J. Zeng, Shen Lin, Zhigang Li, Runchen Sun, Xuexin Yu, Xiaocong Lian, Yan Zhao, Xiangyang Ji, Zhe Zheng
Cardiovascular disease (CVD) may not be detected in time with conventional clinical approaches. Abnormal gait patterns have been associated with pathological conditions and can be monitored continuously by gait video. We aim to test the association between non-contact, video-based gait information and general CVD status. Individuals undergoing confirmatory CVD evaluation were included in a prospective, cross-sectional study. Gait videos were recorded with a Kinect camera. Gait features were extracted from gait videos to correlate with the composite and individual components of CVD, including coronary artery disease, peripheral artery disease, heart failure, and cerebrovascular events. The incremental value of incorporating gait information with traditional CVD clinical variables was also evaluated. Three hundred fifty-two participants were included in the final analysis [mean (standard deviation) age, 59.4 (9.8) years; 25.3% were female]. Compared with the baseline clinical variable model [area under the receiver operating curve (AUC) 0.717, (0.690–0.743)], the gait feature model demonstrated statistically better performance [AUC 0.753, (0.726–0.780)] in predicting the composite CVD, with further incremental value when incorporated with the clinical variables [AUC 0.764, (0.741–0.786)]. Notably, gait features exhibited varied association with different CVD component conditions, especially for peripheral artery disease [AUC 0.752, (0.728–0.775)] and heart failure [0.733, (0.707–0.758)]. Additional analyses also revealed association of gait information with CVD risk factors and the established CVD risk score. We demonstrated the association and predictive value of non-contact, video-based gait information for general CVD status. Further studies for gait video-based daily living CVD monitoring are promising.
{"title":"Association between gait video information and general cardiovascular diseases: a prospective cross-sectional study","authors":"J. Zeng, Shen Lin, Zhigang Li, Runchen Sun, Xuexin Yu, Xiaocong Lian, Yan Zhao, Xiangyang Ji, Zhe Zheng","doi":"10.1093/ehjdh/ztae031","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae031","url":null,"abstract":"\u0000 \u0000 \u0000 Cardiovascular disease (CVD) may not be detected in time with conventional clinical approaches. Abnormal gait patterns have been associated with pathological conditions and can be monitored continuously by gait video. We aim to test the association between non-contact, video-based gait information and general CVD status.\u0000 \u0000 \u0000 \u0000 Individuals undergoing confirmatory CVD evaluation were included in a prospective, cross-sectional study. Gait videos were recorded with a Kinect camera. Gait features were extracted from gait videos to correlate with the composite and individual components of CVD, including coronary artery disease, peripheral artery disease, heart failure, and cerebrovascular events. The incremental value of incorporating gait information with traditional CVD clinical variables was also evaluated. Three hundred fifty-two participants were included in the final analysis [mean (standard deviation) age, 59.4 (9.8) years; 25.3% were female]. Compared with the baseline clinical variable model [area under the receiver operating curve (AUC) 0.717, (0.690–0.743)], the gait feature model demonstrated statistically better performance [AUC 0.753, (0.726–0.780)] in predicting the composite CVD, with further incremental value when incorporated with the clinical variables [AUC 0.764, (0.741–0.786)]. Notably, gait features exhibited varied association with different CVD component conditions, especially for peripheral artery disease [AUC 0.752, (0.728–0.775)] and heart failure [0.733, (0.707–0.758)]. Additional analyses also revealed association of gait information with CVD risk factors and the established CVD risk score.\u0000 \u0000 \u0000 \u0000 We demonstrated the association and predictive value of non-contact, video-based gait information for general CVD status. Further studies for gait video-based daily living CVD monitoring are promising.\u0000","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"65 25","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141121523","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}
A. de Vos, Sophie Troost, Anke Waterschoot, N. Pijls, Marcel van ‘t Veer
Continuous infusion thermodilution is an established technique for the assessment of absolute coronary blood flow and microvascular resistance due to its proven accuracy and reproducibility. However, for this technique to yield reliable measurements, direct and homogenous mixing of injected saline and blood is mandatory. This study aimed to assess and compare the mixing properties of two different microcatheters, namely the Rayflow® and the Finecross® catheter, which are commonly used in the catheterization laboratory. The study employed three different methods to evaluate the mixing properties of the catheters. Firstly, a qualitative assessment of mixing was done using ink injections in an in-vitro bench model of a coronary artery. Secondly, in analogy to the human catheterization laboratory, mixing properties over the length of the coronary artery were assessed semi-quantitatively by temperature measurements in the bench model. Lastly, a quantitative assessment was performed by 3D computational fluid dynamics, where the standard deviation and entropy ratio of the temperature over the cross-section in the coronary artery model were calculated for both catheters. All three evaluation methods demonstrated that the Rayflow catheter's specific design leads to a more optimal, homogeneous mixture of blood and saline over both the cross-section and length of a coronary vessel, as compared to the standard end-hole catheter.
