Intraobserver Reliability Research Articles

Assessing bone density on MRI: comparison between routine MRI sequences and DEXA scans.

Magnetic Resonance Imaging (MRI) is frequently utilised to aid in the comprehensive assessment of back pain, while dual-energy x-ray absorptiometry (DEXA) is the gold standard test for the assessment of bone density. Assessing bone density on MRI could reduce costs and avoid exposing patients to ionising radiation. The aim of this paper is to investigate whether the relative signal intensity of vertebral bodies compared to other structures can detect osteoporosis on MRI. 100 patients that had undergone both a lumbar spine MRI and a DEXA scan were identified. The T1 and T2 signal intensity of L1-L4 vertebral bodies (VB), cerebro-spinal fluid (CSF), and psoas muscle were measured within a 1-cm2 region of interest (ROI), and the signal intensity ratios were calculated. The ratios were stratified as normal, osteopenic, or osteoporotic based on DEXA T-scores. The T1 VB /T1 CSF ratio was significantly higher in the osteoporotic group than the normal and osteopenic groups (p < 0.001). The T1 VB /T1 CSF ratio had excellent discrimination (AUC = 0.841) for the presence of osteoporosis. The Pearson correlation coefficient between the DEXA T-score and the T1VB/T1CSF ratio was -0.474 (p < 0.001). The intra-observer (ICC = 0.910, 95% CI = 0.757-0.966) and inter-observer reliability (ICC = 0.927, 95% CI = 0.824-0.970) were excellent. In our cohort, a T1 VB / T1 CSF ratio of greater than 4 is 66.7% sensitive but 90.0% specific for the presence of osteoporosis. A high T1VB/T1CSF ratio suggests osteoporosis on MRI. Prospective validation is needed to confirm these findings.

The accuracy of multiple regression models for predicting the individual risk of recurrent lateral patellar dislocation

BackgroundRecurrent lateral patellar dislocation (RLPD) poses a significant threat to patients’ quality of life due to knee pain, patellofemoral cartilage damage, and potential traumatic arthritis. Predictive scoring systems have been developed to assess the risk of RLPD; however, their relative accuracy remains uncertain.PurposeTo investigate the accuracy of the multiple regression models to predict the individual risk of recurrent LPD.MethodsThe Patellar Instability probability calculator (PIP), Recurrent Instability of the Patella Score (RIP), and Patellar Instability Severity Score (PIS) scoring rules were measured in 171 patients with a history of patellar dislocation and 171 healthy individuals. Three prediction models were calculated based on the data to predict the risk of recurrent lateral patellar dislocation. The inter-observer and intra-observer reliability of each measurement parameter was evaluated. The predictive capacity of the three-prediction model was investigated using the receiver operating characteristic curve.ResultsIn the case group of 171 patients, PIS accurately predicted recurrent lateral Patella dislocation in 143 patients. RIP was 96, and PIP was 83. The positive predictive values were 92.9%, 64%, and 68% respectively. In the control group of 171 patients, the PIS was validated in 160 patients who would not experience dislocations. RIP was 117, and PIP was 50. The negative predictive values were 85.1%, 60.9%, and 36.2%, respectively. The area under the curve score for the PIS was 0.866, and the RIP was 0.673. the PIP was 0.678.ConclusionRIP and PIP did not work to predict LPD. PIS can accurately predict recurrent lateral patellar dislocation. It can aid doctors in making treatment decisions.Level of evidenceLevel III, retrospective comparative study.

A coronoid-centric classification system of proximal trans-ulnar fracture-dislocations has almost perfect intraobserver and interobserver agreement.

