Pub Date : 2019-09-01DOI: 10.23919/CinC49843.2019.9005881
Alejandro Costoya-Sánchez, A. Climent, I. Hernández-Romero, A. Liberos, F. Fernández‐Avilés, S. Narayan, F. Atienza, M. Guillem, M. Rodrigo
Location of reentrant electrical activity responsible for driving atrial fibrillation (AF) is key to ablative therapies. The aim of this work is to study the effect of the quality of the electrograms (EGMs) used for 3D phase analysis on reentrant activity identification, as well as to develop an algorithm capable of automatically identifying low- quality signals.EGMs signals from 259 episodes obtained from 29 AF patients were recorded using 64-electrode basket catheters. Low-quality EGMs were manually identified. Reentrant activity was identified in 3D phase maps and provided an area under the ROC curve (AUC) of 0.69 when compared to a 2D activation-based method. Reentries located in regions with poor-quality EGMs were then removed, increasing the AUC to 0.80. The EGM classification algorithm showed a similar performance both for low-quality EGM identification (sensitivity 0.91 and specificity 0.80) and for reentrant activity location with 3D phase analysis (AUC 0.80).Discard of reentrant activity identified in regions where EGMs showed low quality significantly improved the specificity of the 3D phase analysis. Besides, EGMs classification according to their quality proved to be possible using time and spectral domain parameters.
{"title":"Automatic Quality Electrogram Assessment Improves Reentrant Activity Identification in Atrial Fibrillation","authors":"Alejandro Costoya-Sánchez, A. Climent, I. Hernández-Romero, A. Liberos, F. Fernández‐Avilés, S. Narayan, F. Atienza, M. Guillem, M. Rodrigo","doi":"10.23919/CinC49843.2019.9005881","DOIUrl":"https://doi.org/10.23919/CinC49843.2019.9005881","url":null,"abstract":"Location of reentrant electrical activity responsible for driving atrial fibrillation (AF) is key to ablative therapies. The aim of this work is to study the effect of the quality of the electrograms (EGMs) used for 3D phase analysis on reentrant activity identification, as well as to develop an algorithm capable of automatically identifying low- quality signals.EGMs signals from 259 episodes obtained from 29 AF patients were recorded using 64-electrode basket catheters. Low-quality EGMs were manually identified. Reentrant activity was identified in 3D phase maps and provided an area under the ROC curve (AUC) of 0.69 when compared to a 2D activation-based method. Reentries located in regions with poor-quality EGMs were then removed, increasing the AUC to 0.80. The EGM classification algorithm showed a similar performance both for low-quality EGM identification (sensitivity 0.91 and specificity 0.80) and for reentrant activity location with 3D phase analysis (AUC 0.80).Discard of reentrant activity identified in regions where EGMs showed low quality significantly improved the specificity of the 3D phase analysis. Besides, EGMs classification according to their quality proved to be possible using time and spectral domain parameters.","PeriodicalId":6697,"journal":{"name":"2019 Computing in Cardiology (CinC)","volume":"34 1","pages":"Page 1-Page 4"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88028830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.23919/CinC49843.2019.9005667
Induparkavi Murugesan, K. Murugesan, Lingeshwaran Balasubramanian, Malathi Arumugam
Despite the rise of Artificial Intelligence (AI) algorithms and their applications in various fields, their utilizations in high-risk fields like healthcare and finance is limited because of the lack of interpretability of their inner workings. Some algorithms are interpretable, but not accurate, whereas some produce accurate results and not decipherable. Research is underway to explore the possibilities to interrogate an AI system, and ask why it makes certain decisions. This paper aims to investigate the decision-making process by AI algorithms in the prediction of sepsis based on patients’ clinical records.We were ranked 59 in the PhysioNet/Computing in Cardiology Challenge 2019 and the utility score obtained on the full test set is 0.131, and our team name was ARUL.