{"title":"Mixing properties of coronary infusion catheters assessed by in-vitro experiments and computational fluid dynamics","authors":"A. de Vos, Sophie Troost, Anke Waterschoot, N. Pijls, Marcel van ‘t Veer","doi":"10.1093/ehjdh/ztae033","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae033","url":null,"abstract":"\u0000 Continuous infusion thermodilution is an established technique for the assessment of absolute coronary blood flow and microvascular resistance due to its proven accuracy and reproducibility. However, for this technique to yield reliable measurements, direct and homogenous mixing of injected saline and blood is mandatory. This study aimed to assess and compare the mixing properties of two different microcatheters, namely the Rayflow® and the Finecross® catheter, which are commonly used in the catheterization laboratory.\u0000 The study employed three different methods to evaluate the mixing properties of the catheters. Firstly, a qualitative assessment of mixing was done using ink injections in an in-vitro bench model of a coronary artery. Secondly, in analogy to the human catheterization laboratory, mixing properties over the length of the coronary artery were assessed semi-quantitatively by temperature measurements in the bench model. Lastly, a quantitative assessment was performed by 3D computational fluid dynamics, where the standard deviation and entropy ratio of the temperature over the cross-section in the coronary artery model were calculated for both catheters.\u0000 All three evaluation methods demonstrated that the Rayflow catheter's specific design leads to a more optimal, homogeneous mixture of blood and saline over both the cross-section and length of a coronary vessel, as compared to the standard end-hole catheter.","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"13 14","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140967015","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}
Jiesuck Park, Joonghee Kim, Si-Hyuck Kang, Jina Lee, Youngtaek Hong, Hyuk-Jae Chang, Youngjin Cho, Y. Yoon
The clinical feasibility of artificial intelligence (AI)-based electrocardiography (ECG) analysis for predicting obstructive coronary artery disease (CAD) has not been sufficiently validated in patients with stable angina, especially in large sample sizes. A deep learning framework for quantitative ECG (QCG) analysis was trained and internally tested to derive risk scores (0–100) for obstructive CAD (QCGObstCAD) and extensive CAD (QCGExtCAD) using 50,756 ECG images from 21,866 patients who underwent coronary artery evaluation for chest pain (invasive coronary or computed tomography angiography). External validation was performed in 4,517 patients with stable angina who underwent coronary imaging to identify obstructive CAD. QCGObstCAD and QCGExtCAD scores were significantly increased in the presence of obstructive and extensive CAD (all p < 0.001), and with increasing degrees of stenosis and disease burden, respectively (all ptrend < 0.001). In internal and external tests, QCGObstCAD exhibited good predictive ability for obstructive CAD (area under the curve [AUC], 0.781 and 0.731, respectively) and severe obstructive CAD (AUC, 0.780 and 0.786, respectively), and QCGExtCAD exhibited good predictive ability for extensive CAD (AUC, 0.689 and 0.784). In the external test, QCGObstCAD and QCGExtCAD scores demonstrated independent and incremental predictive value for obstructive and extensive CAD, respectively, over that with conventional clinical risk factors. QCG scores demonstrated significant associations with lesion characteristics, such as the fractional flow reserve, coronary calcification score, and total plaque volume. AI-based QCG analysis for predicting obstructive CAD in patients with stable angina, including those with severe stenosis and multivessel disease, is feasible.