Fracture-dislocations of the elbow, particularly those that involve a fracture through the proximal ulna, are complex and can be difficult to manage. Moreover, current classification systems often cannot discriminate between Monteggia-variant injury patterns and trans-olecranon fracture-dislocations, particularly when the fracture involves the coronoid. The Mayo classification of proximal trans-ulnar fracture-dislocations categorizes these fractures into 3 types according to what the coronoid is still attached to: trans-olecranon fracture-dislocations (the coronoid is still attached to the ulnar metaphysis); Monteggia-variant fracture-dislocations (the coronoid is still attached to the olecranon); and ulnar basal coronoid fracture-dislocations (the coronoid is not attached to either the olecranon or the ulnar metaphysis). The purpose of this study was to evaluate the intraobserver and interobserver agreement of the Mayo classification system when assessing elbow fracture-dislocations involving the proximal ulna based on radiographs and computed tomography scans. Three fellowship-trained shoulder and elbow surgeons and 2 fellowship-trained orthopedic trauma surgeons blindly and independently evaluated the radiographs and computed tomography scans of 90 consecutive proximal trans-ulnar fracture-dislocations treated at a level I trauma center. The inclusion criteria included subluxation or dislocation of the elbow and/or radioulnar joint with a complete fracture through the proximal ulna. Each surgeon classified all fractures according to the Mayo classification, which is based on what the coronoid remains attached to (ulnar metaphysis, olecranon, or neither). Intraobserver reliability was determined by scrambling the order of the fractures and having each observer classify all the fractures again after a washout period ≥ 6 weeks. Interobserver reliability was obtained to assess the overall agreement between observers. κ Values were calculated for both intraobserver reliability and interobserver reliability. The average intraobserver agreement was 0.87 (almost perfect agreement; range, 0.76-0.91). Interobserver agreement was 0.80 (substantial agreement; range, 0.70-0.90) for the first reading session and 0.89 (almost perfect agreement; range, 0.85-0.93) for the second reading session. The overall average interobserver agreement was 0.85 (almost perfect agreement; range, 0.79-0.91). Classifying proximal trans-ulnar fracture-dislocations based on what the coronoid remains attached to (olecranon, ulnar metaphysis, or neither) was associated with almost perfect intraobserver and interobserver agreement, regardless of trauma vs. shoulder and elbow fellowship training. Further research is needed to determine whether the use of this classification system leads to the application of principles specific to the management of these injuries and translates into better outcomes.

A Modified Approach to Measuring Femoro-Epiphyseal Acetabular Roof (FEAR) Index Has Better Intraobserver and Interobserver Reliability Compared With the Original FEAR Index

PurposeTo propose a modified approach to measuring femoro-epiphyseal acetabular roof (FEAR) index while still abiding by its definition and biomechanical basis, and to compare the intra- and inter-observer reliabilities of the original and the modified FEAR index. To propose a classification for medial sourcil edges. MethodsWe retrospectively reviewed a consecutive series of patients treated with periacetabular osteotomy and/or hip arthroscopy at a single institute. Patients with unilateral or bilateral symptomatic borderline hip(s) were included. Hips with remarkable osteoarthritis, deformities, history of previous surgery, or without symptoms were excluded. A modified FEAR index was defined using a best-fit circle to determine the sourcil line and two ancillary lines connecting femoral head and sourcil edges to determine epiphyseal line. Lateral center-edge angle, Sharp’s angle, Tonnis angle on all hips, as well as FEAR index with original and modified approaches were measured. Intra- and inter-observer reliability were calculated as intraclass correlation coefficients (ICC) for FEAR index with both approaches and other alignments. A classification was proposed to categorize medial sourcil edges. ICC for the two approaches across different sourcil groups were also calculated. ResultsAfter reviewing 411 patients, 49 were finally included. Thirty-two patients (40 hips) were identified as having borderline dysplasia defined by an LCEA of 18 to 25 degrees. Intra-observer ICC for the modified method were good to excellent for borderline hips; poor to excellent for DDH; moderate to excellent for normal hips. As for inter-observer reliability, modified approach outperformed original approach with moderate to good inter-observer reliability (DDH group, ICC=0.650; borderline dysplasia group, ICC=0.813; normal hip group, ICC=0.709). The medial sourcil edge were classified to 3 groups upon its morphology. Type II (39.0%) and III (43.9%) sourcils were the dominant patterns. The sourcil classification had substantial intra-observer agreement (observer 4, kappa=0.68; observer 1, kappa=0.799) and moderate inter-observer agreement (kappa=0.465). Modified approach to FEAR index possessed greater inter-observer reliability in all medial sourcil edge patterns. ConclusionThe modified FEAR index has better intra- and inter-observer reliability compared with the original approach in all hip groups and sourcil groups. Type II and III sourcils accounts for the majority to which the modified approach is better. Level of EvidenceLevel II, Development of diagnostic criteria (consecutive patients with consistently applied reference standard and blinding)

Standardized protocol for quantification of nerve bundle density as a biomarker for endometriosis

