Mechanical loosening of total hip replacement (THR) is primarily diagnosed using radiographs, which are diagnostically challenging and require review by experienced radiologists and orthopaedic surgeons. Automated tools that assist less-experienced clinicians and mitigate human error can reduce the risk of missed or delayed diagnosis. Thus the purposes of this study were to: 1) develop an automated tool to detect mechanical loosening of THR by training a deep convolutional neural network (CNN) using THR x-rays, and 2) visualize the CNN training process to interpret how it functions. A retrospective study was conducted using previously collected imaging data at a single institution with IRB approval. Twenty-three patients with cementless primary THR who underwent revision surgery due to mechanical loosening (either with a loose stem and/or a loose acetabular component) had their hip x-rays evaluated immediately prior to their revision surgery (32 “loose” x-rays). A comparison group was comprised of 23 patients who underwent primary cementless THR surgery with x-rays immediately after their primary surgery (31 “not loose” x-rays). Fig. 1 shows examples of “not loose” and “loose” THR x-ray. DenseNet201-CNN was utilized by swapping the top layer with a binary classifier using 90:10 split-validation [1]. Pre-trained CNN on ImageNet [2] and not pre-trained CNN (initial zero weights) were implemented to compare the results. Saliency maps were implemented to indicate the importance of each pixel of a given x-ray on the CNN's performance [3].INTRODUCTION
METHODS
Joint replacement is one of the most common orthopaedic procedures, with over 2 million surgeries performed each year across the globe. Loss of implant fixation, or aseptic loosening, is the leading cause of revision following primary joint replacement, accounting for ∼25% of all revision cases [1]. However, diagnosis of aseptic loosening and its underlying causes remain challenging due to the low sensitivity and specificity of plain radiographs. To address this, we propose a novel approach inspired by [2] involving the use of a self-sensing bone cement (by imparting strain-dependent electrical conductivity or piezoresistivity) combined with electrical impedance tomography (EIT). Piezoresistivity is imparted to cement via incorporation of micro/nanoscale conductive fillers. Therefore mechanical effects such as loosening and cracks will manifest as a conductivity change of the cement. This work explores if EIT is able to detect strains and cracks within the bone cement volume. Experiments were designed to determine whether EIT combined with piezoresistive cement can be used to detect strains and cracks (Fig. 1). The setup consists of a tank filled with water, 16 electrodes, sample, a loading machine (MTS), and an EIT system. To develop the piezoresistive bone cement, microscale carbon fibers were used with varying CF/PMMA volumetric ratios (VR) from VR = 0.25% to 3.0%. Three conical samples were made to model a loading condition similar to knee implants (Fig. 1). The samples were compressed while the conductivity map of the tank was measured with the EIT system.INTRODUCTION
METHODS
In native knees anterior cruciate ligament (ACL) and asymmetric shape of the tibial articular surface with a convex lateral plateau are responsible for differential medial and lateral femoral rollback. Contemporary ACL retaining total knee arthroplasty (TKA) improves knee function over ACL sacrificing (CR) TKA; however, these implants do not restore the asymmetric tibial articular geometry. This may explain why ACL retention addresses paradoxical anterior sliding seen in CR TKA, but does not fully restore medial pivot motion. To address this, an ACL retaining biomimetic implant, was designed by moving the femoral component through healthy in vivo kinematics obtained from bi-planar fluoroscopy and sequentially removing material from a tibial template. We hypothesized that the biomimetic articular surface together with ACL preservation would better restore activity dependent kinematics of normal knees, than ACL retention alone. Kinematic performance of the biomimetic BCR design (asymmetric tibia with convex lateral surface), a contemporary BCR implant (symmetric shallow dished tibia) and a contemporary CR implant (symmetric dished tibia) was analyzed using KneeSIM software. Chair-sit, deep knee bend, and walking were analyzed. Components were mounted on an average bone model created from magnetic resonance imaging (MRI) data of 40 normal knees. Soft-tissue insertions were defined on the average knee model based on MRI data, and mechanical properties were obtained from literature. Femoral condyle center motions relative to the tibia were tracked to compare different implant designs.INTRODUCTION
