Machine learning aided jump height estimate democratization through smartphone measures.

Introduction: The peak height reached in a countermovement jump is a well established performance parameter. Its estimate is often entrusted to force platforms or body-worn inertial sensors. To date, smartphones may possibly be used as an alternative for estimating jump height, since they natively embed inertial sensors. Methods: For this purpose, 43 participants performed 4 countermovement jumps (172 in total) on two force platforms (gold standard). While jumping, participants held a smartphone in their hands, whose inertial sensor measures were recorded. After peak height was computed for both instrumentations, twenty-nine features were extracted, related to jump biomechanics and to signal time-frequency characteristics, as potential descriptors of soft tissues or involuntary arm swing artifacts. A training set (129 jumps - 75%) was created by randomly selecting elements from the initial dataset, the remaining ones being assigned to the test set (43 jumps - 25%). On the training set only, a Lasso regularization was applied to reduce the number of features, avoiding possible multicollinearity. A multi-layer perceptron with one hidden layer was trained for estimating the jump height from the reduced feature set. Hyperparameters optimization was performed on the multi-layer perceptron using a grid search approach with 5-fold cross validation. The best model was chosen according to the minimum negative mean absolute error. Results: The multi-layer perceptron greatly improved the accuracy (4 cm) and precision (4 cm) of the estimates on the test set with respect to the raw smartphone measures estimates (18 and 16 cm, respectively). Permutation feature importance was performed on the trained model in order to establish the influence that each feature had on the outcome. The peak acceleration and the braking phase duration resulted the most influential features in the final model. Despite not being accurate enough, the height computed through raw smartphone measures was still among the most influential features. Discussion: The study, implementing a smartphone-based method for jump height estimates, paves the way to method release to a broader audience, pursuing a democratization attempt.

Estimating the Standing Long Jump Length from Smartphone Inertial Sensors through Machine Learning Algorithms.

The length of the standing long jump (SLJ) is widely recognized as an indicator of developmental motor competence or sports conditional performance. This work aims at defining a methodology to allow athletes/coaches to easily measure it using the inertial measurement units embedded on a smartphone. A sample group of 114 trained young participants was recruited and asked to perform the instrumented SLJ task. A set of features was identified based on biomechanical knowledge, then Lasso regression allowed the identification of a subset of predictors of the SLJ length that was used as input of different optimized machine learning architectures. Results obtained from the use of the proposed configuration allow an estimate of the SLJ length with a Gaussian Process Regression model with a RMSE of 0.122 m in the test phase, Kendall's τ < 0.1. The proposed models give homoscedastic results, meaning that the error of the models does not depend on the estimated quantity. This study proved the feasibility of using low-cost smartphone sensors to provide an automatic and objective estimate of SLJ performance in ecological settings.

IABP versus Impella Support in Cardiogenic Shock: "In Silico" Study.

Cardiogenic shock (CS) is part of a clinical syndrome consisting of acute left ventricular failure causing severe hypotension leading to inadequate organ and tissue perfusion. The most commonly used devices to support patients affected by CS are Intra-Aortic Balloon Pump (IABP), Impella 2.5 pump and Extracorporeal Membrane Oxygenation. The aim of this study is the comparison between Impella and IABP using CARDIOSIM© software simulator of the cardiovascular system. The results of the simulations included baseline conditions from a virtual patient in CS followed by IABP assistance in synchronised mode with different driving and vacuum pressures. Subsequently, the same baseline conditions were supported by the Impella 2.5 with different rotational speeds. The percentage variation with respect to baseline conditions was calculated for haemodynamic and energetic variables during IABP and Impella assistance. The Impella pump driven with a rotational speed of 50,000 rpm increased the total flow by 4.36% with a reduction in left ventricular end-diastolic volume (LVEDV) by ≅15% to ≅30%. A reduction in left ventricular end systolic volume (LVESV) by ≅10% to ≅18% (≅12% to ≅33%) was observed with IABP (Impella) assistance. The simulation outcome suggests that assistance with the Impella device leads to higher reduction in LVESV, LVEDV, left ventricular external work and left atrial pressure-volume loop area compared to IABP support.

Modelling and Simulation of the Combined Use of IABP and Impella as a Rescue Procedure in Cardiogenic Shock: An Alternative for Non-Transplant Centres?

The treatment of critically ill patients remains an evolving and controversial issue. Mechanical circulatory support through a percutaneous approach for the management of cardiogenic shock has taken place in recent years. The combined use of IABP and the Impella 2.5 device may have a role to play for this group of patients. A simulation approach may help with a quantitative assessment of the achievable degree of assistance. In this paper, we analyse the interaction between the Impella 2.5 pump applied as the LVAD and IABP using the numerical simulator of the cardiovascular system developed in our laboratory. Starting with pathological conditions reproduced using research data, the simulations were performed by setting different rotational speeds for the LVAD and driving the IABP in full mode (1:1) or partial mode (1:2, 1:3 and 1:4). The effects induced by drug administration during the assistance were also simulated. The haemodynamic parameters under investigation were aa follows: mean aortic pressure, systolic and diastolic aortic pressure, mean pulmonary arterial pressure, mean left and right atrial pressure, cardiac output, cardiac index, left and right ventricular end-systolic volume, left ventricular end-diastolic volume and mean coronary blood flow. The energetic variables considered in this study were as follows: left and right ventricular external work and left and right atrial pressure-volume area. The outcome of our simulations shows that the combined use of IABP and Impella 2.5 achieves adequate support in the acute phase of cardiogenic shock compared to each standalone device. This would allow further stabilisation and transfer to a transplant centre should the escalation of treatment be required.

