ensure the accuracy and reliability of the procedure. Water was used as an intermediate medium as its properties are similar to those of demineralized tissue types in the human body. The water-based test systems showed remarkable comparability with human tissue, resulting in accurate results that were transferable to the realistic implant setup. In the cylindrical setup, the varied interlayer thickness ranged from 0.0 to 1.0 mm, while in the planar setup the range was from 0.0 to 2.0 mm. A third more realistic system, a bone-implant system, was examined, in which the same measurements as with the previous two idealized systems were performed. The range of determined layer thicknesses extended from 210 m to 1519 m, thus covering almost the entire range from below 500 m to about 2 mm. The suitability of the measurement method for curved surfaces was successfully demonstrated. The precise measurements performed on realistic materials and curved surfaces met the requirements for potential applications and confirmed the 48 practical usefulness of the method. Conclusion: The developed method for measuring the bone-implant interface was tested using simplified test systems in the range of 200 ţm to 2 mm and then extended to more realistic bone-implant systems with distances from 200 ţm to 1.6 mm. The method was also validated for heterogeneous permeable materials with uneven surfaces, as found in clinical bone-implant systems. The results allow similar measurements on practical bone-implant systems that include soft tissues such as fat, muscle, and skin. Future studies could use the algorithm to characterize interlayer properties and detect biofilm formation. Further improvements to the analytical model and data processing are needed to minimize systematic inaccuracies.