3D Knee Research Articles

Electroconductive, flexible, and printable graphene nanoplate-carbon nanotube-polydimethylsiloxane composite collectors for three-dimensional conformal electrospinning

The three-dimensional (3D) nanofibrous structure demonstrates significant potential in tissue engineering and regenerative medicine. As electrospinning is typically conducted on metal substrates, the integration of a nanofibrous structure with complex 3D non-conductive substrates remains challenging. Therefore, we suggest graphene nanoplate-carbon nanotube-polydimethylsiloxane (GCP) composites as a collector for 3D conformal electrospinning on non-conductive substrates. To prepare electroconductive, flexible, and printable GCP composite collectors, we formulate a GCP ink that is compatible with dip coating and printing processes. Using the GCP ink enables the facile and versatile fabrication of GCP composite collectors on non-conductive substrates, including 3D substrates. By dip coating and printing the GCP ink on a polydimethylsiloxane (PDMS) and water-soluble polyvinyl alcohol (PVA) substrate, 3D or patterned GCP-PDMS composite collectors and hollow GCP composite collectors by removing the PVA substrate are achieved, respectively. Subsequent electrospinning on the GCP-PDMS and hollow GCP composite collector fabricated spatially controlled nanofiber mats with 3D ear and knee cartilage shapes, patterned and aligned configurations, and 3D blood-vessel shape. Finally, as a potential application of artificial ear-cartilage implants, we demonstrate the enhanced biocompatibility and flexibility of 3D nanofiber mat-coated GCP-PDMS composites.

Poster 228: Comparison of Isometric Properties Between the Medial Patellofemoral Complex Fibers

Objectives:The medial patellofemoral complex (MPFC) has a broad attachment on the patella and quadriceps tendon. Recent anatomic studies have demonstrated a corresponding elongated footprint on the medial femur in which the proximal fibers constitute the MQTFL and the distal fibers form the MPFL. Reconstruction of the MPFL and MQTFL have traditionally been described using the midpoint of the femoral footprint (Schottle’s point) while varying the graft’s position on the patella or quadriceps tendon. The purpose of this study was to compare graft isometry of MPFL, MQTFL, and MPFC reconstruction when using a standard femoral tunnel versus the corresponding anatomic femoral tunnel, and to identify the morphologic risk factors that influence each reconstruction technique.Methods:3D digital knee models were created from dynamic CT images of asymptomatic knees of patients with contralateral patellar instability. Knee models were grouped into 10° intervals and graft length, or distance between the defined points on the femur and extensor mechanism were measured in the following conditions: 1) standard MQTFL (sMQTFL), between the midpoint of the femoral footprint and the quadriceps tendon attachment, 2) anatomic MQTFL (aMQTFL), between the proximal femoral footprint and quadriceps tendon attachment, 3) standard MPFL (sMPFL), between the femoral footprint midpoint and the patellar attachment, 4) anatomic MPFL (aMPFL), between the distal femoral footprint and patellar attachment, and 5) MPFC, between the midpoints of the femoral and quadriceps tendon/patellar attachments. Changes in graft length between 0°and 40° flexion were compared for each reconstruction technique and assessed for their relationship to morphologic risk factors using stepwise multiple regression analyses.Results:12 knees were included in this study (3M, 9F; mean age 23.1+/-8.0y). Mean graft length was significantly greater at 0° than 40° in all reconstructive techniques (p<0.001 to p=0.045). Graft length changes showed no significant differences between reconstructive techniques when using corresponding anatomic femoral tunnels (aMQTFL, aMPFL, aMPFC). When compared to aMPFC (8.8mm), graft length changes were greater in sMQTFL (11.3, p=0.005) and smaller in sMPFL (6.9, p=0.037); however 2/12 (16.7%) of the sMPFL had unfavorable length change patterns that were tighter in flexion than extension. Overall, proximal reconstructions (aMQTFL and sMQTFL) were the most sensitive to the presence of morphologic risk factors (R=.933, R2=0.87 and R=0.898,R2=0.81, respectively) with patella alta having the greatest effect. In contrast, the distal reconstructions (aMPFL and sMPFL) did not have a significant relationship with risk factors but trended toward a significant relationship with TTTG distance (p=0.072 and 0.064, respectively).Conclusions:For MQTFL and MPFL reconstructions, using the corresponding anatomic femoral attachment showed more favorable length change patterns over using the currently described standard femoral tunnel. Overall, graft length changes in proximal (quadriceps tendon) reconstructions were more sensitive to the presence of patella alta, while distal (patellar) reconstructions were influenced by TTTG distance. Further studies are needed to identify the optimal reconstructive technique based on each patient-specific morphologic patterns in the treatment of patellar instability.