尽管人工智能(AI)算法的兴起及其在各个领域的应用，但由于其内部工作原理缺乏可解释性，它们在医疗保健和金融等高风险领域的应用受到限制。有些算法是可解释的，但不准确，而有些算法产生准确的结果，但不可破译。研究人员正在探索询问人工智能系统的可能性，并询问它为什么做出某些决定。本文旨在研究基于患者临床记录的AI算法在脓毒症预测中的决策过程。我们在2019年PhysioNet/Computing in Cardiology Challenge中排名第59位，在完整测试集上获得的效用得分为0.131，我们的团队名称为ARUL。
{"title":"Interpretation of Artificial Intelligence Algorithms in the Prediction of Sepsis","authors":"Induparkavi Murugesan, K. Murugesan, Lingeshwaran Balasubramanian, Malathi Arumugam","doi":"10.23919/CinC49843.2019.9005667","DOIUrl":"https://doi.org/10.23919/CinC49843.2019.9005667","url":null,"abstract":"Despite the rise of Artificial Intelligence (AI) algorithms and their applications in various fields, their utilizations in high-risk fields like healthcare and finance is limited because of the lack of interpretability of their inner workings. Some algorithms are interpretable, but not accurate, whereas some produce accurate results and not decipherable. Research is underway to explore the possibilities to interrogate an AI system, and ask why it makes certain decisions. This paper aims to investigate the decision-making process by AI algorithms in the prediction of sepsis based on patients’ clinical records.We were ranked 59 in the PhysioNet/Computing in Cardiology Challenge 2019 and the utility score obtained on the full test set is 0.131, and our team name was ARUL.","PeriodicalId":6697,"journal":{"name":"2019 Computing in Cardiology (CinC)","volume":"55 1","pages":"Page 1-Page 4"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86399389","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.23919/CinC49843.2019.9005634
L. Zhong, S. Leng, Xiaodan Zhao, R. Tan
Impairment of left ventricular (LV) longitudinal function is recognized as an independent predictor of cardiac events in patients with heart failure (HF). 1 Strain imaging derived from speckle tracking echocardiography or feature tracking cardiovascular magnetic resonance (CMR) 2 permit assessments of myocardial function in the longitudinal direction, however, specific competencies and time-consuming protocols are often needed. Therefore, we aim to investigate a fast, semi-automated, and vendor-independent approach for accurate determination of LV longitudinal strain from standard cine CMR images.
{"title":"Feature Tracking for Ventricular Strain Assessment in Heart Failure with Preserved Ejection Fraction","authors":"L. Zhong, S. Leng, Xiaodan Zhao, R. Tan","doi":"10.23919/CinC49843.2019.9005634","DOIUrl":"https://doi.org/10.23919/CinC49843.2019.9005634","url":null,"abstract":"Impairment of left ventricular (LV) longitudinal function is recognized as an independent predictor of cardiac events in patients with heart failure (HF). 1 Strain imaging derived from speckle tracking echocardiography or feature tracking cardiovascular magnetic resonance (CMR) 2 permit assessments of myocardial function in the longitudinal direction, however, specific competencies and time-consuming protocols are often needed. Therefore, we aim to investigate a fast, semi-automated, and vendor-independent approach for accurate determination of LV longitudinal strain from standard cine CMR images.","PeriodicalId":6697,"journal":{"name":"2019 Computing in Cardiology (CinC)","volume":"2004 1","pages":"1-2"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82944193","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.23919/CinC49843.2019.9005624
O. Lahdenoja, Tero Hurnanen, Juho Koskinen, M. Kaisti, Kim Munck, S. Schmidt, T. Koivisto, Mikko Pänkäälä
Previously, using single-axis accelerometers, it has been proposed that in conditions such as traumatic brain injury (TBI) the brain pulsation signal characteristics change, potentially due to changes induced by the impact to the brain. In this paper, we aim to validate the use of a custom built embedded measurement system towards the analysis of the head pulsation signals. The system comprises of several synchronized high sampling rate 3-axis accelerometers and a simultaneous chest ECG. In our case three accelerometers on the surface of human head are used (in left temple, forehead and right temple), while the subject were in supine position. To illustrate that a proper signal quality may be extracted, we derive heart rate (HR) and heart rate variability (HRV) from each sensor and each axis for each of five healthy male volunteers. The results are reported against ECG as the ground truth. This study will build ground for further clinical trial utilizing multi-axial accelerometers to study both healthy and diseased subjects (e.g. TBI patients).
