scholarly journals Application of a machine learning contouring tool for the evaluation of left ventricular strain in clinical practice

2021 ◽  
Vol 22 (Supplement_1) ◽  
Author(s):  
A Kenawy ◽  
MY Khanji ◽  
M Chirvasa ◽  
K Fung ◽  
A Sojoudi ◽  
...  
Keyword(s):  
Machine Learning ◽  
Clinical Practice ◽  
Strain Analysis ◽  
Left Ventricular ◽  
Cine Image ◽  
Time Saving ◽  
Development Fund ◽  
Research And Innovation ◽  
Manual Adjustment ◽  
The Impact

Abstract Funding Acknowledgements Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): AK has been funded by the Egyptian cultural centre and educational bureau of the Egyptian embassy in London and the Ministry of higher education in Egypt. SEP acknowledges support from the “SmartHeart” EPSRC programme grant (www.nihr.ac.uk; EP/P001009/1) and the London Medical Imaging and AI Centre for Value-Based Healthcare. This new centre is one of the UK Centres supported by a £50m investment from the Data to Early Diagnosis and Precision Medicine strand of the government’s Industrial Strategy Challenge Fund, managed and delivered by UK Research and Innovation (UKRI). SEP acknowledges support from the CAP-AI programme, London’s first AI enabling programme focused on stimulating growth in the capital’s AI Sector. CAP-AI is led by Capital Enterprise in partnership with Barts Health NHS Trust and Digital Catapult and is funded by the European Regional Development Fund and Barts Charity. SEP also acts as a paid consultant to Circle Cardiovascular Imaging Inc., Calgary, Canada and Servier onbehalf Barts Heart Centre, St Bartholomew’s Hospital, Barts Health NHS Trust, West Smithfield, London, UK Background Manual contouring of cardiovascular magnetic resonance (CMR) cine images remains common practice and the reference standard for left ventricular (LV) volumes and mass evaluation. However, it is time-consuming and machine learning (ML) may significantly reduce the time required for contouring. Accurate LV contours are the basis for reliable LV strain analysis using tissue tracking. Purpose To assess the impact of a ML contouring tool alone versus expert adjusted contours on LV strain. Methods We retrospectively selected 402 CMR studies with diagnoses of myocardial infarction (n = 108), myocarditis (n = 130) and healthy controls (n = 164) from the Barts BioResource between January 2015 to June 2018. CMR examinations were obtained using 1.5T and 3T scanners (Siemens Healthineers, Germany). We excluded 32 cases due to phase inconsistency between short (SAX) and long axes (LAX) cine images or suboptimal cine image quality. For the remaining 370 cases, steady state free precession cine images for LAX and SAX were analysed by the ML contouring tool (using CVI42 research prototype software 5.11). Manual expert adjustment for the contours was done for each case if considered suboptimal for strain analysis in the reference end-diastolic phase. Strain results from ML and expert adjusted ML methods were compared for strain agreement. Times taken by these methods were recorded and compared against the time taken for standard manual contouring. Results SAX and LAX derived strains by ML and expert adjusted ML methods showed good agreement by Bland-Altman analysis (Figure 1) with excellent coefficient of concordance using Kendall W which is 0.98 for global SAX, radial and circumferential strains (mean difference(MD) = -1.7% (lower and upper limits of agreement (UL,LL) -6.6,3.2), MD = 0.5% (-1.0,2.1)) and is 0.95 for global LAX derived strain (radial and longitudinal, MD = 0.7% (UL,LL -8.7 ,7.4),MD= 0.2% (-1.9,2.5), respectively). Time taken for adjustment of ML contours was significantly shorter than manual contouring (1.35 minutes vs 8.0 minutes, around 590% time saving in ML adjusted method). Conclusions ML contouring compared to expert manual adjustment has a clinically reasonable agreement when used for measuring LV strain. Also, using the ML tool with expert adjustment shows significant time saving for analysis and reporting time compared to entirely manual analysis, favouring its application in routine clinical practice. Abstract Figure.

scholarly journals Predicting the Progression of Mild Cognitive Impairment Using Machine Learning: A Systematic, Quantitative and Critical Review

2020 ◽  
Author(s):  
Manon Ansart ◽  
Stephane Epelbaum ◽  
Giulia Bassignana ◽  
Alexandre Bone ◽  
Simona Bottani ◽  
...  
Keyword(s):  
Machine Learning ◽  
Cognitive Impairment ◽  
Clinical Practice ◽  
Mild Cognitive Impairment ◽  
Parameter Tuning ◽  
Methodological Issues ◽  
Cognitive Assessments ◽  
Data Set ◽  
Mixed Effect ◽  
The Impact