{"title":"AI-Enhanced Electrocardiography Analysis as a Promising Tool for Predicting Obstructive Coronary Artery Disease in Patients with Stable Angina","authors":"Jiesuck Park, Joonghee Kim, Si-Hyuck Kang, Jina Lee, Youngtaek Hong, Hyuk-Jae Chang, Youngjin Cho, Y. Yoon","doi":"10.1093/ehjdh/ztae038","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae038","url":null,"abstract":"\u0000 \u0000 \u0000 The clinical feasibility of artificial intelligence (AI)-based electrocardiography (ECG) analysis for predicting obstructive coronary artery disease (CAD) has not been sufficiently validated in patients with stable angina, especially in large sample sizes.\u0000 \u0000 \u0000 \u0000 A deep learning framework for quantitative ECG (QCG) analysis was trained and internally tested to derive risk scores (0–100) for obstructive CAD (QCGObstCAD) and extensive CAD (QCGExtCAD) using 50,756 ECG images from 21,866 patients who underwent coronary artery evaluation for chest pain (invasive coronary or computed tomography angiography). External validation was performed in 4,517 patients with stable angina who underwent coronary imaging to identify obstructive CAD.\u0000 \u0000 \u0000 \u0000 QCGObstCAD and QCGExtCAD scores were significantly increased in the presence of obstructive and extensive CAD (all p < 0.001), and with increasing degrees of stenosis and disease burden, respectively (all ptrend < 0.001). In internal and external tests, QCGObstCAD exhibited good predictive ability for obstructive CAD (area under the curve [AUC], 0.781 and 0.731, respectively) and severe obstructive CAD (AUC, 0.780 and 0.786, respectively), and QCGExtCAD exhibited good predictive ability for extensive CAD (AUC, 0.689 and 0.784). In the external test, QCGObstCAD and QCGExtCAD scores demonstrated independent and incremental predictive value for obstructive and extensive CAD, respectively, over that with conventional clinical risk factors. QCG scores demonstrated significant associations with lesion characteristics, such as the fractional flow reserve, coronary calcification score, and total plaque volume.\u0000 \u0000 \u0000 \u0000 AI-based QCG analysis for predicting obstructive CAD in patients with stable angina, including those with severe stenosis and multivessel disease, is feasible.\u0000","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"31 42","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140980269","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}
Slobodan Calic, J. Jortveit, Jahn Otto Andersen, Christian Hesbø Eek
Percutaneous coronary intervention (PCI) of chronic total occlusions (CTO) has a lower success rate and a higher complication rate compared to PCI of non-occluded coronary arteries. Co-operation and supervision by a more experienced operator (proctoring) is associated with improved success of CTO-procedures. To assess the feasibility of remote proctoring using web-based communication and mixed reality technology in CTO-procedures. The PCI operator was equipped with a Microsoft HoloLens 2 head mounted display enabling visual and verbal interaction including holographic annotations with a remote proctor. Ten CTO-procedures were performed by a single PCI operator assisted by a remote proctor. Audio and video communication was successfully established in all procedures. All procedures were possible to perform with a Microsoft HoloLens 2 head mounted display. The PCI-operator experienced the remote proctoring as useful. Remote proctoring of CTO-procedures using mixed reality technology was feasible. The impact of the method regarding procedural and patient outcomes needs to be assessed in new studies.
