Identifying Anterior Cruciate Ligament Injuries Through Automated Video Analysis of In-Game Motion Patterns.

Background: Failure to diagnose anterior cruciate ligament (ACL) injury during a game can delay adequate treatment and increase the risk of further injuries. Artificial intelligence (AI) has the potential to be an accurate, cost-efficient, and readily available diagnostic tool for ACL injury in in-game situations. Purpose: To develop an automated video analysis system that uses AI to identify biomechanical patterns associated with ACL injury and to evaluate whether the system can enhance the ability of orthopaedic and sports medicine specialists to identify ACL injuries on video. Study Design: Descriptive laboratory study. Methods: A total of 91 ACL injury and 38 control movement scenes from online available match recordings were analyzed. The videos were processed to identify and track athletes and to estimate their 3-dimensional (3D) poses. Geometric features, including knee flexion, knee and hip abduction, and foot and hip rotation, were extracted from the athletes' 3D poses. A recurrent neural network algorithm was trained to classify ACL injury, using these engineered features as its input. Analysis by 2 orthopaedic surgeons examined whether providing clinical experts with the reconstructed 3D poses and their derived signals could increase their diagnostic accuracy. Results: All AI models performed significantly better than chance. The best model, which used the long short-term memory network with engineered features, demonstrated decision interpretability and good performance (F1 score = 0.63 ± 0.01, area under the receiver operating characteristic curve = 0.88 ± 0.01). The analysis by the 2 orthopaedic surgeons demonstrated improved diagnostic accuracy for ACL injury recognition when provided with system data, resulting in a 0.08 increase in combined F1 scores. Conclusion: Our approach successfully reconstructed the 3D motion of athletes from a single-camera view and derived geometry-based biomechanical features from pose sequences. Our trained AI model was able to automatically detect ACL injuries with relatively good performance and prelabel and highlight regions of interest in video footage. Clinical Relevance: This study demonstrated the feasibility of using AI to automatically evaluate in-game video footage and identify dangerous motion patterns. Further research can explore the full potential of the biomechanical markers and use of the system by nonspecialists, potentially diminishing the rate of missed diagnosis and the detrimental outcomes that follow.

Right ventricular contraction patterns in healthy children using three-dimensional echocardiography.

Background: The right ventricle (RV) has complex geometry and function, with motion along three separate axes-longitudinal, radial, and anteroposterior. Quantitative assessment of RV function by two-dimension echocardiography (2DE) has been limited as a consequence of this complexity, whereas newer three dimensional (3D) analysis offers the potential for more comprehensive assessment of the contributors to RV function. The aims of this study were to quantify the longitudinal, radial and anteroposterior components of global RV function using 3D echocardiography in a cohort of healthy children and to examine maturational changes in these parameters. Methods: Three-dimensional contours of the RV were generated from a cohort of healthy pediatric patients with structurally normal hearts at two centers. Traditional 2D and 3D echo characteristics were recorded. Using offline analysis of 3D datasets, RV motion was decomposed into three components, and ejection fractions (EF) were calculated (longitudinal-LEF; radial-REF; and anteroposterior-AEF). The individual decomposed EF values were indexed against the global RVEF. Strain values were calculated as well. Results: Data from 166 subjects were included in the analysis; median age was 13.5 years (range 0 to 17.4 years). Overall, AEF was greater than REF and LEF (29.2 ± 6.2% vs. 25.1 ± 7.2% and 25.7 ± 6.0%, respectively; p < 0.001). This remained true when indexed to overall EF (49.8 ± 8.7% vs. 43.3 ± 11.6% and 44.4 ± 10%, respectively; p < 0.001). Age-related differences were present for global RVEF, REF, and all components of RV strain. Conclusions: In healthy children, anteroposterior shortening is the dominant component of RV contraction. Evaluation of 3D parameters of the RV in children is feasible and enhances the overall understanding of RV function, which may allow improvements in recognition of dysfunction and assessment of treatment effects in the future.

Assessment of Right Ventricular Mechanics by 3D Transesophageal Echocardiography in the Early Phase of Acute Respiratory Distress Syndrome.