A Comparison of Inertial Measurement Unit and Motion Capture Measurements of Tibiofemoral Kinematics during Simulated Pivot Landings.

Injuries are often associated with rapid body segment movements. We compared Certus motion capture and APDM inertial measurement unit (IMU) measurements of tibiofemoral angle and angular velocity changes during simulated pivot landings (i.e., ~70 ms peak) of nine cadaver knees dissected free of skin, subcutaneous fat, and muscle. Data from a total of 852 trials were compared using the Bland–Altman limits of agreement (LoAs): the Certus system was considered the gold standard measure for the angle change measurements, whereas the IMU was considered the gold standard for angular velocity changes. The results show that, although the mean peak IMU knee joint angle changes were slightly underestimated (2.1° for flexion, 0.2° for internal rotation, and 3.0° for valgus), the LoAs were large, ranging from 35.9% to 49.8%. In the case of the angular velocity changes, Certus had acceptable accuracy in the sagittal plane, with LoAs of ±54.9°/s and ±32.5°/s for the tibia and femur. For these rapid motions, we conclude that, even in the absence of soft tissues, the IMUs could not reliably measure these peak 3D knee angle changes; Certus measurements of peak tibiofemoral angular velocity changes depended on both the magnitude of the velocity and the plane of measurement.

The coronal inclination of the medial tibial plateau affects coronal gait kinematics for varus osteoarthritic knees.

This study aims to (1) measure the kinematics of lower extremity alignment and the bony position relative to the ground during walking, focusing on the coronal plane, and (2) determine the correlation between the kinematics and coronal inclination of the medial tibial plateau (coronal inclination) for healthy and varus knee osteoarthritis (OA). In this study, 43 women (non-OA, 9 knees; early OA, 13 knees; advanced OA, 21 knees; mean age 58 ± 17years) were examined. The knee phenotypes in varus knee OA were varied. Three-dimensional (3D) knee kinematics were calculated in gait analysis by combining the motion capture system and the 3D lower extremity alignment assessment system via biplanar long-leg X-rays, applying the 3D-2D registration technique. The main parameters were the kinematics of the bony axes relative to the ground in the coronal plane during the stance phase of the gait. The differences in overall kinematics were assessed using repeated measures ANOVA with Tukey's post hoc test. The association between kinematic parameters and coronal inclination was evaluated by multiple linear regression after univariate analysis. The tibia tilted laterally during the loading response, and a plateau area subsequently appeared until the terminal stance phase, whereas the femur slowly tilted laterally until the terminal stance phase. The dynamic alignment showed a relatively large varus angular change during the loading response in all groups. The trend of motion was similar among all groups (p = n.s.), although to varying degrees. The coronal inclination was the more dominant factor than the Kellgren-Lawrence (K-L) grades (β = - 0.423, p = 0.005) when the change in dynamic alignment was determined. The TAA plateau area after the loading response implies that the tibial articular surface may become horizontal. The femur slowly tilted laterally until the terminal stance phase in response to the tibial motion. Consequently, the dynamic alignment showed a varus angular change, in which coronal MCT was more involved than K-L grades.

A novel method for 3D knee anatomical landmark localization by combining global and local features

A novel method for 3D knee anatomical landmark localization by combining global and local features

A New Method for Testing the Breaking Force of a Polylactic Acid-Fabric Joint for the Purpose of Making a Protective Garment.

