1
|
Goh CH, Ferdowsi M, Gan MH, Kwan BH, Lim WY, Tee YK, Rosli R, Tan MP. Assessing the efficacy of machine learning algorithms for syncope classification: A systematic review. MethodsX 2024; 12:102508. [PMID: 38162148 PMCID: PMC10755776 DOI: 10.1016/j.mex.2023.102508] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/28/2023] [Accepted: 11/30/2023] [Indexed: 01/03/2024] Open
Abstract
Syncope is a transient loss of consciousness with rapid onset. The aims of the study were to systematically evaluate available machine learning (ML) algorithm for supporting syncope diagnosis to determine their performance compared to existing point scoring protocols. We systematically searched IEEE Xplore, Web of Science, and Elsevier for English articles (Jan 2011 - Sep 2021) on individuals aged five and above, employing ML algorithms in syncope detection with Head-up titl table test (HUTT)-monitored hemodynamic parameters and reported metrics. Extracted data encompassed subject count, age range, syncope protocols, ML type, hemodynamic parameters, and performance metrics. Of the 6301 studies initially identified, 10 studies, involving 1205 participants aged 5 to 82 years, met the inclusion criteria, and formed the basis for it. Selected studies must use ML algorithms in syncope detection with hemodynamic parameters recorded throughout HUTT. The overall ML algorithm performance achieved a sensitivity of 88.8% (95% CI: 79.4-96.1%), specificity of 81.5% (95% CI: 69.8-92.8%) and accuracy of 85.8% (95% CI: 78.6-92.8%). Machine learning improves syncope diagnosis compared to traditional scoring, requiring fewer parameters. Future enhancements with larger databases are anticipated. Integrating ML can curb needless admissions, refine diagnostics, and enhance the quality of life for syncope patients.
Collapse
Affiliation(s)
- Choon-Hian Goh
- Department of Mechatronics and BioMedical Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, 43000 Kajang, Selangor, Malaysia
- Centre for Healthcare Science and Technology, Universiti Tunku Abdul Rahman, 43000 Kajang, Selangor, Malaysia
| | - Mahbuba Ferdowsi
- Department of Mechatronics and BioMedical Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, 43000 Kajang, Selangor, Malaysia
- Centre for Healthcare Science and Technology, Universiti Tunku Abdul Rahman, 43000 Kajang, Selangor, Malaysia
| | - Ming Hong Gan
- Department of Mechatronics and BioMedical Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, 43000 Kajang, Selangor, Malaysia
| | - Ban-Hoe Kwan
- Department of Mechatronics and BioMedical Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, 43000 Kajang, Selangor, Malaysia
- Centre for Healthcare Science and Technology, Universiti Tunku Abdul Rahman, 43000 Kajang, Selangor, Malaysia
| | - Wei Yin Lim
- Electrical and Computer Systems Engineering, School of Engineering and Advanced Engineering Platform, Monash University Malaysia, Bandar Sunway 47500, Selangor, Malaysia
| | - Yee Kai Tee
- Department of Mechatronics and BioMedical Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, 43000 Kajang, Selangor, Malaysia
- Centre for Healthcare Science and Technology, Universiti Tunku Abdul Rahman, 43000 Kajang, Selangor, Malaysia
| | - Roshaslina Rosli
- ACT4Health Services and Consultancy, 47300 Petaling Jaya, Malaysia
| | - Maw Pin Tan
- Ageing and Age-Associated Disorders Research Group, Department of Medicine, Faculty of Medicine, Universiti Malaya, 50603 Kuala Lumpur, Malaysia
- Department Medical Sciences, Faculty of Healthcare and Medical Sciences, Sunway University, 47500 Bandar Sunway, Malaysia
| |
Collapse
|
2
|
Ferdowsi M, Kwan BH, Tan MP, Saedon NI, Subramaniam S, Abu Hashim NFI, Mohd Nasir SS, Zainal Abidin I, Chee KH, Goh CH. Classification of vasovagal syncope from physiological signals on tilt table testing. Biomed Eng Online 2024; 23:37. [PMID: 38555421 PMCID: PMC10981362 DOI: 10.1186/s12938-024-01229-9] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/24/2023] [Accepted: 03/06/2024] [Indexed: 04/02/2024] Open