Aim: To compare global and axial right ventricular ejection fraction in ventilated patients for moderate-to-severe acute respiratory distress syndrome (ARDS) secondary to early SARS-CoV-2 pneumonia or to other causes, and in ventilated patients without ARDS used as reference. Methods: Retrospective single-center cross-sectional study including 64 ventilated patients: 21 with ARDS related to SARS-CoV-2 (group 1), 22 with ARDS unrelated to SARS-CoV-2 (group 2), and 21 without ARDS (control group). Real-time three-dimensional transesophageal echocardiography was performed for hemodynamic assessment within 24 h after admission. Contraction pattern of the right ventricle was decomposed along the three anatomically relevant axes. Relative contribution of each spatial axis was evaluated by calculating ejection fraction along each axis divided by the global right ventricular ejection fraction. Results: Global right ventricular ejection fraction was significantly lower in group 2 than in both group 1 and controls [median: 43% (25-75th percentiles: 40-57) vs. 58% (55-62) and 65% (56-68), respectively: p < 0.001]. Longitudinal shortening had a similar relative contribution to global right ventricular ejection fraction in all groups [group 1: 32% (28-39), group 2: 29% (24-40), control group: 31% (28-38), p = 0.6]. Radial shortening was lower in group 2 when compared to both group 1 and controls [45% (40-53) vs. 57% (51-62) and 56% (50-60), respectively: p = 0.005]. The relative contribution of right ventricular shortening along the anteroposterior axis was not statistically different between groups [group 1: 51% (41-55), group 2: 56% (46-63), control group; 56% (50-64), p = 0.076]. Conclusion: During early hemodynamic assessment, the right ventricular systolic function appears more impaired in ARDS unrelated to SARS-CoV-2 when compared to early stage SARS-CoV-2 ARDS. Radial shortening appears more involved than longitudinal and anteroposterior shortening in patients with ARDS unrelated to SARS-CoV-2 and decreased right ventricular ejection fraction.

Partitioning the Right Ventricle Into 15 Segments and Decomposing Its Motion Using 3D Echocardiography-Based Models: The Updated ReVISION Method.

Three main mechanisms contribute to global right ventricular (RV) function: longitudinal shortening, radial displacement of the RV free wall (bellows effect), and anteroposterior shortening (as a consequence of left ventricular contraction). Since the importance of these mechanisms may vary in different cardiac conditions, a technology being able to assess their relative influence on the global RV pump function could help to clarify the pathophysiology and the mechanical adaptation of the chamber. Previously, we have introduced our 3D echocardiography (3DE)-based solution-the Right VentrIcular Separate wall motIon quantificatiON (ReVISION) method-for the quantification of the relative contribution of the three aforementioned mechanisms to global RV ejection fraction (EF). Since then, our approach has been applied in several clinical scenarios, and its strengths have been demonstrated in the in-depth characterization of RV mechanical pattern and the prognostication of patients even in the face of maintained RV EF. Recently, various new features have been implemented in our software solution to enable the convenient, standardized, and more comprehensive analysis of RV function. Accordingly, in our current technical paper, we aim to provide a detailed description of the latest version of the ReVISION method with special regards to the volumetric partitioning of the RV and the calculation of longitudinal, circumferential, and area strains using 3DE datasets. We also report the results of the comparison between 3DE- and cardiac magnetic resonance imaging-derived RV parameters, where we found a robust agreement in our advanced 3D metrics between the two modalities. In conclusion, the ReVISION method may provide novel insights into global and also segmental RV function by defining parameters that are potentially more sensitive and predictive compared to conventional echocardiographic measurements in the context of different cardiac diseases.

Sex-Specific Patterns of Mortality Predictors Among Patients Undergoing Cardiac Resynchronization Therapy: A Machine Learning Approach.