3D printing is a technology that is increasingly used in the individualization of clothing, especially in the construction of garments for people with disabilities. The paper presents a study on the use of 3D printed knee protectors intended for wheelchair users. Due to the specific purpose of this 3D printed object, the breaking force of the polylactic acid (PLA) combined with 100% cotton and 100% polyester fabric was investigated. This paper will also describe a new method for testing the breaking force of a 3D printed polymer (PLA) combined with an incorporated fabric. Test samples were made, and the input parameters used in 3D printing were defined for testing purposes. A 3D knee protector for wheelchair users was developed based on a digitized model of the human body. The durability of the shape of the 3D printed shield was also tested after washing at temperatures of 40 °C, 50 °C and 60 °C. A clothing model that provides adequate user protection was proposed based on the conducted research. A construction solution has been proposed that enables the application of a 3D printed individualized garment element.

The Permissive Safe Angle of the Tibial Tunnel in Transtibial Posterior Cruciate Ligament Reconstruction: A Three-Dimensional Simulation Study.

ObjectiveTo determine the permissive safe angle (PSA) of the tibial tunnel in transtibial posterior cruciate ligament (PCL) reconstruction based on a three‐dimensional (3D) simulation study.MethodsThis was a computer simulation study of transtibial PCL reconstruction using 3D knee models. CT images of 90 normal knee joints from 2017 to 2020 were collected in this study, and 3D knee models were established based on CT data. The tunnel approaches were subdivided into the anterior 1/3 of the anteromedial tibia (T1), middle 1/2 of the anteromedial tibia (T2), the tibial crest (T3), anterior 1/3 of the anterolateral tibia (T4), middle 1/2 of the anterolateral tibia (T5). Five tibial tunnels (T1–T5) were simulated on the 3D knee models. The PSAs, in different tibial tunnel approaches were measured, and subgroup analyses of sex, age and height were also carried out.ResultsThe mean PSAs of the tibial tunnels with 5 different approaches (T1–T5) were 58.49° ± 6.82°, 61.14° ± 6.69°, 56.12° ± 7.53°, 52.01° ± 8.89° and 49.90° ± 10.53°, respectively. The differences of the mean PSAs between the anteromedial and anterolateral approaches were significant (P < 0.05). However, there was no significant difference of the mean PSA value between the two anteromedial tibial tunnel approaches (T1–T2) (P > 0.05), as well as between the two anterolateral tibial tunnel approaches (T4–T5). The patient's anthropomorphic characteristics of sex, age, and height were not associated with the PSAs.ConclusionsThe PSA varied with the anteromedial, tibial crest and anterolateral approaches for transtibial PCL reconstruction, and surgeons should limit the PCL drill guide by referring to the specific PSA for different surgical approaches.

High fidelity deep learning-based MRI reconstruction with instance-wise discriminative feature matching loss.

To improve reconstruction fidelity of fine structures and textures in deep learning- (DL) based reconstructions. A novel patch-based Unsupervised Feature Loss (UFLoss) is proposed and incorporated into the training of DL-based reconstruction frameworks in order to preserve perceptual similarity and high-order statistics. The UFLoss provides instance-level discrimination by mapping similar instances to similar low-dimensional feature vectors and is trained without any human annotation. By adding an additional loss function on the low-dimensional feature space during training, the reconstruction frameworks from under-sampled or corrupted data can reproduce more realistic images that are closer to the original with finer textures, sharper edges, and improved overall image quality. The performance of the proposed UFLoss is demonstrated on unrolled networks for accelerated two- (2D) and three-dimensional (3D) knee MRI reconstruction with retrospective under-sampling. Quantitative metrics including normalized root mean squared error (NRMSE), structural similarity index (SSIM), and our proposed UFLoss were used to evaluate the performance of the proposed method and compare it with others. In vivo experiments indicate that adding the UFLoss encourages sharper edges and more faithful contrasts compared to traditional and learning-based methods with pure loss. More detailed textures can be seen in both 2D and 3D knee MR images. Quantitative results indicate that reconstruction with UFLoss can provide comparable NRMSE and a higher SSIM while achieving a much lower UFLoss value. We present UFLoss, a patch-based unsupervised learned feature loss, which allows the training of DL-based reconstruction to obtain more detailed texture, finer features, and sharper edges with higher overall image quality under DL-based reconstruction frameworks. (Code available at: https://github.com/mikgroup/UFLoss).