Abstract
BACKGROUND The diagnostic test for vasovagal syncope (VVS), the most common cause of syncope is head-up tilt test (HUTT) assessment. During the test, subjects experienced clinical symptoms such as nausea, sweating, pallor, the feeling of palpitations, being on the verge of passing out, and fainting. The study's goal is to develop an algorithm to classify VVS patients based on physiological signals blood pressure (BP) and electrocardiography (ECG) obtained from the HUTT. METHODS After 10 min of supine rest, the subject was tilted at a 70-degree angle on a tilt table for approximately a total of 35 min. 400 µg of glyceryl trinitrate (GTN) was administered sublingually after the first 20 min and monitoring continued for another 15 min. Mean imputation and K-nearest neighbors (KNN) imputation approaches to handle missing values. Next, feature selection techniques were implemented, including genetic algorithm, recursive feature elimination, and feature importance, to determine the crucial features. The Mann-Whitney U test was then performed to determine the statistical difference between two groups. Patients with VVS are categorized via machine learning models including Support Vector Machine (SVM), Gaussian Naïve Bayes (GNB), Multinomial Naïve Bayes (MNB), KNN, Logistic Regression (LR), and Random Forest (RF). The developed model is interpreted using an explainable artificial intelligence (XAI) model known as partial dependence plot. RESULTS A total of 137 subjects aged between 9 and 93 years were recruited for this study, 54 experienced clinical symptoms were considered positive tests, while the remaining 83 tested negative. Optimal results were obtained by combining the KNN imputation technique and three tilting features with SVM with 90.5% accuracy, 87.0% sensitivity, 92.7% specificity, 88.6% precision, 87.8% F1 score, and 95.4% ROC (receiver operating characteristics) AUC (area under curve). CONCLUSIONS The proposed algorithm effectively classifies VVS patients with over 90% accuracy. However, the study was confined to a small sample size. More clinical datasets are required to ensure that our approach is generalizable.
Collapse
Affiliation(s)
- Mahbuba Ferdowsi
- Department of Mechatronics and Biomedical Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, 43000, Kajang, Malaysia
- Centre for Healthcare Science and Technology, Universiti Tunku Abdul Rahman, 43000, Kajang, Malaysia
| | - Ban-Hoe Kwan
- Department of Mechatronics and Biomedical Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, 43000, Kajang, Malaysia
- Centre for Healthcare Science and Technology, Universiti Tunku Abdul Rahman, 43000, Kajang, Malaysia
| | - Maw Pin Tan
- Ageing and Age-Associated Disorders Research Group, Department of Medicine, Faculty of Medicine, Universiti Malaya, 50603, Kuala Lumpur, Malaysia
| | - Nor' Izzati Saedon
- Ageing and Age-Associated Disorders Research Group, Department of Medicine, Faculty of Medicine, Universiti Malaya, 50603, Kuala Lumpur, Malaysia
| | - Sukanya Subramaniam
- Cardiorespiratory Laboratories, Universiti Malaya Medical Center, 50603, Petaling Jaya, Malaysia
| | | | - Siti Sakinah Mohd Nasir
- Cardiorespiratory Laboratories, Universiti Malaya Medical Center, 50603, Petaling Jaya, Malaysia
| | - Imran Zainal Abidin
- Department of Cardiology, Department of Medicine, Faculty of Medicine, Universiti Malaya, 50603, Kuala Lumpur, Malaysia
| | - Kok Han Chee
- Department of Cardiology, Department of Medicine, Faculty of Medicine, Universiti Malaya, 50603, Kuala Lumpur, Malaysia
| | - Choon-Hian Goh
- Department of Mechatronics and Biomedical Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, 43000, Kajang, Malaysia.
- Centre for Healthcare Science and Technology, Universiti Tunku Abdul Rahman, 43000, Kajang, Malaysia.