IntroductionWe propose a standardized protocol for measurement of nerve bundle density in endometriosis as a potential biomarker, including in deep endometriosis (DE), ovarian endometriomas (OMA) and superficial peritoneal endometriosis (SUP).MethodsThis was a prospective cohort of surgically excised endometriosis samples from Dec 1st 2013 and Dec 31st 2017 at a tertiary referral center for endometriosis in Vancouver, BC, Canada. Surgical data were available from linked patient registry. Protein gene product 9.5 (PGP9.5) was used to identify nerve bundles on immunohistochemistry. PGP9.5 nerve bundles were counted visually. To calculate nerve bundle density, PGP9.5 nerve bundle count was divided by the tissue surface area (total on the slide). All samples were assessed using NHS Elements software for semi-automated measurement of the tissue surface area. For a subset of samples, high power fields (HPFs) were also counted as manual measurement of the tissue surface area. Intraclass correlation was used to assess intra observer and inter observer reliability. Generalized linear mixed model (GLMM) with random intercepts only was conducted to assess differences in PGP9.5 nerve bundle density by endometriosis type (DE, OMA, SUP).ResultsIn total, 236 tissue samples out of 121 participants were available for analysis in the current study. Semi-automated surface area measurement could be performed in 94.5% of the samples and showed good correlation with manually counted HPFs (Spearman's rho = 0.781, p &amp;lt; 0.001). To assess intra observer reliability, 11 samples were assessed twice by the same observer; to assess inter observer reliability, 11 random samples were blindly assessed by two observers. Intra observer reliability and inter observer reliability for nerve bundle density were excellent: 0.979 and 0.985, respectively. PGP9.5 nerve bundle density varied among samples and no nerve bundles could be found in 24.6% of the samples. GLMM showed a significant difference in PGP9.5 nerve bundle density between the different endometriosis types (X2 = 87.6, P &amp;lt; 0.001 after adjusting for hormonal therapy, with higher density in DE and SUP in comparison to OMA).ConclusionA standardized protocol is presented to measure PGP9.5 nerve bundle density in endometriosis, which may serve as a biomarker reflecting local neurogenesis in the endometriosis microenvironment.

Reliability and reproducibility of CBCT assessment of mandibular changes before and after treatment for Class III growing patients - an easy and quick way for evaluation.

The objective of this study was to evaluate intraobserver reliability and inter-observer reproducibility of a 3-dimensional (3D) assessment method for mandibular changes of growing patients after orthodontic treatment for Class III malocclusion.Methods Cone-beam computed tomography (CBCT) scans were performed before and after orthodontic treatment for 27 patients. During the scan, the patient was positioned such that his/her mandibular plane was parallel to floor. Three observers independently worked on the DICOM data, reconstructed the pre- and post-treatment 3D models in software, selected the stable anatomical structures (basal bone area from the lingual surface of the symphysis to the distal aspect of the first molars) to guide the automated superimposition process. Then, each observer registered 14 anatomical landmarks on the virtual models, for three times after suitable interval, to generate 3 sets of coordinates; the mean was taken as the coordinates for that particular landmark. The intraobserver reliability and inter-observer reproducibility of the method were analyzed.Results The ICCs was > 0.90 for 25 (92.6%) of the intraobserver assessments. The precision of the measurement method was < 0.3mm in 24 (88.9%) cases. The interobserver reproducibility errors were < 0.3mm in 21 of the 27 cases.Conclusions The intraobserver reliability and inter-observer reproducibility of 3D assessment of mandibular changes using the virtual models were excellent.

A Clinical Classification of Cervical Ossification of the Posterior Longitudinal Ligament to Guide Surgical Strategy.

A clinical classification of cervical ossification of the posterior longitudinal ligament (COPLL) was developed based on imaging findings. This study aimed to establish a clinical classification for COPLL and provide corresponding surgery strategies for each subtype. A practical and reliable classification is needed to guide the treatment of COPLL. This study retrospectively reviewed plain radiographs, computed tomography scans, and magnetic resonance images of patients diagnosed with COPLL between 2018 and 2022 at Shanghai Changzheng Hospital. The types of COPLL were classified according to the location, morphology, and canal-occupying ratio (OR) of the ossification mass. Interobserver and intraobserver reliability were evaluated using Cohen's kappa. A total of 1000 cases were included, which were classified into five types: focal type (F type), short-sequential type (S type), long-sequential type (L type), high type (H type), and mixed type (M type). In addition, each type could be classified into subtype 1 or subtype 2 according to the canal-OR. Then each type could be further classified into other subtypes according to location and morphology. The interobserver reliabilities in the first and second rounds were 0.853 and 0.887, respectively. The intraobserver reliability was 0.888. The authors classified COPLL into a system comprised of five types and several subtypes according to canal-OR, location, and morphology. Surgical strategies for each subtype are also suggested. This provides a theoretical guide for the description and surgical management of COPLL.

