Polyethylene wear represents a significant risk factor for the long-term success of knee arthroplasty [1]. This work aimed to develop and in vivo validate an automated algorithm for accurate and precise AI based wear measurement in knee arthroplasty using clinical AP radiographs for scientifically meaningful multi-centre studies.

Twenty postoperative radiographs (knee joint AP in standing position) after knee arthroplasty were analysed using the novel algorithm. A convolutional neural network-based segmentation is used to localize the implant components on the X-Ray, and a 2D-3D registration of the CAD implant models precisely calculates the three-dimensional position and orientation of the implants in the joint at the time of acquisition. From this, the minimal distance between the involved implant components is determined, and its postoperative change over time enables the determination of wear in the radiographs.

The measured minimum inlay height of 335 unloaded inlays excluding the weight-induced deformation, served as ground truth for validation and was compared to the algorithmically calculated component distances from 20 radiographs.

With an average weight of 94 kg in the studied TKA patient cohort, it was determined that an average inlay height of 6.160 mm is expected in the patient. Based on the radiographs, the algorithm calculated a minimum component distance of 6.158 mm (SD = 81 µm), which deviated by 2 µm in comparison to the expected inlay height.

An automated method was presented that allows accurate and precise determination of the inlay height and subsequently the wear in knee arthroplasty based on a clinical radiograph and the CAD models. Precision and accuracy are comparable to the current gold standard RSA [2], but without relying on special radiographic setups. The developed method can therefore be used to objectively investigate novel implant materials with meaningful clinical cohorts, thus improving the quality of patient care.

Current implant designs and materials provide a high grade of quality and safety, but aseptic implant loosening is still the main reason for total hip revision. Highly cross-linked polyethylene (HX-PE) is used successfully in total hip replacements (THR) since several years. The good wear properties lead to a reduction of wear debris and may contribute to a longer survival time of the THRs. Furthermore, thin HX-PE liner allows the use of larger femoral heads associated with a decreased risk of dislocation and an improved range of motion. However, the cross-linking process is associated with a loss of mechanical properties of the polyethylene material which compromise the use of thin HX-PE liner in terms of high stress situations.

The aim of the present study was the experimental wear analysis of HX-PE liner under steep acetabular cup position. Furthermore, a finite element analysis (FEA) was performed in order to calculate the stress within the HX-PE material in case of steep cup position under physiological loading.

Experimental wear testing was performed for 5 Mio load cycles, using highly cross-linked polyethylene (HX-PE) acetabular liner combined with 44 mm ceramic femoral heads at a standard position of the acetabular cup (30° inclination) according to ISO 14242 as well as at 60° cup inclination. The wall thickness of the HX-PE liner was 3.8 mm. A hip wear simulator, according to ISO 14242 (EndoLab GmbH, Rosenheim, Germany), was used and wear was determined gravimetrically. Moreover, finite element models of the THR system at standard and steep cup position was created by Abaqus/CAE (Dessault Systemes Providence, USA). Using the finite element software Abaqus (Dessault Systemes Providence, USA) the total hip implants were physiologically loaded with maximum force of the gait cycle (3.0 kN). Thereby, the stresses within the HX-PE material were analysed.

The average gravimetrical wear rates of the HX-PE liners at standard implant position (30°) and 60° cup inclination showed small wear amounts of 3.15 ± 0.32 mg and 1.92 ± 1.00 mg per million cycles, respectively. The FEA revealed a clear increase of stresses at the HX-PE liner with respect to steep cup position (von Mises stress of 8.78 MPa) compared to ISO standard implant position (von Mises stress of 5.70 MPa).

The wear simulator tests could not demonstrate significant differences of gravimetrical wear amount of HX-PE liners under steep hip cup position compared to standard implant position. The small contact surface between the femoral head and the SX-PE liner during the wear testing may lead to the low wear rate of the misaligned acetabluar cup. Moreover, the FEA showed that the effect of a misaligned acetabular cup on the stresses within the polyethylene liner can be critical. Although an increase of wear could not be detected a steeper acetabular cup position using thin HX-PE liners should be avoided due to higher stresses preventing implant failure in clinical application.

Clinically applied methods of assessing implant fixation and implant loosening are of sub-optimal precision, leading to the risk of unsecure indication of revision surgery and late recognition of bone defects. Loosening diagnosis involving measuring the eigenfrequencies of implants has its roots in the field of dentistry. The changing of the eigenfrequencies of the implant-bone-system due to the loosening state can be measured as vibrations or structure-borne sound. In research, vibrometry was studied using an external shaker to excite the femur-stem-system of total hip replacements and to measure the resulting frequencies by integrated accelerometers or by ultrasound. Since proper excitation of implant components seems a major challenge in vibrometry, we developed a non-invasive method of internal excitation creating an acoustic source directly inside the implant.

In the concept proposed for clinical use, an oscillator is integrated in the implant, e.g. the femoral stem of a total hip replacement. The oscillator consists of a magnetic or magnetisable spherical body which is fixed on a flat steel spring and is excited electromagnetically by a coil placed outside the patient. The oscillator impinges inside the implant and excites this to vibrate in its eigenfrequency. The excitation within the bending modes of the implant leads to a sound emission to the surrounding bone and soft tissue. The sound waves are detected by an acoustic sensor which is applied on the patient's skin. Differences in the signal generated result from varying level of implant fixation.

The sensor principle was tested in porcine foreleg specimens with a custom-made implant. Influence of the measurement location at the porcine skin and different levels of fixation were investigated (press-fit, slight loosening, advanced loosening) and compared to the pull-out strength of the implant. Evaluation of different parameters, especially the frequency spectrum resulted in differences of up to 12% for the comparison between press-fit and slight loosening, and 30% between press-fit and advanced loosening. A significant correlation between the measured frequency and the pull-out strength for different levels of fixation was found.

Based on these findings, an animal study with sensor-equipped bone implants was initiated using a rabbit model. The implants comprised an octagonal cross-section and were implanted into a circular drill hole at the distal femur. Thereby, definite gaps were realized between bone and implant initially. After implantation, the bone growth around the implant started and the gaps were successively closed over postoperative period. Consequently, since the tests had been started with a loose implant followed by its bony integration, a reverse loosening situation was simulated. In weekly measurements of the eigenfrequencies using the excitation and sensor system, the acoustic signals were followed up. Finally, after periods of 4 and 12 weeks after implantation, the animals were sacrificed and pull-out tests of the implants were performed to measure the implant fixation. The measured implant fixation strengths at the endpoint of each animal trial were correlated with the acoustic signals recorded.