{"title":"Remote proctoring in complex percutaneous coronary intervention aided by mixed reality technology","authors":"Slobodan Calic, J. Jortveit, Jahn Otto Andersen, Christian Hesbø Eek","doi":"10.1093/ehjdh/ztae037","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae037","url":null,"abstract":"\u0000 \u0000 \u0000 Percutaneous coronary intervention (PCI) of chronic total occlusions (CTO) has a lower success rate and a higher complication rate compared to PCI of non-occluded coronary arteries. Co-operation and supervision by a more experienced operator (proctoring) is associated with improved success of CTO-procedures.\u0000 \u0000 \u0000 \u0000 To assess the feasibility of remote proctoring using web-based communication and mixed reality technology in CTO-procedures.\u0000 \u0000 \u0000 \u0000 The PCI operator was equipped with a Microsoft HoloLens 2 head mounted display enabling visual and verbal interaction including holographic annotations with a remote proctor.\u0000 \u0000 \u0000 \u0000 Ten CTO-procedures were performed by a single PCI operator assisted by a remote proctor. Audio and video communication was successfully established in all procedures. All procedures were possible to perform with a Microsoft HoloLens 2 head mounted display. The PCI-operator experienced the remote proctoring as useful.\u0000 \u0000 \u0000 \u0000 Remote proctoring of CTO-procedures using mixed reality technology was feasible. The impact of the method regarding procedural and patient outcomes needs to be assessed in new studies.\u0000","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"16 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140981478","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}
James Howard, Neethu Vasudevan, Shantanu Sarkar, Sean Landman, J. Koehler, Daniel Keene
Implantable loop recorders (ILRs) provide continuous single-lead ambulatory electrocardiogram (aECG) monitoring. It is unknown whether these aECGs could be used to identify worsening heart failure. We linked ILR aECG from Medtronic device database to the LVEF measurements in Optum® de-identified electronic health record dataset. We trained an AI algorithm (aECG-CNN) on a dataset of 35,741 aECGs from 2247 patients to identify left ventricular ejection fraction (LVEF) ≤ 40% and assessed its performance using the area under the receiver operating characteristic curve (AUROC). aECG-CNN was then used to identify patients with increasing risk of heart-failure hospitalization in a real-world cohort of 909 patients with prior heart failure diagnosis. This dataset provided 12,467 follow up monthly evaluations, with 201 heart failure hospitalizations. For every month, time series features from these predictions were used to categorize patients into high and low risk groups and predict heart failure hospitalization in the next month. The risk of heart-failure hospitalization in the next 30 days was significantly higher in the cohort that aECG-CNN identified as high risk (hazard ratio 1·89; 95% confidence interval 1·28-2·79; p = 0·001) compared to low risk, even after adjusting patient demographics. (Hazard ratio 1·88, 1.27 to 2·79 p = 0·002). An AI algorithm trained to detect LVEF ≤40% using ILR aECGs can also readily identify patients at increased risk for HF hospitalizations by monitoring changes in the probability of heart failure over 30 days.
{"title":"Dynamic Risk Stratification of worsening heart failure using a Deep learning enabled Implanted Ambulatory Single lead ECG","authors":"James Howard, Neethu Vasudevan, Shantanu Sarkar, Sean Landman, J. Koehler, Daniel Keene","doi":"10.1093/ehjdh/ztae035","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae035","url":null,"abstract":"\u0000 \u0000 \u0000 Implantable loop recorders (ILRs) provide continuous single-lead ambulatory electrocardiogram (aECG) monitoring. It is unknown whether these aECGs could be used to identify worsening heart failure.\u0000 \u0000 \u0000 \u0000 We linked ILR aECG from Medtronic device database to the LVEF measurements in Optum® de-identified electronic health record dataset. We trained an AI algorithm (aECG-CNN) on a dataset of 35,741 aECGs from 2247 patients to identify left ventricular ejection fraction (LVEF) ≤ 40% and assessed its performance using the area under the receiver operating characteristic curve (AUROC). aECG-CNN was then used to identify patients with increasing risk of heart-failure hospitalization in a real-world cohort of 909 patients with prior heart failure diagnosis. This dataset provided 12,467 follow up monthly evaluations, with 201 heart failure hospitalizations. For every month, time series features from these predictions were used to categorize patients into high and low risk groups and predict heart failure hospitalization in the next month. The risk of heart-failure hospitalization in the next 30 days was significantly higher in the cohort that aECG-CNN identified as high risk (hazard ratio 1·89; 95% confidence interval 1·28-2·79; p = 0·001) compared to low risk, even after adjusting patient demographics. (Hazard ratio 1·88, 1.27 to 2·79 p = 0·002).\u0000 \u0000 \u0000 \u0000 An AI algorithm trained to detect LVEF ≤40% using ILR aECGs can also readily identify patients at increased risk for HF hospitalizations by monitoring changes in the probability of heart failure over 30 days.\u0000","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":" 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140997880","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}
J. Hughes, S. Somani, P. Elias, J. Tooley, A. J. Rogers, T. Poterucha, C. Haggerty, Michael Salerno, D. Ouyang, E. Ashley, James Zou, M. Perez
Deep learning methods have recently gained success in detecting left ventricular systolic dysfunction (LVSD) from electrocardiogram waveforms. Despite their impressive accuracy, they are difficult to interpret and deploy broadly in the clinical setting. We set out to determine whether simpler models based on standard electrocardiogram measurements could detect LVSD with similar accuracy to deep learning models. Using an observational dataset of 40,994 matched 12-lead electrocardiograms (ECGs) and transthoracic echocardiograms, we trained a range of models with increasing complexity to detect LVSD based on ECG waveforms and derived measurements. The training data was acquired from Stanford University Medical Center. External validation data was acquired from Columbia Medical Center and the UK Biobank. The Stanford dataset consisted of 40,994 matched ECGs and echocardiograms of which 9.72% had LVSD. A random forest model using 555 discrete, automated measurements achieves an area under the receiver operator characteristic curve (AUC) of 0.92 (0.91-0.93), similar to a deep learning waveform model with an AUC of 0.94 (0.93-0.94). A logistic regression model based on 5 measurements achieves high performance (AUC of 0.86 (0.85-0.87)), close to a deep learning model and better than NT-proBNP. Finally, we find that simpler models are more portable across sites, with experiments at two independent, external sites. Our study demonstrates the value of simple electrocardiographic models which perform nearly as well as deep learning models while being much easier to implement and interpret.