Intra-Observer and Inter-Observer Reliability of Ankle Circumference Measurement in Patients with Diabetic Foot: A Prospective Observational Study.

Inflammation, being a typical response to vascular tissue alterations, induces variations in tissue oxygen diffusion pressure. Diabetic microangiopathy, an inflammatory process, is characterized by an increase in vascular flow at rest, reduced venous and arteriolar responses, and increased capillary permeability, resulting in oedema development, decreased transcutaneous oxygen pressure, and increased transcutaneous carbon dioxide pressure. This phenomenon potentially hampers ulcer healing. Although the figure-of-eight method has proven to be a reliable, valid, quick, and efficient test for assessing foot and ankle measurements in patients with oedema and compromised skin integrity, it has not been studied in patients with diabetic foot. The aim of this study was to determine and compare the intra- and inter-observer variabilities of the figure-of-eight method in patients with diabetic foot. A prospective observational and cross-sectional study was undertaken, involving sixty-one subjects from a specialized Diabetic Foot Unit. Three investigators with varying levels of experience independently measured the subjects to assess both intra-observer and inter-observer variability. The evaluation was conducted using the Intraclass Correlation Coefficient (ICC). In the statistical analysis, an ICC of 0.93, adjusted using a 95% confidence interval (CI), was obtained for inter-observer reliability ICC, indicating excellent reliability among observers. Furthermore, an ICC of 0.98 with a 95% CI was obtained for the intra-observer reliability analysis, indicating excellent reliability. The results support using this test during the clinical management of oedema in patients with diabetic foot. The absence of an objective, fast, and readily available diagnostic method for oedema in diabetic foot patients in clinical practice might pose a limitation. Subsequent research should tackle this issue and explore the correlation between ankle perimeter measurements and other clinical outcomes in diabetic foot patients, including wound healing and quality of life.

Automated decision support for Hallux Valgus treatment options using anteroposterior foot radiographs

BACKGROUND Assessment of the potential utility of deep learning with subsequent image analysis to automate the measurement of hallux valgus and intermetatarsal angles from radiographs to serve as a preoperative aid in establishing hallux valgus severity for clinical decision-making. AIM To investigate the accuracy of automated measurements of angles of hallux valgus from radiographs for further integration with the preoperative planning process. METHODS The data comprises 265 consecutive digital anteroposterior weightbearing foot radiographs. 181 radiographs were utilized for training (161) and validating (20) a U-Net neural network to achieve a mean Sørensen–Dice index &gt; 97% on bone segmentation. 84 test radiographs were used for manual (computer assisted) and automated measurements of hallux valgus severity determined by hallux valgus (HVA) and intermetatarsal angles (IMA). The reliability of manual and computer-based measurements was calculated using the interclass correlation coefficient (ICC) and standard error of measurement (SEM). Inter- and intraobserver reliability coefficients were also compared. An operative treatment recommendation was then applied to compare results between automated and manual angle measurements. RESULTS Very high reliability was achieved for HVA and IMA between the manual measurements of three independent clinicians. For HVA, the ICC between manual measurements was 0.96-0.99. For IMA, ICC was 0.78-0.95. Comparing manual against automated computer measurement, the reliability was high as well. For HVA, absolute agreement ICC and consistency ICC were 0.97, and SEM was 0.32. For IMA, absolute agreement ICC was 0.75, consistency ICC was 0.89, and SEM was 0.21. Additionally, a strong correlation (0.80) was observed between our approach and traditional clinical adjudication for preoperative planning of hallux valgus, according to an operative treatment algorithm proposed by EFORT. CONCLUSION The proposed automated, artificial intelligence assisted determination of hallux valgus angles based on deep learning holds great potential as an accurate and efficient tool, with comparable accuracy to manual measurements by expert clinicians. Our approach can be effectively implemented in clinical practice to determine the angles of hallux valgus from radiographs, classify the deformity severity, streamline preoperative decision-making prior to corrective surgery.