METHODS
Dual Mobility (DM) implants have gained popularity for the treatment and prevention of hip dislocation, with increased stability provided by a large diameter mobile liner. However, distal regions of the liner can impinge on soft-tissues like hip capsule and iliopsoas, leading to anterior hip pain. Additionally, soft-tissue impingement may trap the mobile liner, leading to excessive loading of the liner rim, from engagement with the femoral stem, and subsequent intra-prosthetic dislocation. The hypothesis of this study was that reducing the liner profile below the equator (contoured design) can mitigate soft-tissue impingement without compromising inner-head pull-out resistance and overall hip joint stability ( The interaction of conventional and contoured liners with anterior soft-tissues was evaluated in 10 cadaveric hips (5 specimens; 2 male, 3 female; age 65 ± 10 yrs; liner diameter 42–48mm) via visual observation and fluoroscopic imaging. A metal wire was sutured to the deep fibers of the iliopsoas tendon/muscle, and metal wires were embedded in the mobile liners for fluoroscopic visualization Resistance to inner-head pull-out was evaluated via Finite Element Analysis (FEA) by simulating a full cycle of insertion of the inner head into the mobile liner and subsequent pullout. The femoral head, acetabular shell, and stem were modeled as rigid, while the mobile liner was modeled as plastically deformable. Hip joint stability was evaluated by dynamic simulations in for two dislocation modes: (A) Posterior dislocation (at 90° hip flexion) with internal hip rotation; (B) Posterior dislocation (starting at 90° flexion) with combined hip flexion and adduction. A 44 mm diameter conventional and a 44 mm contoured liner were evaluated during these tests.Introduction
Methods
Dual-mobility (DM) liners have increased popularity due to the range of motion and stability provided by these implants. However, larger head diameters have been associated with anterior hip pain, due to surrounding soft-tissue impingement, particularly the iliopsoas. To address this, an anatomically contoured dual mobility (ACDM) liner was designed by reducing the volume of the liner below the equator (Fig1). Previous cadaver studies have shown that the ACDM significantly reduces iliopsoas tenting and trapping of the liner compared to conventional designs. We created a finite element study based on previous cadaver testing to further analyze the effectiveness of the ACDM design in reducing soft-tissue impingement, specifically the tendon-liner contact pressure and the tendon stress. The finite element model was developed within COMSOL 4.3b. The psoas tendon was modelled as a Yeoh hyper-elastic Material, which uses 3 constants (c1-c3), density (1.73g/cm3) and a bulk modulus (26GPa)[Hirokawa,2000]. In a previous, separate study, the average stiffness of 10 psoas tendon samples (5 cadavers), were measured to be 339[N/mm] in the linear region with average width and thickness of 14mmX4mm. The 3 constants were tuned to match experimental uniaxial test data, and were 5[GPa], 0[Gpa], and 46[GPa] for c1, c2, and c3 respectively. The implant components were rigidly modeled relative to the psoas. Cadaver specific CT models were used to create the FEA geometry. The insertion points for the Psoas were digitally determined on the proximal end of the lesser trochanter, and the psoas notch on the pelvis for hip flexion angles of −15°, 0°, 15° and 30°. These insertion points determined the length of the psoas and its relative position to the femoral head in 3D. The specific liner size and position for each cadaver was determined by implant planning with the CT models. In this abstract, we only present data for 2 specimens (left/right hips) with 44mm conventional DM, and 44mm ACDM, matching specimen anatomy. A 500N tensile load was applied to the psoas tendon proximally to simulate moderate physiological loading, the average/max stresses and contact pressures between the psoas and the two liner designs were determined.Introduction
Methods