{"title":"Head Pulsation Signal Analysis for 3-Axis Head-Worn Accelerometers","authors":"O. Lahdenoja, Tero Hurnanen, Juho Koskinen, M. Kaisti, Kim Munck, S. Schmidt, T. Koivisto, Mikko Pänkäälä","doi":"10.23919/CinC49843.2019.9005624","DOIUrl":"https://doi.org/10.23919/CinC49843.2019.9005624","url":null,"abstract":"Previously, using single-axis accelerometers, it has been proposed that in conditions such as traumatic brain injury (TBI) the brain pulsation signal characteristics change, potentially due to changes induced by the impact to the brain. In this paper, we aim to validate the use of a custom built embedded measurement system towards the analysis of the head pulsation signals. The system comprises of several synchronized high sampling rate 3-axis accelerometers and a simultaneous chest ECG. In our case three accelerometers on the surface of human head are used (in left temple, forehead and right temple), while the subject were in supine position. To illustrate that a proper signal quality may be extracted, we derive heart rate (HR) and heart rate variability (HRV) from each sensor and each axis for each of five healthy male volunteers. The results are reported against ECG as the ground truth. This study will build ground for further clinical trial utilizing multi-axial accelerometers to study both healthy and diseased subjects (e.g. TBI patients).","PeriodicalId":6697,"journal":{"name":"2019 Computing in Cardiology (CinC)","volume":"10 1","pages":"Page 1-Page 4"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89214836","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
To get early prediction of sepsis, we propose to extract more time-dependent characteristics that retain the temporal evolvement information of the underlying biomedical dynamic system, including differential, integration, time-dependent statistics, variations and convolutions.Considering that two categories are unbalanced in the training set, we employed easy ensemble algorithm to get multiple base learners. As for the base learner, we tried three models: random forest, XGBoost and LightGBM. By boosting the results of multiple base learners, we constructed our ensemble model.Our team which name is njuedu ranked 25th in the official test and scored 0.282 in full test set.Since the submitted model version only used training set A to train our model, the model had a higher score of 0.401 in test set A, and 0.278 in test set B, and only -0.207 points in test set C.
{"title":"Using Features Extracted From Vital Time Series for Early Prediction of Sepsis","authors":"Qiang Yu, Xiaolin Huang, Weifeng Li, Cheng Wang, Ying Chen, Yun Ge","doi":"10.23919/CinC49843.2019.9005646","DOIUrl":"https://doi.org/10.23919/CinC49843.2019.9005646","url":null,"abstract":"To get early prediction of sepsis, we propose to extract more time-dependent characteristics that retain the temporal evolvement information of the underlying biomedical dynamic system, including differential, integration, time-dependent statistics, variations and convolutions.Considering that two categories are unbalanced in the training set, we employed easy ensemble algorithm to get multiple base learners. As for the base learner, we tried three models: random forest, XGBoost and LightGBM. By boosting the results of multiple base learners, we constructed our ensemble model.Our team which name is njuedu ranked 25th in the official test and scored 0.282 in full test set.Since the submitted model version only used training set A to train our model, the model had a higher score of 0.401 in test set A, and 0.278 in test set B, and only -0.207 points in test set C.","PeriodicalId":6697,"journal":{"name":"2019 Computing in Cardiology (CinC)","volume":"1 1","pages":"Page 1-Page 4"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88576055","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.23919/CinC49843.2019.9005870
R. Vullings
Congenital heart disease (CHD) is one of the main problems that can occur during pregnancy. Annually, 300.000 babies die during pregnancy or infancy because of CHD. Early detection of CHD leads to reduced mortality and morbidity, but is hampered by the relatively low detection rates (i.e. <60%) of current CHD screening technology. This detection rate could be improved by complementing echocardiographic screening with assessment of the fetal electrocardiogram (ECG).In this study, the fetal ECG was measured non-invasively, with electrodes on the maternal abdomen, in almost 400 fetuses, 30% of which had known CHD. The fetal ECG measurements were processed to yield a 3-dimensional fetal vectorcardiogram. A deep neural network was trained to classify this fetal vectorcardiogram as either originating from a healthy fetus or CHD. The network was evaluated on a test set of about 100 patients, showing a CHD detection accuracy of 76%. Non-invasive fetal electrocardiography therefore shows clear potential in diagnosis of CHD and should be considered as supplementary technology next to echocardiography.