Background: The relative importance of variables explaining sex-related differences in outcomes is scarcely explored in patients undergoing cardiac resynchronization therapy (CRT). We sought to implement and evaluate machine learning (ML) algorithms for the prediction of 1- and 3-year all-cause mortality in CRT patients. We also aimed to assess the sex-specific differences in predictors of mortality utilizing ML. Methods: Using a retrospective registry of 2,191 CRT patients, ML models were implemented in 6 partially overlapping patient subsets (all patients, females, or males with 1- or 3-year follow-up). Each cohort was randomly split into training (80%) and test sets (20%). After hyperparameter tuning in the training sets, the best performing algorithm was evaluated in the test sets. Model discrimination was quantified using the area under the receiver-operating characteristic curves (AUC). The most important predictors were identified using the permutation feature importances method. Results: Conditional inference random forest exhibited the best performance with AUCs of 0.728 (0.645-0.802) and 0.732 (0.681-0.784) for the prediction of 1- and 3-year mortality, respectively. Etiology of heart failure, NYHA class, left ventricular ejection fraction, and QRS morphology had higher predictive power, whereas hemoglobin was less important in females compared to males. The importance of atrial fibrillation and age increased, while the importance of serum creatinine decreased from 1- to 3-year follow-up in both sexes. Conclusions: Using ML techniques in combination with easily obtainable clinical features, our models effectively predicted 1- and 3-year all-cause mortality in CRT patients. Sex-specific patterns of predictors were identified, showing a dynamic variation over time.

Importance of Nonlongitudinal Motion Components in Right Ventricular Function: Three-Dimensional Echocardiographic Study in Healthy Volunteers.

BACKGROUND: Global right ventricular (RV) function is determined by the interplay of different motion components related to the myofiber architecture, and the relative importance of these components is still not thoroughly characterized. The aims of this study were to quantify the relative contributions of longitudinal, radial, and anteroposterior motion components to global RV function and to examine their determining factors in a large cohort of healthy volunteers using three-dimensional echocardiography. METHODS: Three hundred healthy adults with a balanced age range and an equal sex distribution were investigated at two centers. A three-dimensional mesh model of the right ventricle was generated, and its motion was decomposed along the three anatomically relevant axes. Multiplicative relative contributions were measured by dividing the ejection fraction (EF) values generated by shortening in the longitudinal, radial, and anteroposterior directions by global RV EF (longitudinal EF index [LEFi], radial EF index [REFi], and anteroposterior EF index, respectively). The circumferential contribution was defined as shortening in the radial and anteroposterior directions, omitting only longitudinal shortening. RESULTS: Circumferential EF index was markedly higher compared with LEFi (79 ± 7% vs 47 ± 9%, P < .001). LEFi (47 ± 9%) and anteroposterior EF index (49 ± 7%) were found to be similar in the pooled population, whereas REFi (44 ± 10%) was lower (P < .001). In younger individuals (20-39 years of age), the relative contribution of longitudinal shortening was significantly higher compared with the radial component; however, in the older age groups, LEFi and REFi were comparable. Age, body surface area, heart rate, and RV end-diastolic volume were independent predictors of LEFi and REFi, but all with opposite effects on the two motion directions. CONCLUSIONS: In contrast to the traditional viewpoint, the contributions of the radial and anteroposterior motion directions may be of comparable significance with that of longitudinal shortening in determining global RV function. Standard parameters referring only to longitudinal shortening of the right ventricle may be inadequate to characterize RV function thoroughly.

Right ventricular mechanical pattern in patients undergoing mitral valve surgery: a predictor of post-operative dysfunction?

AIMS: The PREPARE-MVR study (PRediction of Early PostoperAtive Right vEntricular failure in Mitral Valve Replacement/Repair patients) sought to investigate the alterations of right ventricular (RV) contraction pattern in patients undergoing mitral valve replacement/repair (MVR) and to explore the associations between pre-operative RV mechanics and early post-operative RV dysfunction (RVD). METHODS AND RESULTS: We prospectively enrolled 42 patients (63 ± 11 years, 69% men) undergoing open-heart MVR. Transthoracic three-dimensional (3D) echocardiography was performed pre-operatively, at intensive care unit discharge, and 6 months after surgery. The 3D model of the RV was reconstructed, and RV ejection fraction (RVEF) was calculated. We decomposed the motion of the ventricle to compute longitudinal ejection fraction (LEF) and radial ejection fraction (REF). Pulmonary artery catheterization was performed to monitor RV stroke work index (RVSWi). RVEF was slightly decreased after MVR [52 (50-55) vs. 51 (46-54)%; P = 0.001], whereas RV contraction pattern changed notably. Before MVR, the longitudinal shortening was the main contributor to global systolic RV function [LEF/RVEF vs. REF/RVEF; 0.53 (0.47-0.58) vs. 0.33 (0.22-0.42); P < 0.001]. Post-operatively, the radial motion became dominant [0.33 (0.28-0.43) vs. 0.46 (0.37-0.51); P = 0.004]. However, this shift was temporary as 6 months later the two components contributed equally to global RV function [0.44 (0.38-0.50) vs. 0.41 (0.36-0.49); P = 0.775]. Pre-operative LEF was an independent predictor of post-operative RVD defined as RVSWi < 300 mmHg⋅mL/m2 [OR = 1.33 (95% CI: 1.08-1.77), P < 0.05]. CONCLUSIONS: MVR induces a significant shift in the RV mechanical pattern. Advanced indices of RV mechanics are associated with invasively measured parameters of RV contractility and may predict post-operative RVD.