{"title":"Simple Models Versus Deep Learning in Detecting Low Ejection Fraction From The Electrocardiogram","authors":"J. Hughes, S. Somani, P. Elias, J. Tooley, A. J. Rogers, T. Poterucha, C. Haggerty, Michael Salerno, D. Ouyang, E. Ashley, James Zou, M. Perez","doi":"10.1093/ehjdh/ztae034","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae034","url":null,"abstract":"\u0000 \u0000 \u0000 Deep learning methods have recently gained success in detecting left ventricular systolic dysfunction (LVSD) from electrocardiogram waveforms. Despite their impressive accuracy, they are difficult to interpret and deploy broadly in the clinical setting. We set out to determine whether simpler models based on standard electrocardiogram measurements could detect LVSD with similar accuracy to deep learning models.\u0000 \u0000 \u0000 \u0000 Using an observational dataset of 40,994 matched 12-lead electrocardiograms (ECGs) and transthoracic echocardiograms, we trained a range of models with increasing complexity to detect LVSD based on ECG waveforms and derived measurements. The training data was acquired from Stanford University Medical Center. External validation data was acquired from Columbia Medical Center and the UK Biobank. The Stanford dataset consisted of 40,994 matched ECGs and echocardiograms of which 9.72% had LVSD. A random forest model using 555 discrete, automated measurements achieves an area under the receiver operator characteristic curve (AUC) of 0.92 (0.91-0.93), similar to a deep learning waveform model with an AUC of 0.94 (0.93-0.94). A logistic regression model based on 5 measurements achieves high performance (AUC of 0.86 (0.85-0.87)), close to a deep learning model and better than NT-proBNP. Finally, we find that simpler models are more portable across sites, with experiments at two independent, external sites.\u0000 \u0000 \u0000 \u0000 Our study demonstrates the value of simple electrocardiographic models which perform nearly as well as deep learning models while being much easier to implement and interpret.\u0000","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"3 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140654063","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}
Nurgül Keser, J. Lumens, Lukasz Koltowski, Gerd Hindricks, N. Bruining
{"title":"Radical Health Festival Helsinki 2024 Preview: Navigating the Future of Healthcare","authors":"Nurgül Keser, J. Lumens, Lukasz Koltowski, Gerd Hindricks, N. Bruining","doi":"10.1093/ehjdh/ztae030","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae030","url":null,"abstract":"","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"55 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140666318","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}
{"title":"Smartphone use and cerebro-cardio-vascular health: opportunity or public health threat?","authors":"Yvan Devaux, G. Fagherazzi, Christian Montag","doi":"10.1093/ehjdh/ztae032","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae032","url":null,"abstract":"","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"81 12","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140670422","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}
Dor Hadida Barzilai, M. Cohen-Shelly, Vera Sorin, Eyal Zimlichman, E. Massalha, Thomas G Allison, Eyal Klang
Coronary artery disease (CAD) is a leading health challenge worldwide. Exercise stress testing is a foundational non-invasive diagnostic tool. Nonetheless, its variable accuracy prompts the exploration of more reliable methods. Recent advances in machine learning (ML), including deep learning (DL) and natural language processing (NLP), have shown potential in refining the interpretation of stress testing data. Adhering to PRISMA guidelines, we conducted a systematic review of ML applications in stress electrocardiogram (ECG) and stress echocardiography for CAD prognosis. MEDLINE, Web of Science, and the Cochrane Library were used as databases. We analysed the ML models, outcomes, and performance metrics. Overall, seven relevant studies were identified. ML application in stress ECGs resulted in sensitivity and specificity improvements. Some models achieved above 96% in both metrics and reducing false positives by up to 21%. In stress echocardiography, ML models demonstrated an increase in diagnostic precision. Some models achieved specificity and sensitivity rates of up to 92.7% and 84.4%, respectively. NLP applications enabled categorization of stress echocardiography reports, with accuracies nearing 98%. Limitations include a small, retrospective study pool and the exclusion of nuclear stress testing, due to its well-documented status. This review indicates AI applications potential in refining CAD stress testing assessment. Further development for real-world use is warranted.