{"title":"Fetal Electrocardiography and Deep Learning for Prenatal Detection of Congenital Heart Disease","authors":"R. Vullings","doi":"10.23919/CinC49843.2019.9005870","DOIUrl":"https://doi.org/10.23919/CinC49843.2019.9005870","url":null,"abstract":"Congenital heart disease (CHD) is one of the main problems that can occur during pregnancy. Annually, 300.000 babies die during pregnancy or infancy because of CHD. Early detection of CHD leads to reduced mortality and morbidity, but is hampered by the relatively low detection rates (i.e. <60%) of current CHD screening technology. This detection rate could be improved by complementing echocardiographic screening with assessment of the fetal electrocardiogram (ECG).In this study, the fetal ECG was measured non-invasively, with electrodes on the maternal abdomen, in almost 400 fetuses, 30% of which had known CHD. The fetal ECG measurements were processed to yield a 3-dimensional fetal vectorcardiogram. A deep neural network was trained to classify this fetal vectorcardiogram as either originating from a healthy fetus or CHD. The network was evaluated on a test set of about 100 patients, showing a CHD detection accuracy of 76%. Non-invasive fetal electrocardiography therefore shows clear potential in diagnosis of CHD and should be considered as supplementary technology next to echocardiography.","PeriodicalId":6697,"journal":{"name":"2019 Computing in Cardiology (CinC)","volume":"49 1","pages":"Page 1-Page 4"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90580406","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.23919/CinC49843.2019.9005876
M. Vizcardo, A. Ravelo, Miriam Manrique, P. Gomis
Chagas disease American trypanosomiasis is caused by a flagellated parasite: Trypanosoma cruzi, transmitted by an insect of the genus Triatoma and also by blood transfusions. In Latin America, the number of infected people is approximately 6 million, with a population exposed to the risk of infection of 550000. It is our interest to develop a non-invasive and low-cost methodology, capable of detecting any early cardiac alteration that also allows us to see dysautononia or dysfunction within 24 hours and with this it could be used to detect any cardiac alteration caused by T early Cruzi. For this, we analyzed the 24- hour Holter ECG records in 107 patients with ECG abnormalities (CH2), 102 patients without ECG alterations (CH1) who had positive serological results for Chagas disease and 83 volunteers without positive serological results for Chagas disease (CONTROL). Approximate entropy was used to quantify the regularity of electrocardiograms (ECG) in the three groups. We analyzed 288 ECG segments per patient. Significant differences were found between the CONTROL-CH1, CONTROL-CH2 and CH1- CH2 groups.