DADP: Dynamic abnormality detection and progression for longitudinal knee magnetic resonance images from the Osteoarthritis Initiative.

Osteoarthritis (OA) is the most common disabling joint disease. Magnetic resonance (MR) imaging has been commonly used to assess knee joint degeneration due to its distinct advantage in detecting morphologic cartilage changes. Although several statistical methods over conventional radiography have been developed to perform quantitative cartilage analyses, little work has been done capturing the development and progression of cartilage lesions (or abnormal regions) and how they naturally progress. There are two major challenges, including (i) the lack of building spatial-temporal correspondences and correlations in cartilage thickness and (ii) the spatio-temporal heterogeneity in abnormal regions. The goal of this work is to propose a dynamic abnormality detection and progression (DADP) framework for quantitative cartilage analysis, while addressing the two challenges. First, spatial correspondences are established on flattened 2D cartilage thickness maps extracted from 3D knee MR images both across time within each subject and across all subjects. Second, a dynamic functional mixed effects model (DFMEM) is proposed to quantify abnormality progression across time points and subjects, while accounting for the spatio-temporal heterogeneity. We systematically evaluate our DADP using simulations and real data from the Osteoarthritis Initiative (OAI). Our results show that DADP not only effectively detects subject-specific dynamic abnormal regions, but also provides population-level statistical disease mapping and subgroup analysis.

Three-Dimensional Modeling and Simulation of Normal Knee Joint Motion

Abstract Three-dimensional (3D) modeling and simulation have important applications in orthopedic surgery. This technology can be used to create a 3D model of anatomic structures that can be used as a template to perform surgical simulations preoperatively. This study was aimed to apply 3D modeling to simulate flexion of a human knee based on the normal knee joint motion that includes tibial internal rotation and femoral rollback under high degree knee flexion. A 3D knee joint model was constructed from a two-dimensional computed tomography image of a normal knee joint. Based on the reconstructed 3D model of the knee joint, simulation of knee joint flexion at 90,100,110,120,130, and 140 degrees was conducted while replicating the normal knee joint motion. The rotation, translation, and tilt wizards allow the simulation of a complex motion that happens at the human knee joint. Simulation of knee joint motion allows replication of the complex motion during a gait cycle and can offer an alternative option to in vivo experiments to simulate surgical techniques, analyse knee implants and conduct biomechanical studies. This study has demonstrated that 3D modeling and simulation technology is an important tool to study the relationship between the degree of knee flexion and femoral tunnel orientation in anterior cruciate ligament (ACL) reconstruction. Accordingly, the joint flexion angle, and the sagittal and axial drill angles should be taken into consideration in ACL reconstruction.

3D Knee Kinematic Parameters Effectively Diagnose Knee Osteoarthritis and Assess Its Therapeutic Strategy

Knee osteoarthritis (KOA) is a worldwide disease leading to knee function loss and disorders. However, traditional assessment by X‐ray cannot assess patients’ knee functions and disorders dynamically, making it impossible to achieve a direct functional assessment of KOA. To solve this problem, here it is shown that 3D knee gait parameters could be used to diagnose KOA and guide its therapeutic strategy through direct functional assessment. We employ a total of 1201 participants, and successfully build and validate diagnostic and predictive models for KOA diagnosis and therapeutic strategy using an artificial intelligence (AI)‐based method, logistic regression, a kind of interpretable machine learning. Four diagnostic models are successfully established including angular (AM), translational (TM), composite (CM), and ATCM (a parallel conjoint model of AM, TM, and CM) model with a Youden index of 0.7312, 0.6689, 0.8214, and 0.7492, respectively. The same AI‐based method is also used to develop medical decision classification (MDC) for predicting whether a KOA patient needs operative intervention or not. MDC has a Youden index, sensitivity, and specificity of 0.8886, 92.11%, and 96.75%, respectively. These findings contribute to new knowledge of knee kinematics and KOA diagnosis and represent a new approach to accurate KOA diagnosis and assessment.