We performed a systematic review of studies focusing on the automatic prediction of the progression of mild cognitive impairment to Alzheimer's disease (AD) dementia, and a quantitative analysis of the methodological choices impacting performance. This review included 172 articles, from which 234 experiments were extracted. For each of them, we reported the used data set, the feature types, the algorithm type, performance and potential methodological issues. The impact of these characteristics on the performance was evaluated using a multivariate mixed effect linear regressions. We found that using cognitive, fluorodeoxyglucose-positron emission tomography or potentially electroencephalography and magnetoencephalography variables significantly improved predictive performance compared to not including them, whereas including other modalities, in particular T1 magnetic resonance imaging, did not show a significant effect. The good performance of cognitive assessments questions the wide use of imaging for predicting the progression to AD and advocates for exploring further fine domain-specific cognitive assessments. We also identified several methodological issues, including the absence of a test set, or its use for feature selection or parameter tuning in nearly a fourth of the papers. Other issues, found in 15% of the studies, cast doubts on the relevance of the method to clinical practice. We also highlight that short-term predictions are likely not to be better than predicting that subjects stay stable over time. These issues highlight the importance of adhering to good practices for the use of machine learning as a decision support system for the clinical practice.

scholarly journals Explainable Machine-Learning Predictions To Support Personalized Cardiology Strategies

2021 ◽  
Author(s):  
De Rong Loh ◽  
Si Yong Yeo ◽  
Ru San Tan ◽  
Fei Gao ◽  
Angela S Koh
Keyword(s):  
Physical Activity ◽  
Machine Learning ◽  
Older Adults ◽  
Mean Squared Error ◽  
Absolute Error ◽  
Left Ventricular ◽  
Physical Parameters ◽  
Activity Levels ◽  
Physical Activity Levels ◽  
The Impact

Abstract Aims A widely practiced intervention to modify cardiac health, the effect of physical activity on older adults is likely heterogeneous. While machine learning (ML) models that combine various systemic signals may aid in predictive modeling, the inability to rationalize predictions at a patient personalized level is a major shortcoming in the current field of ML. Methods and Results We applied a novel methodology, Shapley Additive Explanations (SHAP), on a dataset of older adults n = 86 (mean age 72 ± 4 years) whose physical activity levels were studied alongside changes in their left ventricular (LV) structure. SHAP was tested to provide intelligible visualization on the magnitude of the impact of the features in their physical activity levels on their LV structure. As proof of concept, using repeated K-cross validation on the train set (n = 68), we found the Random Forest Regressor with the most optimal hyperparameters, which achieved the lowest mean squared error. With the trained model, we evaluated its performance by reporting its mean absolute error and plotting the correlation on the test set (n = 18). Based on collective force plot, individually numbered patients are indicated on the horizontal axis, and each bandwidth implies the magnitude (i.e., effect) of physical parameters (higher in red; lower in blue) towards prediction of their LV structure. Conclusions As a tool that identified specific features in physical activity that predicted cardiac structure on a per patient level, our findings support a role for explainable ML to be incorporated into personalized cardiology strategies.

Abstract 4340: Left Ventricular Dyssynchrony Measured by Speckle-Tracking Radial Strain Analysis Predicts Survival After Resynchronization Therapy

Circulation ◽  
2008 ◽  
Vol 118 (suppl_18) ◽  
Author(s):  
Rutger J Van Bommel ◽  
Victoria Delgado ◽  
Claudia Ypenburg ◽  
Sjoerd A Mollema ◽  
C. Jan Willem Borleffs ◽  
...  
Keyword(s):  
Heart Failure ◽  
Speckle Tracking ◽  
Radial Strain ◽  
Strain Analysis ◽  
Left Ventricular ◽  
Cardiac Resynchronization ◽  
Resynchronization Therapy ◽  
Cox Regression Analysis ◽  
Lv Dyssynchrony ◽  
The Impact