Machine learning-based mortality prediction of patients undergoing cardiac resynchronization therapy: the SEMMELWEIS-CRT score.

AIMS: Our aim was to develop a machine learning (ML)-based risk stratification system to predict 1-, 2-, 3-, 4-, and 5-year all-cause mortality from pre-implant parameters of patients undergoing cardiac resynchronization therapy (CRT). METHODS AND RESULTS: Multiple ML models were trained on a retrospective database of 1510 patients undergoing CRT implantation to predict 1- to 5-year all-cause mortality. Thirty-three pre-implant clinical features were selected to train the models. The best performing model [SEMMELWEIS-CRT score (perSonalizEd assessMent of estiMatEd risk of mortaLity With machinE learnIng in patientS undergoing CRT implantation)], along with pre-existing scores (Seattle Heart Failure Model, VALID-CRT, EAARN, ScREEN, and CRT-score), was tested on an independent cohort of 158 patients. There were 805 (53%) deaths in the training cohort and 80 (51%) deaths in the test cohort during the 5-year follow-up period. Among the trained classifiers, random forest demonstrated the best performance. For the prediction of 1-, 2-, 3-, 4-, and 5-year mortality, the areas under the receiver operating characteristic curves of the SEMMELWEIS-CRT score were 0.768 (95% CI: 0.674-0.861; P < 0.001), 0.793 (95% CI: 0.718-0.867; P < 0.001), 0.785 (95% CI: 0.711-0.859; P < 0.001), 0.776 (95% CI: 0.703-0.849; P < 0.001), and 0.803 (95% CI: 0.733-0.872; P < 0.001), respectively. The discriminative ability of our model was superior to other evaluated scores. CONCLUSION: The SEMMELWEIS-CRT score (available at semmelweiscrtscore.com) exhibited good discriminative capabilities for the prediction of all-cause death in CRT patients and outperformed the already existing risk scores. By capturing the non-linear association of predictors, the utilization of ML approaches may facilitate optimal candidate selection and prognostication of patients undergoing CRT implantation.

Exercise-induced shift in right ventricular contraction pattern: novel marker of athlete's heart?