EFFECTS OF A PROPHYLACTIC KNEE SLEEVE ON THE ANTERIOR CRUCIATE LIGAMENT AND LOWER EXTREMITY BIOMECHANICS: AN EXAMINATION USING MUSCULOSKELETAL SIMULATION

The current study aimed using a two-experiment musculoskeletal simulation-based approach, measuring anterior cruciate ligament (ACL) biomechanics, knee joint kinematics, and lower extremity joint loading to examine the effects of both a prophylactic knee sleeve on (1) a sport-specific change of direction movement in female footballers and (2) a single leg landing in male footballers. Experiment 1 examined 12 female university first team level footballers (age [Formula: see text] years, height [Formula: see text][Formula: see text]m, body mass [Formula: see text][Formula: see text]kg) undertaking a [Formula: see text] cutting movement in sleeve and no-sleeve conditions. Experiment 2 examined 10 male university first team level footballers (age [Formula: see text] years, height [Formula: see text][Formula: see text]m, body mass [Formula: see text][Formula: see text]kg) undertaking a single leg drop jump landing in sleeve and no-sleeve conditions. In each experiment, data was collected in a biomechanics laboratory and three-dimensional (3D) motion capture and ground reaction force (GRF) information was collected. 3D kinematics, 3D knee kinetics, and ACL ligament forces/strains were measured using musculoskeletal simulation, and participants were also asked to subjectively rate the knee sleeve in terms of both comfort and stability. Experiment 1 showed that the sleeve condition was associated with greater ACL strain ([Formula: see text]% and no-[Formula: see text]%) and forces ([Formula: see text] BW and no-[Formula: see text] BW). In addition, the brace condition also enhanced lateral compressive tibiofemoral ([Formula: see text] BW and no-[Formula: see text] BW) and total compressive tibiofemoral force ([Formula: see text] BW and no-[Formula: see text] BW). Finally, for the subjective ratings, participants indicated that the knee sleeve significantly improved perceived comfort and stability. Experiment 2 did not reveal any statistical differences between knee sleeve and no-sleeve conditions nor any effects of the knee sleeve on subjective ratings of comfort or stability. Therefore, the findings from the current investigation suggest that the prophylactic knee sleeve examined in the current investigation does not appear to reduce the biomechanical parameters linked to the aetiology of knee pathologies in male/female footballers.

Measuring Knee Joint Laxity in Three Degrees-of-Freedom In Vivo Using a Robotics- and Image-Based Technology.

Accurate and reliable information about three-dimensional (3D) knee joint laxity can prevent misdiagnosis and avoid incorrect treatments. Nevertheless, knee laxity assessments presented in the literature suffer from significant drawbacks such as soft tissue artifacts, restricting the knee within the measurement, and the absence of quantitative knee ligament property information. In this study, we demonstrated the applicability of a novel methodology for measuring 3D knee laxity, combining robotics- and image-based technology. As such technology has never been applied to healthy living subjects, the aims of this study were to develop novel technology to measure 3D knee laxity in vivo and to provide proof-of-concept 3D knee laxity measurements. To measure tibiofemoral movements, four healthy subjects were placed on a custom-built arthrometer located inside a low dose biplanar X-ray system with an approximately 60 deg knee flexion angle. Anteroposterior and mediolateral translation as well as internal and external rotation loads were subsequently applied to the unconstrained leg, which was placed inside a pneumatic cast boot. Bone contours were segmented in the obtained X-rays, to which subject-specific bone geometries from magnetic resonance imaging (MRI) scans were registered. Afterward, tibiofemoral poses were computed. Measurements of primary and secondary laxity revealed considerable interpersonal differences. The method differs from those available by the ability to accurately track secondary laxity of the unrestricted knee and to apply coupled forces in multiple planes. Our methodology can provide reliable information for academic knee ligament research as well as for clinical diagnostics in the future.

CORR Insights®: What Is the Maximum Tibial Tunnel Angle for Transtibial PCL Reconstruction? A Comparison Based on Virtual Radiographs, CT Images, and 3D Knee Models.