{"title":"Application of the Entropy of Approximation for the nonlinear characterization in patients with Chagas Disease","authors":"M. Vizcardo, A. Ravelo, Miriam Manrique, P. Gomis","doi":"10.23919/CinC49843.2019.9005876","DOIUrl":"https://doi.org/10.23919/CinC49843.2019.9005876","url":null,"abstract":"Chagas disease American trypanosomiasis is caused by a flagellated parasite: Trypanosoma cruzi, transmitted by an insect of the genus Triatoma and also by blood transfusions. In Latin America, the number of infected people is approximately 6 million, with a population exposed to the risk of infection of 550000. It is our interest to develop a non-invasive and low-cost methodology, capable of detecting any early cardiac alteration that also allows us to see dysautononia or dysfunction within 24 hours and with this it could be used to detect any cardiac alteration caused by T early Cruzi. For this, we analyzed the 24- hour Holter ECG records in 107 patients with ECG abnormalities (CH2), 102 patients without ECG alterations (CH1) who had positive serological results for Chagas disease and 83 volunteers without positive serological results for Chagas disease (CONTROL). Approximate entropy was used to quantify the regularity of electrocardiograms (ECG) in the three groups. We analyzed 288 ECG segments per patient. Significant differences were found between the CONTROL-CH1, CONTROL-CH2 and CH1- CH2 groups.","PeriodicalId":6697,"journal":{"name":"2019 Computing in Cardiology (CinC)","volume":"17 1","pages":"Page 1-Page 4"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89570317","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.23919/CinC49843.2019.9005679
M. Macedo, Dario Augusto Borges Oliveira, M. A. Gutierrez
Coronary artery disease (CAD) is intrinsically related to presence of atherosclerotic plaques. The rupture of this plaques is responsible for most acute coronary events. Intracoronary optical coherence tomography (IOCT) enables a detailed high-resolution visualization of micro-structural changes of the arterial wall in vivo. In this paper, we introduce a new way of identifying atherosclerotic plaques using 1D Convolutional Neural Networks (CNN) analyzing only the lumen contour. Training and test were performed with 1600 IOCT frames from in vivo patients. In our tests, we achieved f1-score of 95% for atherosclerotic plaque recognition. The results allow us to report an interesting correlation between the lumen contour geometry and the presence of plaques in the vascular wall observed through IOCT exams. The use of lumen contour for plaque detection opens two new perspectives: assisting specialists in the task of detecting plaques visually by paying special attention to the lumen and allowing methods to work in real time to detect plaques using efficient methods that use less information and deliver accurate results.
{"title":"Atherosclerotic Plaques Recognition in Intracoronary Optical Images Using Neural Networks","authors":"M. Macedo, Dario Augusto Borges Oliveira, M. A. Gutierrez","doi":"10.23919/CinC49843.2019.9005679","DOIUrl":"https://doi.org/10.23919/CinC49843.2019.9005679","url":null,"abstract":"Coronary artery disease (CAD) is intrinsically related to presence of atherosclerotic plaques. The rupture of this plaques is responsible for most acute coronary events. Intracoronary optical coherence tomography (IOCT) enables a detailed high-resolution visualization of micro-structural changes of the arterial wall in vivo. In this paper, we introduce a new way of identifying atherosclerotic plaques using 1D Convolutional Neural Networks (CNN) analyzing only the lumen contour. Training and test were performed with 1600 IOCT frames from in vivo patients. In our tests, we achieved f1-score of 95% for atherosclerotic plaque recognition. The results allow us to report an interesting correlation between the lumen contour geometry and the presence of plaques in the vascular wall observed through IOCT exams. The use of lumen contour for plaque detection opens two new perspectives: assisting specialists in the task of detecting plaques visually by paying special attention to the lumen and allowing methods to work in real time to detect plaques using efficient methods that use less information and deliver accurate results.","PeriodicalId":6697,"journal":{"name":"2019 Computing in Cardiology (CinC)","volume":"3 1","pages":"Page 1-Page 4"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89265843","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.23919/CinC49843.2019.9005791
César Navarro, M. Kurth, M. Ruddock, S. Fishlock, J. Mclaughlin
Our previous work demonstrated that algorithms combining high sensitivity cardiac troponin T (hs-cTnT) and heart-type fatty acid-binding protein (H-FABP) may help in ruling out Acute Myocardial Infarction (AMI). For those algorithms, the hs-cTnT thresholds were adopted from the ESC guidelines. This time, we present a data-driven approach that also explores hs-cTnT thresholds.The results show a significant improvement when compared to previous algorithms reported. Using a cohort of n = 360 patients (288 Non-AMI and 72 AMI), a rule-out algorithm used at presentation identified more low-risk patients who presented with chest pain of suspected cardiac origin than the standard ESC algorithm: (199/288 (69.1%) vs. 83/288 (28.8%) (p <0.0005)), respectively.According to our data, our algorithm at the emergency department, would identify additional non-AMI patients in comparison to the ESC algorithm, potentially reducing the number of hospital admissions by 42%.