Data about the functional adaptation of the right ventricle (RV) to intense exercise are limited. Our aim was to characterize the RV mechanical pattern in top-level athletes using three-dimensional echocardiography. A total of 60 elite water polo athletes (19 ± 4 yr, 17 ± 6 h of training/wk, 50% women and 50% men) and 40 healthy sedentary control subjects were enrolled. We measured the RV end-diastolic volume index (RVEDVi) and ejection fraction (RVEF) using dedicated software. Furthermore, we determined RV global longitudinal (RV GLS) and circumferential strain (RV GCS) and the relative contribution of longitudinal ejection fraction (LEF) and radial ejection fraction (REF) to RVEF using the ReVISION method. Athletes also underwent cardiopulmonary exercise testing [O2 consumption (V̇o2)/kg]. Athletes had significantly higher RVEDVi compared with control subjects (athletes vs. control subjects, 88 ± 11 vs. 65 ± 10 ml/m2, P < 0.001); however, they also demonstrated lower RVEF (56 ± 4% vs. 61 ± 5%, P < 0.001). RV GLS was comparable between the two groups (-22 ± 5% vs. -23 ± 5%, P = 0.24), whereas RV GCS was significantly lower in athletes (-21 ± 4% vs. -26 ± 7%, P < 0.001). Athletes had higher LEF and lower REF contribution to RVEF (LEF/RVEF: 0.50 ± 0.07 vs. 0.42 ± 0.07, P < 0.001; REF/RVEF: 0.33 ± 0.08 vs. 0.45 ± 0.08, P < 0.001). Moreover, the pattern of RV functional shift correlated with V̇o2/kg (LEF/RVEF: r = 0.30, P < 0.05; REF/RVEF: r = -0.27, P < 0.05). RV mechanical adaptation to long-term intense exercise implies a functional shift; the relative contribution of longitudinal motion to global function was increased, whereas the radial shortening was significantly decreased, in athletes. Moreover, this functional pattern correlates with aerobic exercise performance, representing a potential new resting marker of an athlete's heart. NEW & NOTEWORTHY Intensive regular physical exercise results in significant changes of right ventricular morphology and function. By separate quantification of the right ventricular longitudinal and radial function, a relative dominance of longitudinal motion and a decrease in radial motion can be observed compared with sedentary controls. Moreover, this contraction pattern correlates with cardiopulmonary fitness. According to these results, this functional shift of the right ventricle may represent a novel marker of an athlete's heart.

Dominance of free wall radial motion in global right ventricular function of heart transplant recipients.

Assessment of right ventricular (RV) function using conventional echocardiography might be inadequate as the radial motion of the RV free wall is often neglected. Our aim was to quantify the longitudinal and the radial components of RV function using three-dimensional (3D) echocardiography in heart transplant (HTX) recipients. Fifty-one HTX patients in stable cardiovascular condition without history of relevant rejection episode or chronic allograft vasculopathy and 30 healthy volunteers were enrolled. RV end-diastolic (EDV) volume and total ejection fraction (TEF) were measured by 3D echocardiography. Furthermore, we quantified longitudinal (LEF) and radial ejection fraction (REF) by decomposing the motion of the RV using the ReVISION method. RV EDV did not differ between groups (HTX vs control; 96 ± 27 vs 97 ± 2 mL). In HTX patients, TEF was lower, however, tricuspid annular plane systolic excursion (TAPSE) decreased to a greater extent (TEF: 47 ± 7 vs 54 ± 4% [-13%], TAPSE: 11 ± 5 vs 21 ± 4 mm [-48%], P < .0001). In HTX patients, REF/TEF ratio was significantly higher compared to LEF/TEF (REF/TEF vs LEF/TEF: 0.58 ± 0.10 vs 0.27 ± 0.08, P < .0001), while in controls the REF/TEF and LEF/TEF ratio was similar (0.45 ± 0.07 vs 0.47 ± 0.07). Current results confirm the superiority of radial motion in determining RV function in HTX patients. Parameters incorporating the radial motion are recommended to assess RV function in HTX recipients.

Quantification of the relative contribution of the different right ventricular wall motion components to right ventricular ejection fraction: the ReVISION method.

Three major mechanisms contribute to right ventricular (RV) pump function: (i) shortening of the longitudinal axis with traction of the tricuspid annulus towards the apex; (ii) inward movement of the RV free wall; (iii) bulging of the interventricular septum into the RV and stretching the free wall over the septum. The relative contribution of the aforementioned mechanisms to RV pump function may change in different pathological conditions.Our aim was to develop a custom method to separately assess the extent of longitudinal, radial and anteroposterior displacement of the RV walls and to quantify their relative contribution to global RV ejection fraction using 3D data sets obtained by echocardiography.Accordingly, we decomposed the movement of the exported RV beutel wall in a vertex based manner. The volumes of the beutels accounting for the RV wall motion in only one direction (either longitudinal, radial, or anteroposterior) were calculated at each time frame using the signed tetrahedron method. Then, the relative contribution of the RV wall motion along the three different directions to global RV ejection fraction was calculated either as the ratio of the given direction's ejection fraction to global ejection fraction and as the frame-by-frame RV volume change (∆V/∆t) along the three motion directions.The ReVISION (Right VentrIcular Separate wall motIon quantificatiON) method may contribute to a better understanding of the pathophysiology of RV mechanical adaptations to different loading conditions and diseases.