QUANTIFICATION OF ERROR IN THE MANUAL SEGMENTATION OF ANKLE CT SCANS

AbstractObjectivesThe fidelity of a 3D model created using image segmentation must be precisely quantified and evaluated for the model to be trusted for use in subsequent biomechanical studies such as finite element analysis. The bones within the ankle joint vary significantly in size and shape. The purpose of this study was to test the hypothesis that the accuracy and reliability of a segmented bone geometry is independent of the particular bone being measured.MethodsComputed tomography (CT) scan data (slice thickness 1 mm, pixel size 808±7 µm) from three anonymous patients was used for the development of the ankle geometries (consisting of the tibia, fibula, talus, calcaneus, and navicular bones) using Simpleware Scan IP software (Synopsys, Exeter, UK). Each CT scan was segmented 4 times by an inexperienced undergraduate, resulting in a total of 12 geometry assemblies. An experienced researcher segmented each scan once, and this was used as the ‘gold standard’ to quantify the accuracy. The solid bone geometries were imported into CAD software (Inventor 2023, Autodesk, CA, USA) for measurement of the surface area and volume of each bone, and the distances between bones (tibia to talus, talus to navicular, talus to calcaneus, and tibia to fibula) were carried out. The intra-class coefficient (ICC) was used to assess intra-observer reliability. Bland Altman plots were employed as a statistical measure for criteria validity (accuracy) [1].ResultsThe average ICC score was 0.93, which is regarded as a high reliability score for an inexperienced user. The talus to navicular and talus to tibia separations, which had the smallest distances, showed a slight decrease in reliability and this was observed for all separations shorter than 2 mm. According to the Bland-Altman plots, more than 95% of the data points were inside the borders of agreement, which is an excellent indication of accuracy. The bias percentage (average error percentage) varied between 1% and 4% and was constant across all parameters, with the proportion rising for short distance separations.ConclusionsThe current study demonstrates that an inexperienced undergraduate, with access to software manuals, can segment an ankle CT scan with excellent reliability. The present study also concluded that all five bones were segmented with high levels of accuracy, and this was not influenced by bone volume or type. The only factor found to influence the reliability was the magnitude of distance between bones, where if this was smaller than 2 mm it reduced the reliability, indicating the influence of CT scan resolution on the segmentation reliability.Declaration of Interest(b) declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported:I declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research project.

Effects of tactical dimension and situational variables in throw-ins on the offensive performance in football.

The aim of the study was to describe the usual practices in the execution of throw-ins by La Liga teams during the 2021-2022 season, identify tactical indicators related to the outcome of plays that start with a throw-in, calculate their predictive power, and finally analyse the influence of situational variables on the effectiveness of these plays. A total of 2,658 throw-ins, during 80 matches were analysed. Two UEFA PRO coaches designed an ad hoc observation instrument "Thrinfoot" and two observers coded the data after a training process. Inter and intra-observer reliability was calculated using Cohen´s Kappa coefficient, revealing almost perfect agreement. Multinomial logistic regression was applied to predict the throw-ins outcome (p<0.05). Results showed how fast throw-ins (OR = 0.7, p<0.05), without pressing (OR = 0.4, p<0.001), short and backwards (OR = 0.3, p<0.01) in the central zone (OR = 0.6, p<0.01) and made in the 16´-30´ (OR = 0.6, p<0.01), 61´-75´ (OR = 0.7, p< 0.05) periods, presented higher probabilities of continuing with possession. Match status losing>2 (OR = 4.1, p< 0.05) showed higher probabilities of success. On the other hand, throw-ins from the defensive zone presented higher probabilities of unsuccess (OR = 8.6, p<0.01) and losing possession (OR = 1.8, p<0.01). Finally, the bottom teams showed the highest probability of losing the ball. In conclusion, tactical indicators such as duration, press, distance, direction and zone were identified as key performance indicators and the situational variables team quality, match status and time influence the outcome of throw-ins. These findings provide valuable insights to coaches regarding the factors that influence the outcome of throw-ins. This allows them to design optimal strategies for both executing and defending these plays based on the game situation and their immediate aims.

Novel Modic grading scoring system and its clinical validation: a preliminary investigation.