Since the introduction of cardiac resynchronization therapy (CRT) in heart failure patients, many echocardiographic criteria, including left ventricular (LV) dyssynchrony, have been investigated in improving selection of suitable candidates. A novel method for the assessment of LV dyssynchrony is speckle-tracking radial strain analysis. Aim of this study was to investigate the impact of pre-implantation, speckle-tracking derived LV dyssynchrony on survival in patients treated with CRT A total of 537 consecutive patients undergoing CRT at our center were included. In all patients, speckle-tracking radial strain analysis was applied to standard LV short-axis images. Significant LV dyssynchrony was defined as a delay between the anteroseptal and posterior segments ≥130 ms. The primary endpoint was all-cause mortality Mean LV dyssynchrony in all 537 patients was 138±105 ms and 251 patients (47%) had predefined significant LV dyssynchrony ≥130 ms. For survival analysis, mean follow-up in the study population was 34±20 months. Within this period 145 patients (27%) died. Main cause of death remained heart failure (61% of all deaths). Multivariate Cox regression analysis with correction for age, gender, etiology, QRS duration, NYHA class, quality of life score, distance covered in the 6-minute walking test, LV volumes and LVEF, demonstrated a significant survival benefit for patients with significant LV dyssynchrony ≥130 ms (HR 0.63, 95% C.I. 0.43–0.92, p=0.017, Figure ). Presence of significant LV mechanical dyssynchrony, measured with speckle-tracking radial strain, is associated with improved survival probability after CRT

scholarly journals P1383 The effects of doxorubicin on left and right ventricular strain in patients with lymphoma: insights from a retrospective study

2020 ◽  
Vol 21 (Supplement_1) ◽  
Author(s):  
S Vahabi ◽  
M Stewart ◽  
A Kasim ◽  
H Hancock ◽  
M Norouzi ◽  
...  
Keyword(s):  
Global Longitudinal Strain ◽  
Radial Strain ◽  
Strain Analysis ◽  
Left Ventricular ◽  
Research Association ◽  
Significant Deterioration ◽  
Development Fund ◽  
Monitoring Strategies ◽  
Right Ventricular Strain ◽  
The Right

Abstract Funding Acknowledgements South Tees Research and Development Fund (UK) Background Anthracyclines are a cornerstone in the management of lymphoma. However, their use is associated with cardiotoxicity. Speckle tracking echocardiography (STE) has been established as a valid measure of quantifying cardiac function. However, most studies to this date have focused predominantly on left ventricular (LV) global longitudinal strain (GLS) with only a limited number assessing the right ventricle (RV) and other LV strain parameters. Purpose Using 2D STE, we assessed the effects of anthracyclines on LV and RV strain parameters, focusing on LV endocardial (GLS), LV myocardial GLS (myoGLS), LV radial strain (GRS), RV endocardial (RV GLS), myocardial GLS (RV myoGLS), and RV free wall strain (RVFWS). Methods We retrospectively collected data on patients treated for lymphoma between 2015-2018. Two groups (G) were defined: those with a conventional drop in LV ejection fraction (EF), (G1, n = 11) and those without (G2, n = 24). Echocardiograms were performed pre-chemotherapy (T0), mid-treatment (T1), and post-chemotherapy (T2) and were analysed offline using vendor-independent software (TomTec 2D CPA). LV and RV strain analysis was performed in both groups. This study was ethically approved by Health Research Association (REC Reference 18/SS/0139). Results Mean age was 61 ± 16 years (G1) and 65 ± 12 years (G2). 18% (G1) and 17% (G2) of patients had a history of IHD in each group. Mean cumulative dose of doxorubicin was 280 ± 31 mg/m2 (G1) and 280± 48mg/m2 (G2). In both groups, there was no significant change in LV or RV strain parameters from T0 to T1. In G1, between T1 and T2, patients exhibited a significant deterioration in LV GLS (-19.7 ± 2.6% vs.-15.6 ± 2.5% p < 0.0005), and LV myoGLS (-17.3 ± 2.2% vs. -14.1 ± 2.9% p = 0.02). There was also a measurable decline in RV strain parameters between T1 to T2 (RV GLS, -23.1 ± 4.7% vs. -18.8 ± 4.2% p = 0.028) and (RV myoGLS -21.5 ± 5.2% vs -17.3 ± 3.6% p = 0.013). When analysed from T0 to T2, the changes in RV strain were more marked; RV GLS (-25.2 ± 4.9% vs. -18.8 ± 4.2% p = 0.005), RV myoGLS (-22.4 ± 5.2% vs-17.3 ± 3.6% p = 0.005), and RVFWS (-28.8 ± 5.7% vs. -20.9 ± 6.7% p = 0.001). In G2, no change was observed in LV GLS (-20.4 ± 2.3% vs. -19.6 ± 2.8% p = 0.66), LV myoGLS (-18.8 ± 2.5% vs. -17.5 ± 3.1% p = 0.18), RV GLS (-24.2 ± 2.3% vs. -23.1 ± 2.7% p = 0.42), RV myoGLS (-22.4 ± 2.9% vs. -20.6 ± 2.8% p = 0.09), RVFWS (-26.7 ± 4.6% vs. -25.2 ± 3.6% p = 1.0). GRS did not significantly change in either G1 or G2 during treatment. Conclusion In patients with a reduction in LVEF, this analysis demonstrated a significant reduction in LV strain parameters secondary to anthracycline treatment. Novel strain parameters did not change in the normal EF group, or predate EF/GLS decline in those with reduced LVEF. Measurable RV dysfunction was noted in those with LV deterioration, highlighting the global cardiac insult of anthracycline treatment. Preventative and monitoring strategies in cardio-oncology should not overlook RV function.