Evaluation of Different Cannulation Strategies for Aortic Arch Surgery Using a Cardiovascular Numerical Simulator.

Aortic disease has a significant impact on quality of life. The involvement of the aortic arch requires the preservation of blood supply to the brain during surgery. Deep hypothermic circulatory arrest is an established technique for this purpose, although neurological injury remains high. Additional techniques have been used to reduce risk, although controversy still remains. A three-way cannulation approach, including both carotid arteries and the femoral artery or the ascending aorta, has been used successfully for aortic arch replacement and redo procedures. We developed circuits of the circulation to simulate blood flow during this type of cannulation set up. The CARDIOSIM© cardiovascular simulation platform was used to analyse the effect on haemodynamic and energetic parameters and the benefit derived in terms of organ perfusion pressure and flow. Our simulation approach based on lumped-parameter modelling, pressure-volume analysis and modified time-varying elastance provides a theoretical background to a three-way cannulation strategy for aortic arch surgery with correlation to the observed clinical practice.

Ventricular and Atrial Pressure-Volume Loops: Analysis of the Effects Induced by Right Centrifugal Pump Assistance.

The main indications for right ventricular assist device (RVAD) support are right heart failure after implantation of a left ventricular assist device (LVAD) or early graft failure following heart transplantation. We sought to study the effects induced by different RVAD connections when right ventricular elastance (EesRIGHT) was modified using numerical simulations based on atrial and ventricular pressure-volume analysis. We considered the effects induced by continuous-flow RVAD support on left/right ventricular/atrial loops when EesRIGHT changed from 0.3 to 0.8 mmHg/mL during in-series or parallel pump connection. Pump rotational speed was also addressed. Parallel RVAD support at 4000 rpm with EesRIGHT = 0.3 mmHg/mL generated percentage changes up to 60% for left ventricular pressure-volume area and external work; up to 20% for left ventricular ESV and up to 25% for left ventricular EDV; up to 50% change in left atrial pressure-volume area (PVLAL-A) and only a 3% change in right atrial pressure-volume area (PVLAR-A). Percentage variation was lower when EesRIGHT = 0.8 mmHg/mL. Early recognition of right ventricular failure followed by aggressive treatment is desirable, so as to achieve a more favourable outcome. RVAD support remains an option for advanced right ventricular failure, although the onset of major adverse events may preclude its use.

CARDIOSIM©: The First Italian Software Platform for Simulation of the Cardiovascular System and Mechanical Circulatory and Ventilatory Support.

This review is devoted to presenting the history of the CARDIOSIM© software simulator platform, which was developed in Italy to simulate the human cardiovascular and respiratory systems. The first version of CARDIOSIM© was developed at the Institute of Biomedical Technologies of the National Research Council in Rome. The first platform version published in 1991 ran on a PC with a disk operating system (MS-DOS) and was developed using the Turbo Basic language. The latest version runs on PC with Microsoft Windows 10 operating system; it is implemented in Visual Basic and C++ languages. The platform has a modular structure consisting of seven different general sections, which can be assembled to reproduce the most important pathophysiological conditions. One or more zero-dimensional (0-D) modules have been implemented in the platform for each section. The different modules can be assembled to reproduce part or the whole circulation according to Starling's law of the heart. Different mechanical ventilatory and circulatory devices have been implemented in the platform, including thoracic artificial lungs, ECMO, IABPs, pulsatile and continuous right and left ventricular assist devices, biventricular pacemakers and biventricular assist devices. CARDIOSIM© is used in clinical and educational environments.

Telehealth: A winning weapon to face the COVID-19 outbreak for patients with pulmonary arterial hypertension.

BACKGROUND: COVID-19 pandemic severely affected national health systems, altering the modality and the type of care of patients with acute and chronic diseases. To minimize the risk of exposure to SARS-CoV2 for patients and health professionals, face-to-face visits were cancelled or postponed and the use of telemedicine was strongly encouraged. This reorganization involved especially patients with rare diseases needing periodic comprehensive assessment, such as pulmonary arterial hypertension (PAH). MAIN BODY: The paper reports a proposal of strategy adopted for patients followed at our PAH center in Rome, where patients management was diversified based on clinical risk according to the European Society of Cardiology/European Respiratory Society PH guidelines-derived score and the REVEAL 2.0 score. A close monitoring and support of these patients were made possible by policy changes reducing barriers to telehealth access and promoting the use of telemedicine. Synchronous/asynchronous modalities and remote monitoring were used to collect and transfer medical data in order to guide physicians in therapeutic-decision making. Conversely, the use of implantable monitors providing hemodynamic information and echocardiography-mobile devices wirelessly connecting was limited by the poor experience existing in this setting. Large surveys and clinical trials are welcome to test the potential benefit of the optimal balance between traditional PAH management and telemedicine opportunities. CONCLUSION: Italy was found unprepared to manage the dramatic effects caused by COVID-19 on healthcare systems. In this emergency situation telemedicine represented a promising tool especially in rare diseases as PAH, but was limited by its scattered availability and legal and ethical issues. Cohesive partnership of health care providers with regional public health officials is needed to prioritize PAH patients for telemedicine by dedicated tools.