To propose a novel Modic grading scoring system and explore the relationship between the Modic grading score and disc degeneration, disc herniation, disc height, and clinical symptom scores. In total, 194 patients were included in the study. The new Modic grading scoring system included four indicators: invaded vertebral height, invaded endplate length, endplate morphology, and grade of endplate defects. The severity of Modic changes was visually quantified by numerical scores, and the kappa value was used to verify the interobserver and intraobserver reliability. Spearman correlation analysis was used to explore the relationship between the Modic grading score and intervertebral disc degeneration, disc herniation, disc height, and clinical symptom scores. The interobserver and intraobserver reliability showed substantial to almost perfect agreement in the new Modic grading scoring system. The Modic grading score was positively correlated with intervertebral disc degeneration (r = 0.757, p < 0.001) and negatively correlated with the intervertebral disc height index (r = - 0.231, p < 0.001). There was no significant correlation between the Modic grading scoring system and disc herniation (r = 0.369, p = 0.249). Additionally, there was no significant correlation between the Modic grading score and the Japanese Orthopaedic Association score (r = - 0.349, p = 0.25), Oswestry Disability Index score (r = 0.246, p = 0.11), or visual analogue scale score (r = 0.315, p = 0.35). The new Modic grading scoring system had good interobserver and intraobserver reliability. The Modic grading score was positively correlated with intervertebral disc degeneration and negatively correlated with the intervertebral disc height.

MRI-based measurements of spondylolisthesis and kyphosis in degenerative cervical myelopathy

BackgroundTo provide normative data and to determine accuracy and reliability of preoperative measurements of spondylolisthesis and kyphosis on supine static magnetic resonance imaging (MRI) of patients with degenerative cervical myelopathy.MethodsT2-weighted midsagittal images of the cervical spine were in 100 cases reviewed twice by one junior observer, with an interval of 3 months, and once by a senior observer. The spondylolisthesis slip (SSlip, mm) and the modified K-line interval (mK-line INT, mm) were assessed for accuracy with the standard error of measurement (SEm) and the minimum detectable change (MDC). Intraobserver and interobserver reliability levels were determined using the intraclass correlation coefficient (ICC).ResultsThe SEm was 0.5 mm (95% CI 0.4–0.6) for spondylolisthesis and 0.6 mm (95% CI 0.5–0.7) for kyphosis. The MDC, i.e., the smallest difference between two examinations that can be detected with statistical certainty, was 1.5 mm (95% CI 1.2–1.8) for spondylolisthesis and 1.6 mm (95% CI 1.3–1.8) for kyphosis. The highest reliability levels were seen between the second observation of the junior examiner and the senior observer (ICC = 0.80 [95% CI 0.70–0.87] and ICC = 0.96 [95% CI 0.94–0.98] for SSlip and mK-line INT, respectively).ConclusionsThis study provides normative values of alignment measurements of spondylolisthesis and kyphosis in DCM patients. It further shows the importance of taking measurement errors into account when defining cut-off values for cervical deformity parameters and their potential clinical application in surgical decision-making.

Assessment of a novel deep learning-based software developed for automatic feature extraction and grading of radiographic knee osteoarthritis

BackgroundThe Kellgren-Lawrence (KL) grading system is the most widely used method to classify the severity of osteoarthritis (OA) of the knee. However, due to ambiguity of terminology, the KL system showed inferior inter- and intra-observer reliability. For a more reliable evaluation, we recently developed novel deep learning (DL) software known as MediAI-OA to extract each radiographic feature of knee OA and to grade OA severity based on the KL system.MethodsThis research used data from the Osteoarthritis Initiative for training and validation of MediAI-OA. 44,193 radiographs and 810 radiographs were set as the training data and used as validation data, respectively. This AI model was developed to automatically quantify the degree of joint space narrowing (JSN) of medial and lateral tibiofemoral joint, to automatically detect osteophytes in four regions (medial distal femur, lateral distal femur, medial proximal tibia and lateral proximal tibia) of the knee joint, to classify the KL grade, and present the results of these three OA features together. The model was tested by using 400 test datasets, and the results were compared to the ground truth. The accuracy of the JSN quantification and osteophyte detection was evaluated. The KL grade classification performance was evaluated by precision, recall, F1 score, accuracy, and Cohen's kappa coefficient. In addition, we defined KL grade 2 or higher as clinically significant OA, and accuracy of OA diagnosis were obtained.ResultsThe mean squared error of JSN rate quantification was 0.067 and average osteophyte detection accuracy of the MediAI-OA was 0.84. The accuracy of KL grading was 0.83, and the kappa coefficient between the AI model and ground truth was 0.768, which demonstrated substantial consistency. The OA diagnosis accuracy of this software was 0.92.ConclusionsThe novel DL software known as MediAI-OA demonstrated satisfactory performance comparable to that of experienced orthopedic surgeons and radiologists for analyzing features of knee OA, KL grading and OA diagnosis. Therefore, reliable KL grading can be performed and the burden of the radiologist can be reduced by using MediAI-OA.