The large diameter mobile polyethylene liner of the dual mobility implant provides increased resistance to hip dislocation. However, a problem specific to the dual mobility system is intra-prosthetic dislocation (IPD), secondary to loss of the retentive rim, causing the inner head to dissociate from the polyethylene liner. We hypothesized that impingement of the polyethylene liner with the surrounding soft-tissue inhibits liner motion, thereby facilitating load transfer from the femoral neck to the liner and leading to loss of retentive rim over time. This mechanism of soft-tissue impingement with the liner was evaluated via cadaver experiments, and retrievals were used to assess polyethylene rim damage. Total hip arthroplasty was performed on 10 cadaver hips using 3D printed dual mobility components. A metal wire was sutured to the posterior surface (underside) of the iliopsoas, and metal wires were embedded into grooves on the outer surface of the liner and inner head to identify these structures under fluoroscopy. Tension was applied to the iliopsoas to move the femur from maximum hyperextension to 90° of flexion for the purpose of visualizing the iliopsoas and capsule interaction with the mobile liner. The interaction of the mobile liner with the iliopsoas was studied using fluoroscopy and direct visual observation. Fifteen retrieved dual mobility liners were assessed for rim edge and rim chamfer damage. Rim edge damage was defined as any evidence of contact, and rim chamfer damage was classified into six categories: impact ribs on the chamfer surface, loss of machining marks, scratching or pitting, rim deformation causing a raised lip, a rounded rim edge, or embedded metal debris.Introduction
Methods
In native knees the anterior cruciate ligament (ACL) plays a major role in joint stability and kinematics. Sacrificing the ACL in contemporary total knee arthroplasty (TKA) is known to cause abnormal knee motion, and reduced function. Hence, there is growing interest in the development of ACL retaining TKA implants. Accommodation of ACL insertion around the tibial eminence is a challenge with these designs. Therefore, a reproducible and practical test setup is necessary to characterize the strength of the ACL/bone construct in ACL retaining implants. Seminal work showed importance of loading the ACL along its anatomical orientation. However, prior setups designed for this purpose are complex and difficult to incorporate into a standardized test for wide adoption. The goal of this study was to develop a standardized and anatomically relevant test setup for repeatable strength assessment of ACL construct using basic force-displacement testing equipment. Cadaver knees were positioned with the ACL oriented along the loading axis and being the only connection between femur and tibia. 15° knee flexion was selected based on highest ACL tensions reported in literature. Therefore, the fixtures were adjusted accordingly to retain 15° knee flexion when the ACL was tensioned. The test protocol included 10 cycles of preconditioning between 6N and 60N at 1mm/s, followed by continuous distraction at 1mm/s until failure (INTRODUCTION
METHODS
Mechanical tissue properties of some ligaments and tendons have been described in the literature. However, to our knowledge no data exists describing the tensile properties of the Iliopsoas tendon. The iliopsoas complex is in very close proximity to the hip joint running through the psoas notch from the inner side of the pelvis to the lesser trochanter on the posterior aspect of the proximal femur. The tendon muscle complex wraps around the anterior aspect of the femoral head. Hip joint intervention such as total hip arthroplasty (THA) can interfere with iliopsoas function and contact mechanics, and thereby play a major role in the clinically known condition of anterior hip pain. For computer simulations such as finite element analysis (FEA) precise knowledge of soft-tissue mechanical properties is crucial for accurate models and therefore, the goal of this study was to describe the iliopsoas tensile properties using uniaxial testing equipment. Ten iliopsoas tendons were harvested from five specimens (2 male, 3 female; 82.4 yrs ±7.4 yrs) and then carefully cleaned from any fat and muscle tissue. Two freeze clamps were fixed to each end of the tendon sample. The clamps were submerged in liquid nitrogen for 30 seconds to prevent tendon slip and attached to the test frame and load cell via carabiners allowing the tendon to rotate around its long axis. Width, thickness and initial gauge length of each tendon were measured before testing. The test protocol included 10 cycles of preconditioning between 6 N and 60 N at 0.4 mm/s, followed by continuous distraction at 0.4 mm/s until failure. For each tendon the linear stiffness was determined by fitting a straight line to the liner region on the force-displacement curve (INTRODUCTION
METHODS