Stratifying the prognostic capability of cardiovascular magnetic resonance in severe aortic stenosis: a machine learning approach

2020 ◽  
Vol 41 (Supplement_2) ◽  
Author(s):  
S Kwak ◽  
R Everett ◽  
T Ko ◽  
H Lee ◽  
W Lee ◽  
...  
Keyword(s):  
Machine Learning ◽  
Cardiovascular Magnetic Resonance ◽  
Magnetic Resonance ◽  
Aortic Stenosis ◽  
Volume Fraction ◽  
Left Ventricular ◽  
Funding Source ◽  
Ventricular Ejection Fraction ◽  
Increased Risk ◽  
The Impact

Abstract Background Cardiovascular magnetic resonance (CMR) demonstrates promise in improving patient risk stratification in aortic stenosis (AS). We explored whether machine learning might provide further insights into the prognostic capability of CMR parameters. Methods Severe AS patients (n=440) undergoing AVR were prospectively enrolled across 10 international sites, and CMR performed prior to AVR. A machine learning prediction model using a random survival forest (RSF) was trained with 29 variables, including 13 CMR, 4 echocardiography, and 12 clinical parameters, using post-AVR mortality as an outcome. The impact of the important variables on the outcome (partial dependency) was examined. Results The most predictive CMR parameters in the RSF model were the extracellular volume fraction (ECV%), followed by right ventricular ejection fraction (RVEF), late gadolinium enhancement (LGE%), and indexed left ventricular end-diastolic volume (LVEDVi). Regarding the partial effects, the predicted mortality increased strongly once the ECV% exceeded 26.5% (Figure 1A). The LGE% was associated with an increased risk of mortality, which reached a plateau beyond the level of 2% (Figure 1C). There were U-shaped relationships between mortality and both RVEF and LVEDVi, with the lowest mortality seen at RVEF 70% and LVEDVi 68ml/m2 (Figure 1B, D). These trends of predicted outcomes by each variable were verified in the Kaplan-Meier curves and Cox analyses (Table). In both Cox and RSF models, the predictability was substantially increased when these four CMR parameters were added to conventional clinical risk factors. An AS-CMR risk score comprised of these four parameters presented a stepwise increase in mortality with increasing adverse CMR features (p<0.001). Conclusions Our machine learning analysis using RSF has identified ECV%, RVEF, LGE%, and LVEDVi as key prognostic markers in severe AS with a nonlinear influence of each parameter on mortality post-AVR. Figure 1 Funding Acknowledgement Type of funding source: Public grant(s) – National budget only. Main funding source(s): This study was supported by grants from the Korean Health Technology R & D Project, Ministry of Health, Welfare & Family Affairs, Republic of Korea (HI16C0225 and HI15C0399) and the National Institute for Health Research (NIHR) infrastructure at Leeds.