A Novel Radiological Classification System of the Proximal Humerus.

Different anatomical variants have been reported for the proximal and distal femur. Given this context, the goal of our study was to answer the following question: Can we also identify different anatomical variants of the proximal humerus?Two hundred anteroposterior (AP) radiographs of the proximal humerus with an equal gender distribution and equal laterality per gender were reviewed. The metaphyseal diameter of the proximal humerus at the level of the anatomical neck (X) and the intramedullary diameter at 10cm from the apex of the greater tuberosity (Y) were measured. A new ratio was established, based on both measurements (Y/X). Radiographs showed different anatomical variants: type A: Y/X<0.3, type B: Y/X=0.3-0.4, type C: Y/X>0.4. Two observers reviewed the AP radiographs independently and blindly in 2 different sessions.Three different anatomical groups (A, B and C) were identified based on the 25th and 75th percentiles. A higher percentage of type C was observed among females and a higher percentage of type A among males. A high inter-observer reliability was noted, with a Cronbach's alpha of 0.97 (ICC 0.96-0.98). The intra-observer reliability for observer 1 had a Cronbach's alpha of 0.98.A novel radiological classification of the proximal humerus has been established based on 3 different anatomical types (A, B and C). Further studies are needed to establish whether the novel classification system can be used as an indicator for aseptic loosening of cemented or cementless total shoulder arthroplasty.

Contextualizing Evaluation of Performance in Volleyball: Introducing Contextual Individual Contribution Coefficients to Assess Technical Actions.

In this study, we proposed novel metrics for evaluating volleyball technical performance in relation to the action context. To assess each player's relative participation, we also introduced relative contextual coefficients. We analyzed 20 games played by the world's top eight teams during the 2019 FIVB Women's Club World Championship, using Data Volley software and Python programming. We evaluated inter- and intra-observer reliability and used binomial logistic regression models to estimate each variable's probability of having contributed to the team's set win. We calculated estimated confidence intervals, standard errors, and Wald values; and we employed Akaike's and Bayesian criteria to evaluate the model's goodness of fit. We identified optimal cut-off points using receiver operating characteristic curves, and we found that the proposed contextual evaluation coefficients prevented overestimation of a player's technical performance in uneven situations. We addressed the issue in which the winning team may be the one that scored the fewest points, and we were able to predict a team's victory with confidence. The proposed coefficients made multiple technical and contextual aspects of the game easily accessible and comprehensible, offering significant beneficial implications for coaches and players.

Intraobserver and Interobserver Reliability of the Snyder and Expanded SLAP Classification System: A Video Study

Background: Superior labral anterior and posterior (SLAP) tears are a common finding in overhead athletes. The original classification system produced by Snyder in 1990 contained 4 types of SLAP tears and was later expanded to 10 types. The classification has been challenging because of inconsistencies between surgeons making diagnoses and treatments based on the diagnosis. Furthermore, patient factors—such as age and sports played—affect the treatment algorithms, even across similarly classified SLAP tears. Purpose: To (1) assess the interobserver and intraobserver reliability of the Snyder and expanded SLAP (ESLAP) classification systems and (2) determine the consistency of treatment for a given SLAP tear depending on different clinical scenarios. Study Design: Cohort study (diagnosis); Level of evidence, 3. Methods: A total of 20 arthroscopic surgical videos and magnetic resonance imaging scans of patients with SLAP tears were sent to 20 orthopaedic sports medicine surgeons at various stages of training. Surgeons were asked to identify the type of SLAP tear using the Snyder and ESLAP classifications. Surgeons were then asked to determine the treatment for a SLAP tear using 4 clinical scenarios: (1) in the throwing arm of an 18-year-old pitcher; (2) in the dominant arm of an 18-year-old overhead athlete; (3) a 35-year-old overhead athlete; (4) or a 50-year-old overhead athlete. Responses were recorded, and the cases were shuffled and sent back 6 weeks after the initial responses. Results were then analyzed using the Fleiss kappa coefficient (κ) to determine interobserver and intraobserver degrees of agreement. Results: There was moderate intraobserver reliability in both the Snyder and ESLAP classifications (κ = 0.52) and fair interobserver reliability for both classification systems (Snyder, κ = 0.31; ESLAP, κ = 0.30; P &lt; .0001) among all surgeons. Additionally, there was only fair agreement (κ = 0.30; P &lt; .0001) for the treatment modalities chosen by the reviewers for each case. Conclusion: This study demonstrated that SLAP tears remain a challenging problem for orthopaedic surgeons in diagnostics and treatment plans. Therefore, care should be taken in the preoperative discussion with the patient to consider all the possible treatment options because this may affect the postoperative recovery period and patient expectations.