1Orthopaedic Surgeon, Mammoth Orthopedic Institute, Mammoth Lakes, CA, USA B. B. Gilmer, Mammoth Orthopedic Institute, 85 Sierra Park Road, Mammoth Lakes, CA 93546, USA, Email: [email protected] This CORR Insights® is a commentary on the article “What Is the Maximum Tibial Tunnel Angle for Transtibial PCL Reconstruction? A Comparison Based on Virtual Radiographs, CT Images, and 3D Knee Models” by Teng and colleagues available at: DOI: 10.1097/CORR.0000000000002111. The author certifies that there are no funding or commercial associations (consultancies, stock ownership, equity interest, patent/licensing arrangements, etc.) that might pose a conflict of interest in connection with the submitted article related to the author or any immediate family members. All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research® editors and board members are on file with the publication and can be viewed on request. The opinions expressed are those of the writer, and do not reflect the opinion or policy of CORR® or The Association of Bone and Joint Surgeons®.

Concurrent Validity and Reliability of Two-dimensional Frontal Plane Knee Measurements during Multi-directional Cutting Maneuvers.

BackgroundExcessive knee valgus has been strongly suggested as a contributing key factor for anterior cruciate ligament (ACL) injuries. Three-dimensional (3D) motion analysis is considered the “gold standard” to assess joint kinematics, however, this is difficult for on-field assessments and for clinical setting.PurposeTo investigate the concurrent validity of 2D measurements of knee valgus angle during cutting in different directions and to explore intra-rater and inter-rater reliability of the 2D measurements.Study DesignDescriptive laboratory studyMethodSeven recreational soccer players participated in this study. Participants performed three trials of cutting maneuvers in three different directions (30º, 60º, and 90º) with the dominant leg. Cutting maneuvers were recorded simultaneously with a video camera and a ViconTM motion capture system. Knee valgus angle from 2D and 3D measurements at initial contact and at peak vertical ground reaction force (vGRF) were extracted. The Pearson’s correlation was used to explore the relationship between the 2D and 3D measurements, and reliability of the 2D measurements were performed using intraclass correlation coefficients (ICC).ResultSignificant correlations between 2D and 3D knee valgus measurements were noted for 60º (r = 0.45) and 90º (r = 0.77) cutting maneuvers at initial contact. At peak vGRF, significant correlations between 2D and 3D knee valgus measurements were noted for 30º, 60º, and 90º cutting maneuvers (r=0.45, r=0.74, r=0.78), respectively. Good-to-excellent intra-rater and inter-rater reliability of the 2D knee valgus measurements was observed during cutting in all directions (ICCs: 0.821-0.997).ConclusionModerate-to-strong correlation between 2D and 3D knee valgus measurements during 60°-90° cutting maneuvers, and good-to-excellent intra-rater and excellent inter-rater reliability for the 2D measurements in the present study supports the use of 2D knee valgus measurements in the evaluation of targeted interventions, although the limitations of examining cutting maneuvers using 2D measurement in complex movement still need to be considered.Level of Evidence3

Do Individuals with History of Patellofemoral Pain Walk and Squat Similarly to Healthy Controls? A 3D Kinematic Analysis During Pain Remission Phase.

BackgroundPatellofemoral pain (PFP) is typically accompanied by changes in movement pattern. However, it is unclear if these changes persist in the remission phase of symptoms. Investigating movement patterns in individuals in remission phase of PFP may help to further guide the rehabilitation process and to understand whether changes are due to high levels of pain or related to other factors.PurposeTo compare 3D kinematics during walking and the single leg squat (SLS) between individuals with history of PFP in remission phase and a control group without history of lower limb injuries and PFP.Study DesignCross-sectional case-control study.MethodsIndividuals with onset of PFP for at least one year and in phase of remission of symptoms (experimental group [EG]; n=13, 30±8 years) were compared to a control group (CG, n=13, 28±7 years). A 10-camera motion analysis system (Vicon-Nexus®) was used to record 3D ankle, knee, hip and trunk angles during walking and SLS.ResultsThe EG presented less ankle dorsiflexion, knee and hip flexion during the stance phase of walking compared to the CG (p=0.005, large effect size ηp2 = 0.141). During the SLS, no between-group differences were observed for the ankle, knee and hip angles at the peak of knee flexion (p>0.05). A trend for increased trunk range of movement in the EG compared to the CG was observed (p=0.075, medium effect size ηp2 = 0.127).ConclusionThe results of this study indicate less movement in the sagittal plane during walking, and a trend towards more movement of the trunk during SLS in the EG compared to the CG. The participants of the EG had minimal symptoms, to the point of not classifying them as pathological. However, the between-group differences suggest that even in the remission phase, kinematic differences persist for some reason and may contribute to the recurring pain in PFP individuals.Level of EvidenceLevel 3.