{"title":"An Algorithm Based on Combining hs-cTnT and H-FABP for Ruling Out Acute Myocardial Infarction","authors":"César Navarro, M. Kurth, M. Ruddock, S. Fishlock, J. Mclaughlin","doi":"10.23919/CinC49843.2019.9005791","DOIUrl":"https://doi.org/10.23919/CinC49843.2019.9005791","url":null,"abstract":"Our previous work demonstrated that algorithms combining high sensitivity cardiac troponin T (hs-cTnT) and heart-type fatty acid-binding protein (H-FABP) may help in ruling out Acute Myocardial Infarction (AMI). For those algorithms, the hs-cTnT thresholds were adopted from the ESC guidelines. This time, we present a data-driven approach that also explores hs-cTnT thresholds.The results show a significant improvement when compared to previous algorithms reported. Using a cohort of n = 360 patients (288 Non-AMI and 72 AMI), a rule-out algorithm used at presentation identified more low-risk patients who presented with chest pain of suspected cardiac origin than the standard ESC algorithm: (199/288 (69.1%) vs. 83/288 (28.8%) (p <0.0005)), respectively.According to our data, our algorithm at the emergency department, would identify additional non-AMI patients in comparison to the ESC algorithm, potentially reducing the number of hospital admissions by 42%.","PeriodicalId":6697,"journal":{"name":"2019 Computing in Cardiology (CinC)","volume":"175 1","pages":"Page 1-Page 4"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75395132","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-09-01DOI: 10.23919/CinC49843.2019.9005516
Shivnarayan Patidar
Early prediction of sepsis is of utmost importance to provide optimal care at an early stage. This work aims to use machine learning for early prediction of sepsis using ratio and power-based feature transformation. The feature transformation and feature selection process is optimized by applying a genetic algorithm (GA) based approach to extract the information specific to the sepsis from the given raw patient covariates that maximizes the underlying classification performance in terms of utility score. The proposed method begins with filling the missing values in the training dataset. Then, GA is applied strategically to identify influential ratio and power-based features from the raw patient covariates. The utility score is maximized as an objective of the optimization. RusBoost is used with default settings for underlying classification during optimization. Subsequently, an optimal RusBoost model is developed with a set of 55 identified features. Independent performance evaluation of the proposed method with the 2019 PhysioNet/CinC Challenge dataset has officially achieved 19th rank with a utility score of 30.9% on the full hidden test data. This work appears as Shivpatidar on the leader-board. The proposed early warning system has potential clinical value in critical care clinics.
{"title":"Diagnosis of Sepsis Using Ratio Based Features","authors":"Shivnarayan Patidar","doi":"10.23919/CinC49843.2019.9005516","DOIUrl":"https://doi.org/10.23919/CinC49843.2019.9005516","url":null,"abstract":"Early prediction of sepsis is of utmost importance to provide optimal care at an early stage. This work aims to use machine learning for early prediction of sepsis using ratio and power-based feature transformation. The feature transformation and feature selection process is optimized by applying a genetic algorithm (GA) based approach to extract the information specific to the sepsis from the given raw patient covariates that maximizes the underlying classification performance in terms of utility score. The proposed method begins with filling the missing values in the training dataset. Then, GA is applied strategically to identify influential ratio and power-based features from the raw patient covariates. The utility score is maximized as an objective of the optimization. RusBoost is used with default settings for underlying classification during optimization. Subsequently, an optimal RusBoost model is developed with a set of 55 identified features. Independent performance evaluation of the proposed method with the 2019 PhysioNet/CinC Challenge dataset has officially achieved 19th rank with a utility score of 30.9% on the full hidden test data. This work appears as Shivpatidar on the leader-board. The proposed early warning system has potential clinical value in critical care clinics.","PeriodicalId":6697,"journal":{"name":"2019 Computing in Cardiology (CinC)","volume":"21 1","pages":"Page 1-Page 4"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85068622","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：