The internal rotation traction radiograph does not improve the reliability in the AO classification system for pertrochanteric fractures. An inter- and intra-observer reliability assessment

IntroductionThe most universal method for classifying pertrochanteric fractures is the AO/OTA classification. These fractures are classified into different categories according to the features found in the anteroposterior radiograph of the hip. Anteroposterior radiograph of the hip with internal rotation traction can improve the characterization of the fracture. Inter- and intra-observer reliability in any classification is essential to achieve a homogeneous agreement for decision making. Our objective is assessing the overall reliability and by level of experience of the new AO/OTA classification of pertrochanteric fractures. Materials and methodsA hospital registry was used to collect patients with pertrochanteric hip fracture who had anteroposterior radiograph of the hip with and without internal rotation traction. We selected six evaluators stratified by levels of expertise in orthopedic trauma, leaving three groups: advanced, intermediate and beginner. Radiographs were sent through electronic forms and inter- and intra-observer reliability was calculated using the kappa (K) statistic. Results115 (one hundred fifteen) patients were included, each with their corresponding anteroposterior radiograph of the hip with and without internal rotation traction. Overall inter- and intra-observer reliability was moderate on both anteroposterior radiographs of the hip with and without internal rotation traction. Regarding the different levels of experience, the advanced level group reached a substantial inter- and intra-observer reliability in both anteroposterior radiographs with and without traction, while the rest of the groups with lower level of experience obtained a lesser reliability. ConclusionOur study found that the internal rotation traction x-ray did not improve the reliability of the new AO/OTA classification for pertrochanteric fractures, as assessed by inter- and intra-observer agreement, in either the overall group or in groups divided by experience level.

A stretchable hardness sensor for the assessment of skin disease in systemic sclerosis

ObjectiveTo determine the validity of a hardness sensor to objectively assess skin induration in patients with systemic sclerosis, and to compare the hardness sensor with the modified Rodnan skin score...

Abstract 7 — Deep Learning Differentiation of Inflammatory Lesions in Sacroiliac Joint MRI Based on Spondyloarthritis Research Consortium of Canada (SPARCC) System

Objective To develop a deep learning algorithm for grading sacroiliitis based on SPARCC in magnetic resonance imaging (MRI). Method A total of 996 images with inflammatory lesions from 210 participants with MRI sacroiliitis were used for training and validation. The testing cohort consisted of 18 participants with and 19 without MRI sacroiliitis. One hundred and fifty four images from the testing cohort had inflammatory lesions identified by a pre-trained algorithm from our previous study[1]. The ground truth was defined by manually outlined regions of interests (ROIs) consisting of bone marrow edema (BME) at the sacroiliac joint. The performance of the deep learning pipeline in predicting the SPARCC score was compared to manual interpretation by two experienced readers. Result The intra-observer reliability and the Pearson coefficient between the SPARCC scores from two experienced readers and the deep learning pipeline were 0.83 and 0.86, respectively. The sensitivities in identifying all inflammatory lesions, deep lesions, and intense lesions were 0.83, 0.79 and 0.81, respectively. The Dice coefficients of the sacrum and ilium segmentation were 0.82 and 0.80, respectively. The accuracies of identifying the SI joint and reference vessel were 0.90 and 0.88, respectively. Conclusion The performance of AI algorithms in SPARCC scoring was compatible with manual scoring by experienced readers. This proposed deep learning pipeline could be the first demonstration of a complete and SPARCC-informed deep-learning approach in scoring STIR images in SpA.