A study of the effects of hail, snow and PM on Potential Gradient

2020 ◽  
Author(s):  
Konstantinos Kourtidis ◽  
Athanassios Karagioras ◽  
Eleni Papadopoulou ◽  
Nikos Mihalopoulos ◽  
Iasonas Stavroulas
Keyword(s):  
Potential Gradient ◽  
Atmospheric Composition ◽  
The European Union ◽  
Development Fund ◽  
European Regional Development Fund ◽  
Research And Innovation ◽  
Innovation Infrastructure ◽  
Preliminary Study ◽  
The Impact ◽  
European Regional

<p>We present here the study of six hail events and five snow events in Xanthi, N. Greece, on Potential Gradient (PG). All hail events occurred in the spring-summer season of the years 2011-2018. A decrease in PG has been observed which has been around 2000-3000 V/m during the three hail events which occurred concurrently with rain. In three events with no rain, the decrease has been varying between 60 and 6000 V/m. In the case of only 60 V/m drop, no concurrent drop in temperature has been observed, while for the other cases it appears that for each degree drop in temperature the drop in PG is 1000 V/m, hence it appears that the intensity of the hail event regulates the drop in PG, although we do not have hail amount measurements to validate this. Regarding snow events,  the situation is more complicated, with PG fluctuating rapidly between high positive and high negative values. We present also a preliminary study of the impact of PM1.0 and PM2.5 on PG from measurements performed during 2019. We acknowledge support of this work by the project “PANhellenic infrastructure for Atmospheric Composition and climatE change” (MIS 5021516) which is implemented under the Action “Reinforcement of the Research and Innovation Infrastructure”, funded by the Operational Programme "Competitiveness, Entrepreneurship and Innovation" (NSRF 2014-2020) and co-financed by Greece and the European Union (European Regional Development Fund).</p>

scholarly journals Impact of apical foreshortening on deformation measurements: a report from the EACVI-ASE Strain Standardization Task Force

2019 ◽  
Author(s):  
Serkan Ünlü ◽  
Jürgen Duchenne ◽  
Oana Mirea ◽  
Efstathios D Pagourelias ◽  
Stéphanie Bézy ◽  
...  
Keyword(s):  
Ejection Fraction ◽  
Longitudinal Strain ◽  
Task Force ◽  
Strain Analysis ◽  
Left Ventricular ◽  
Lv Function ◽  
Substantial Impact ◽  
Endocardial Layer ◽  
Relative Change ◽  
The Impact

Abstract Aims Foreshortening of apical views is a common problem in echocardiography. It results in an abnormally thick false apex and a shortened left ventricular (LV) long axis. We sought to evaluate the impact of foreshortened (FS) on LV ejection fraction (LVEF) and layer-specific 2D speckle tracking based segmental (S) and global (G) longitudinal strain (LS) measurements. Methods and results We examined 72 participants using a GE Vivid E9 system. FS apical views were collected from an imaging window one rib-space higher than the optimal images. Ejection fraction as well as layer-specific GLS and SLS measurements were analysed by GE EchoPAC v201 and TomTec Image Arena 4.6 and compared between optimal and FS images. On average, LV long axis was 10% shorter in FS images than in optimal images. FS induced a relative change in LVEF of 3.3% and 6.9% for GE and TomTec, respectively (both, P < 0.001). Endocardial GLS was 9.0% higher with GE and 23.2% with TomTec (P < 0.001). Midwall GLS measurements were less affected (7.8% for GE and 14.1% for TomTec, respectively, both P < 0.001). Segmental strain analysis revealed that the mid-ventricular and apical segments were more affected by foreshortening, and endocardial measurements were more affected than midwall. Conclusion Optimal image geometry is crucial for accurate LV function assessment. Foreshorhening of apical views has a substantial impact on longitudinal strain measurements, predominantly in the apex and in the endocardial layer. Our data suggest that measuring midwall strain might therefore be the more robust approach for clinical routine use.

Translating Principles of Speech Science to Clinical Practice: Current and Future Trends in Craniofacial Disorders

2008 ◽  
Vol 18 (1) ◽  
pp. 31-40 ◽  
Author(s):  
David J. Zajac
Keyword(s):  
Clinical Practice ◽  
Speech Intelligibility ◽  
Current Practice ◽  
Computer Assisted ◽  
Behavioral Management ◽  
Single Word ◽  
Future Directions ◽  
Speech Science ◽  
The Impact ◽  
Opinion Article

Abstract The purpose of this opinion article is to review the impact of the principles and technology of speech science on clinical practice in the area of craniofacial disorders. Current practice relative to (a) speech aerodynamic assessment, (b) computer-assisted single-word speech intelligibility testing, and (c) behavioral management of hypernasal resonance are reviewed. Future directions and/or refinement of each area are also identified. It is suggested that both challenging and rewarding times are in store for clinical researchers in craniofacial disorders.

Sign in / Sign up

Export Citation Format

Share Document