What Is the Maximum Tibial Tunnel Angle for Transtibial PCL Reconstruction? A Comparison Based on Virtual Radiographs, CT Images, and 3D Knee Models.

To minimize the killer turn caused by the sharp margin of the tibial tunnel exit in transtibial PCL reconstruction, surgeons tend to maximize the angle of the tibial tunnel in relation to the tibial plateau. However, to date, no consensus has been reached regarding the maximum angle for the PCL tibial tunnel. In this study we sought (1) to determine the maximum tibial tunnel angle for the anteromedial and anterolateral approaches in transtibial PCL reconstruction; (2) to compare the differences in the maximum angle based on three measurement methods: virtual radiographs, CT images, and three-dimensional (3D) knee models; and (3) to conduct a correlation analysis to determine whether patient anthropomorphic factors (age, sex, height, and BMI) are associated with the maximum tibial tunnel angle. Between January 2018 and December 2020, 625 patients who underwent CT scanning for knee injuries were retrospectively reviewed in our institution. Inclusion criteria were patients 18 to 60 years of age with a Kellgren-Lawrence grade of knee osteoarthritis less than 1 and CT images that clearly showed the PCL tibial attachment. Exclusion criteria were patients with a history of tibial plateau fracture, PCL injuries, tumor, and deformity around the knee. Finally, 104 patients (43 males and 61 females, median age: 38 [range 24 to 56] years, height: 165 ± 9 cm, median BMI: 23 kg/cm2 [range 17 to 31]) were included for analysis. CT data were used to create virtual 3D knee models, and virtual true lateral knee radiographs were obtained by rotating the 3D knee models. Virtual 3D knee models were used as an in vitro standard method to assess the true maximum tibial tunnel angle of anteromedial and anterolateral approaches in transtibial PCL reconstruction. The tibial tunnel's entry was placed 1.5 cm anteromedial and anterolateral to the tibial tubercle for the two approaches. To obtain the maximum angle, a 10-mm- diameter tibial tunnel was simulated by making the tibial tunnel near the posterior tibial cortex. The maximum tibial tunnel angle, tibial tunnel lengths, and perpendicular distances of the tunnel's entry point to the tibial plateau were measured on virtual radiographs, CT images, and virtual 3D knee models. One-way ANOVA was used to compare the differences in the maximum angle among groups, and correlation analysis was performed to identify the relationship of the maximum angle and anthropomorphic factors (age, sex, height, and BMI). The maximum angle of the PCL tibial tunnel relative to the tibial plateau was greater in the anteromedial group than the anterolateral group (58° ± 8° versus 50° ± 8°, mean difference 8° [95% CI 6° to 10°]; p < 0.001). The maximum angle of the PCL tibial tunnel was greater in the virtual radiograph group than the CT image (68° ± 6° versus 49° ± 5°, mean difference 19° [95% CI 17° to 21°]; p < 0.001), the anteromedial approach (68° ± 6° versus 58° ± 8°, mean difference 10° [95% CI 8° to 12°]; p < 0.001), and the anterolateral approach (68° ± 6° versus 50° ± 8°, mean difference 18° [95% CI 16° to 20°]; p < 0.001), but no difference was found between the CT image and the anterolateral groups (49° ± 5° versus 50° ± 8°, mean difference -1° [95% CI -4° to 1°]; p = 0.79). We found no patient anthropomorphic characteristics (age, sex, height, and BMI) that were associated with the maximum angle. Surgeons should note that the mean maximum angle of the tibial tunnel relative to the tibial plateau was greater in the anteromedial than anterolateral approach in PCL reconstruction, and the maximum angle might be overestimated on virtual radiographs and underestimated on CT images. To perform PCL reconstruction more safely, the findings of this study suggest that the PCL drill system should be set differently for the anteromedial and anterolateral approaches, and the maximum angle measured by intraoperative fluoroscopy should be reduced 10° for the anteromedial approach and 18° for the anterolateral approach. Future clinical or cadaveric studies are needed to validate our findings.