冠状动脉疾病（CAD）是全球面临的主要健康挑战。运动压力测试是一种基本的无创诊断工具。然而，其准确性参差不齐，这促使人们探索更可靠的方法。机器学习（ML）领域的最新进展，包括深度学习（DL）和自然语言处理（NLP），已显示出完善压力测试数据解读的潜力。 根据 PRISMA 指南，我们对压力心电图（ECG）和压力超声心动图中用于 CAD 预后的 ML 应用进行了系统性回顾。我们使用了 MEDLINE、Web of Science 和 Cochrane Library 作为数据库。我们分析了 ML 模型、结果和性能指标。 总共确定了七项相关研究。在压力心电图中应用 ML 提高了灵敏度和特异性。一些模型在这两项指标上都达到了 96% 以上，并将假阳性降低了 21%。在负荷超声心动图中，ML 模型提高了诊断精确度。一些模型的特异性和灵敏度分别高达 92.7% 和 84.4%。NLP 应用实现了压力超声心动图报告的分类，准确率接近 98%。不足之处包括：回顾性研究的规模较小，而且由于核应力测试的地位已得到充分证明，因此未将其包括在内。 本综述显示了人工智能在完善 CAD 压力测试评估方面的应用潜力。有必要进一步开发用于真实世界。
{"title":"Machine Learning in Cardiac Stress Test Interpretation: A Systematic Review","authors":"Dor Hadida Barzilai, M. Cohen-Shelly, Vera Sorin, Eyal Zimlichman, E. Massalha, Thomas G Allison, Eyal Klang","doi":"10.1093/ehjdh/ztae027","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae027","url":null,"abstract":"\u0000 \u0000 \u0000 Coronary artery disease (CAD) is a leading health challenge worldwide. Exercise stress testing is a foundational non-invasive diagnostic tool. Nonetheless, its variable accuracy prompts the exploration of more reliable methods. Recent advances in machine learning (ML), including deep learning (DL) and natural language processing (NLP), have shown potential in refining the interpretation of stress testing data.\u0000 \u0000 \u0000 \u0000 Adhering to PRISMA guidelines, we conducted a systematic review of ML applications in stress electrocardiogram (ECG) and stress echocardiography for CAD prognosis. MEDLINE, Web of Science, and the Cochrane Library were used as databases. We analysed the ML models, outcomes, and performance metrics.\u0000 \u0000 \u0000 \u0000 Overall, seven relevant studies were identified.\u0000 ML application in stress ECGs resulted in sensitivity and specificity improvements. Some models achieved above 96% in both metrics and reducing false positives by up to 21%. In stress echocardiography, ML models demonstrated an increase in diagnostic precision. Some models achieved specificity and sensitivity rates of up to 92.7% and 84.4%, respectively. NLP applications enabled categorization of stress echocardiography reports, with accuracies nearing 98%. Limitations include a small, retrospective study pool and the exclusion of nuclear stress testing, due to its well-documented status.\u0000 \u0000 \u0000 \u0000 This review indicates AI applications potential in refining CAD stress testing assessment. Further development for real-world use is warranted.\u0000","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":" 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140690799","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}
P. Calvert, Mark T. Mills, Kelly Howarth, Sini Aykara, Lindsay Lunt, Helen Brewer, David Green, Janet Green, Simon Moore, Jude Almutawa, Dominik Linz, G. Lip, Derick Todd, Dhiraj Gupta
Direct current cardioversion (DCCV) is a commonly utilised rhythm control technique for atrial fibrillation (AF). Follow-up typically comprises a hospital visit for 12-lead ECG two weeks post-DCCV. We report the feasibility, costs and environmental benefit of remote photoplethysmography (PPG) monitoring as an alternative. We retrospectively analysed DCCV cases at our centre from May 2020 to October 2022. Patients were stratified into those with remote PPG follow-up and those with traditional 12-lead ECG follow-up. Monitoring type was decided by the specialist nurse performing the DCCV at the time of the procedure after discussing with the patient and offering them both options if appropriate. Outcomes included the proportion of patients who underwent PPG monitoring, patient compliance and experience, and cost, travel and environmental impact. 