ECMO Assistance during Mechanical Ventilation: Effects Induced on Energetic and Haemodynamic Variables.

BACKGROUND AND OBJECTIVE: Simulation in cardiovascular medicine may help clinicians understand the important events occurring during mechanical ventilation and circulatory support. During the COVID-19 pandemic, a significant number of patients have required hospital admission to tertiary referral centres for concomitant mechanical ventilation and extracorporeal membrane oxygenation (ECMO). Nevertheless, the management of ventilated patients on circulatory support can be quite challenging. Therefore, we sought to review the management of these patients based on the analysis of haemodynamic and energetic parameters using numerical simulations generated by a software package named CARDIOSIM©. METHODS: New modules of the systemic circulation and ECMO were implemented in CARDIOSIM© platform. This is a modular software simulator of the cardiovascular system used in research, clinical and e-learning environment. The new structure of the developed modules is based on the concept of lumped (0-D) numerical modelling. Different ECMO configurations have been connected to the cardiovascular network to reproduce Veno-Arterial (VA) and Veno-Venous (VV) ECMO assistance. The advantages and limitations of different ECMO cannulation strategies have been considered. We have used literature data to validate the effects of a combined ventilation and ECMO support strategy. RESULTS: The results have shown that our simulations reproduced the typical effects induced during mechanical ventilation and ECMO assistance. We focused our attention on ECMO with triple cannulation such as Veno-Ventricular-Arterial (VV-A) and Veno-Atrial-Arterial (VA-A) configurations to improve the hemodynamic and energetic conditions of a virtual patient. Simulations of VV-A and VA-A assistance with and without mechanical ventilation have generated specific effects on cardiac output, coupling of arterial and ventricular elastance for both ventricles, mean pulmonary pressure, external work and pressure volume area. CONCLUSION: The new modules of the systemic circulation and ECMO support allowed the study of the effects induced by concomitant mechanical ventilation and circulatory support. Based on our clinical experience during the COVID-19 pandemic, numerical simulations may help clinicians with data analysis and treatment optimisation of patients requiring both mechanical ventilation and circulatory support.

Computational Simulator Models and Invasive Hemodynamic Monitoring as Tools for Precision Medicine in Pulmonary Arterial Hypertension.

Precision medicine, providing the right therapeutic strategy for the right patient, could revolutionize management and prognosis of patients affected by cardiovascular diseases. Big data and artificial intelligence are pivotal for the realization of this ambitious design. In the setting of pulmonary arterial hypertension (PAH), the use of computational models and data derived from ambulatory implantable hemodynamic monitors could provide useful information for tailored treatment, as requested by precision medicine.

Intra-aortic balloon counterpulsation timing: A new numerical model for programming and training in the clinical environment.

BACKGROUND AND OBJECTIVE: The intra-aortic balloon pump (IABP) is the most widely available device for short-term mechanical circulatory support, often used to wean off cardiopulmonary bypass or combined with extra-corporeal membrane oxygenation support or as a bridge to a left ventricular assist device. Although based on a relatively simple principle, its complex interaction with the cardiovascular system remains challenging and open to debate. The aim of this work was focused on the development of a new numerical model of IABP. METHODS: The new model was implemented in CARDIOSIM©, which is a modular software simulator of the cardiovascular system used in research and e-learning environment. The IABP is inserted into the systemic bed divided in aortic, thoracic and two abdominal tracts modelled with resistances, inertances and compliances. The effect induced by the balloon is reproduced in each tract of the aorta by the presence of compliances connected to PIABP generator and resistances. PIABP generator reproduces the balloon pressure with the option to change IABP timing. We have used literature data to validate the potential of this new numerical model. RESULTS: The results have shown that our simulations reproduced the typical effects induced during IABP assistance. We have also simulated the effects induced by the device on the hemodynamic variables when the IABP ratio was set to 1:1, 1:2, 1:4 and 1:8. The outcome of these simulations is in accordance with literature data measured in the clinical environment. CONCLUSIONS: The new IABP module is easy to manage and can be used as a training tool in a clinical setting. Although based on literature data, the outcome of the simulations is encouraging. Additional work is ongoing with a view to further validate its features. The configuration of CARDIOSIM© presented in this work allows the simulation of the effects induced by mechanical ventilatory assistance. This facility may have significant importance in the management of patients affected by COVID-19 when they require mechanical circulatory support devices.