Detecting Small Anatomical Structures in 3D Knee MRI Segmentation by Fully Convolutional Networks

Accurately identifying the pixels of small organs or lesions from magnetic resonance imaging (MRI) has a critical impact on clinical diagnosis. U-net is the most well-known and commonly used neural network for image segmentation. However, the small anatomical structures in medical images cannot be well recognised by U-net. This paper explores the performance of the U-net architectures in knee MRI segmentation to find a relative structure that can obtain high accuracies for both small and large anatomical structures. To maximise the utilities of U-net architecture, we apply three types of components, residual blocks, squeeze-and-excitation (SE) blocks, and dense blocks, to construct four variants of U-net, namely U-net variants. Among these variants, our experiments show that SE blocks can improve the segmentation accuracies of small labels. We adopt DeepLabv3plus architecture for 3D medical image segmentation by equipping SE blocks based on this discovery. The experimental results show that U-net with SE block achieves higher accuracy in parts of small anatomical structures. In contrast, DeepLabv3plus with SE block performs better on the average dice coefficient of small and large labels.

Healthy Knee Kinematic Phenotypes Identification Based on a Clustering Data Analysis

The purpose of this study is to identify healthy phenotypes in knee kinematics based on clustering data analysis. Our analysis uses the 3D knee kinematics curves, namely, flexion/extension, abduction/adduction, and tibial internal/external rotation, measured via a KneeKG™ system during a gait task. We investigated two data representation approaches that are based on the joint analysis of the three dimensions. The first is a global approach that is considered a concatenation of the kinematic data without any dimensionality reduction. The second is a local approach that is considered a set of 69 biomechanical parameters of interest extracted from the 3D kinematic curves. The data representations are followed by a clustering process, based on the BIRCH (balanced iterative reducing and clustering using hierarchies) discriminant model, to separate 3D knee kinematics into homogeneous groups or clusters. Phenotypes were obtained by averaging those groups. We validated the clusters using inter-cluster correlation and statistical hypothesis tests. The simulation results showed that the global approach is more efficient, and it allows the identification of three descriptive 3D kinematic phenotypes within a healthy knee population.

A new 3D software for analysis and planning of lower limb and patellofemoral alignment: Reliability and accuracy

BackgroundThe new software, mediCAD® 3D Knee Sport (mediCAD Hectec GmbH, Altdorf/Landshut, Germany), promises to combine automated digital 3D bone model generation, 3D analysis of lower limb geometry including analysis of the patellofemoral joint, and osteotomy planning. The aim of this study was to evaluate its reliability and accuracy. MethodsIn this retrospective multi-observer study, three post-mortem CTs were analysed by three observers at three points in time. Reliability was evaluated by calculating the intraclass correlation coefficient (ICC) of interobserver agreement. Accuracy was evaluated using the mean deviation D from the mean and the standard deviation SD from D. ResultsTen of 18 alignment parameters showed excellent, two good and three moderate interobserver agreement. Poor agreement was found for the mechanical medial proximal tibial angle, the trochlear sulcus angle and trochlea depth. Mean interobserver ICC of all parameters ranged from 0.32 to 0.99. Fifteen of 18 parameters showed a low mean deviation D from the mean of < 2 mm / 2°. Three parameters related to the patellofemoral joint showed medium or high D (patella tilt, trochlear sulcus angle, patellar ridge angle). These parameters also showed the highest values for the SD of D. The trochlear sulcus angle was found to be the only parameter with high mean deviation (D ≥ 5 mm/5°) with D being 5.67 ± 3.23°. ConclusionsThe current version of the software achieves good interobserver reliability and accuracy with the exception of a few measurement parameters.