416 patients underwent 461 acutely successful DCCV procedures. 246 underwent PPG follow-up whilst 214 underwent ECG follow-up. Patient compliance was high (PPG 89.4% vs ECG 89.8%; p > 0.999) and the majority of PPG users (90%) found the app easy to use. Sinus rhythm was maintained in 71.1% (PPG) and 64.7% (ECG) of patients (p = 0.161). 29 (11.8%) PPG patients subsequently required an ECG either due to non-compliance, technical failure or inconclusive PPG readings. Despite this, mean healthcare costs (£47.91 vs £135 per patient; p < 0.001) and median cost to the patient (£0 vs £5.97; p < 0.001) were lower with PPG. Median travel time per patient (0 vs 44min; p < 0.001) and CO2 emissions (0 vs 3.59kg; p < 0.001) were also lower with PPG. No safety issues were identified. Remote PPG monitoring is a viable method of assessing for arrhythmia recurrence post-DCCV. This approach may save patients significant travel time, reduce environmental CO2 emission and be cost saving in a publicly-funded healthcare system.
{"title":"Remote Rhythm Monitoring using a Photoplethysmography Smartphone Application after Cardioversion for Atrial Fibrillation","authors":"P. Calvert, Mark T. Mills, Kelly Howarth, Sini Aykara, Lindsay Lunt, Helen Brewer, David Green, Janet Green, Simon Moore, Jude Almutawa, Dominik Linz, G. Lip, Derick Todd, Dhiraj Gupta","doi":"10.1093/ehjdh/ztae028","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae028","url":null,"abstract":"\u0000 \u0000 \u0000 Direct current cardioversion (DCCV) is a commonly utilised rhythm control technique for atrial fibrillation (AF). Follow-up typically comprises a hospital visit for 12-lead ECG two weeks post-DCCV. We report the feasibility, costs and environmental benefit of remote photoplethysmography (PPG) monitoring as an alternative.\u0000 \u0000 \u0000 \u0000 We retrospectively analysed DCCV cases at our centre from May 2020 to October 2022. Patients were stratified into those with remote PPG follow-up and those with traditional 12-lead ECG follow-up. Monitoring type was decided by the specialist nurse performing the DCCV at the time of the procedure after discussing with the patient and offering them both options if appropriate. Outcomes included the proportion of patients who underwent PPG monitoring, patient compliance and experience, and cost, travel and environmental impact.\u0000 \u0000 \u0000 \u0000 416 patients underwent 461 acutely successful DCCV procedures. 246 underwent PPG follow-up whilst 214 underwent ECG follow-up. Patient compliance was high (PPG 89.4% vs ECG 89.8%; p > 0.999) and the majority of PPG users (90%) found the app easy to use. Sinus rhythm was maintained in 71.1% (PPG) and 64.7% (ECG) of patients (p = 0.161). 29 (11.8%) PPG patients subsequently required an ECG either due to non-compliance, technical failure or inconclusive PPG readings. Despite this, mean healthcare costs (£47.91 vs £135 per patient; p < 0.001) and median cost to the patient (£0 vs £5.97; p < 0.001) were lower with PPG. Median travel time per patient (0 vs 44min; p < 0.001) and CO2 emissions (0 vs 3.59kg; p < 0.001) were also lower with PPG. No safety issues were identified.\u0000 \u0000 \u0000 \u0000 Remote PPG monitoring is a viable method of assessing for arrhythmia recurrence post-DCCV. This approach may save patients significant travel time, reduce environmental CO2 emission and be cost saving in a publicly-funded healthcare system.\u0000","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"340 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140703223","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：