| |
Collapse
|
3
|
Svennberg E, Tjong F, Goette A, Akoum N, Di Biase L, Bordachar P, Boriani G, Burri H, Conte G, Deharo JC, Deneke T, Drossart I, Duncker D, Han JK, Heidbuchel H, Jais P, de Oliviera Figueiredo MJ, Linz D, Lip GYH, Malaczynska-Rajpold K, Márquez M, Ploem C, Soejima K, Stiles MK, Wierda E, Vernooy K, Leclercq C, Meyer C, Pisani C, Pak HN, Gupta D, Pürerfellner H, Crijns HJGM, Chavez EA, Willems S, Waldmann V, Dekker L, Wan E, Kavoor P, Turagam MK, Sinner M. How to use digital devices to detect and manage arrhythmias: an EHRA practical guide. Europace 2022; 24:979-1005. [PMID: 35368065 DOI: 10.1093/europace/euac038] [Citation(s) in RCA: 101] [Impact Index Per Article: 50.5] [Reference Citation Analysis] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/16/2022] Open
Affiliation(s)
- Emma Svennberg
- Department of Medicine Huddinge, Karolinska Institutet, Stockholm, Sweden
| | - Fleur Tjong
- Heart Center, Amsterdam University Medical Centers, University of Amsterdam, Amsterdam, The Netherlands
| | - Andreas Goette
- St. Vincenz Hospital Paderborn, Paderborn, Germany
- MAESTRIA Consortium/AFNET, Münster, Germany
| | - Nazem Akoum
- Heart Institute, University of Washington School of Medicine, Seattle, WA, USA
| | - Luigi Di Biase
- Albert Einstein College of Medicine at Montefiore Hospital, New York, NY, USA
| | | | - Giuseppe Boriani
- Cardiology Division, Department of Biomedical, Metabolic and Neural Sciences, University of Modena and Reggio Emilia, Policlinico di Modena, Modena, Italy
| | - Haran Burri
- Cardiology Department, University Hospital of Geneva, Geneva, Switzerland
| | - Giulio Conte
- Cardiocentro Ticino Institute, Ente Ospedaliero Cantonale, Lugano, Switzerland
| | - Jean Claude Deharo
- Assistance Publique-Hôpitaux de Marseille, Centre Hospitalier Universitaire La Timone, Service de Cardiologie, Marseille, France
- Aix Marseille Université, C2VN, Marseille, France
| | - Thomas Deneke
- Heart Center Bad Neustadt, Bad Neustadt an der Saale, Germany
| | - Inga Drossart
- European Society of Cardiology, Sophia Antipolis, France
- ESC Patient Forum, Sophia Antipolis, France
| | - David Duncker
- Hannover Heart Rhythm Center, Department of Cardiology and Angiology, Hannover Medical School, Hannover, Germany
| | - Janet K Han
- Cardiac Arrhythmia Centers, Veterans Affairs Greater Los Angeles Healthcare System and University of California, Los Angeles, CA, USA
| | - Hein Heidbuchel
- Department of Cardiology, Antwerp University Hospital, Antwerp, Belgium
- Cardiovascular Research Group, Antwerp University, Antwerp, Belgium
| | - Pierre Jais
- Bordeaux University Hospital, Bordeaux, France
| | | | - Dominik Linz
- Department of Cardiology, Maastricht University Medical Centre and Cardiovascular Research Institute Maastricht, Maastricht, the Netherlands
| | - Gregory Y H Lip
- Liverpool Centre for Cardiovascular Science, University of Liverpool and Liverpool Heart & Chest Hospital, Liverpool, UK
- Department of Clinical Medicine, Aalborg University, Aalborg, Denmark
| | | | - Manlio Márquez
- Department of Electrocardiology, Instituto Nacional de Cardiología, Mexico City, Mexico
| | - Corrette Ploem
- Department of Ethics, Law and Medical Humanities, Amsterdam University Medical Centers, University of Amsterdam, Amsterdam, The Netherlands
| | - Kyoko Soejima
- Kyorin University School of Medicine, Mitaka, Tokyo, Japan
| | - Martin K Stiles
- Waikato Clinical School, University of Auckland, Hamilton, New Zealand
| | - Eric Wierda
- Department of Cardiology, Dijklander Hospital, Hoorn, the Netherlands
| | - Kevin Vernooy
- Department of Cardiology, Cardiovascular Research Institute Maastricht (CARIM), Maastricht University Medical Center, Maastricht, the Netherlands
| | | | - Christian Meyer
- Division of Cardiology/Angiology/Intensive Care, EVK Düsseldorf, Teaching Hospital University of Düsseldorf, Düsseldorf, Germany
| | - Cristiano Pisani
- Arrhythmia Unit, Heart Institute, InCor, University of São Paulo Medical School, São Paulo, Brazil
| | - Hui Nam Pak
- Yonsei University, Severance Cardiovascular Hospital, Yonsei University Health System, Seoul, Republic of Korea
| | - Dhiraj Gupta
- Faculty of Health and Life Sciences, Liverpool Heart and Chest Hospital, University of Liverpool, Liverpool, UK
| | | | - H J G M Crijns
- Em. Professor of Cardiology, University of Maastricht, Maastricht, Netherlands
| | - Edgar Antezana Chavez
- Division of Cardiology, Hospital General de Agudos Dr. Cosme Argerich, Pi y Margall 750, C1155AHB Buenos Aires, Argentina
- Division of Cardiology, Hospital Belga, Antezana 455, C0000 Cochabamba, Bolivia
| | | | - Victor Waldmann
- Electrophysiology Unit, European Georges Pompidou Hospital, Paris, France
- Adult Congenital Heart Disease Unit, European Georges Pompidou Hospital, Paris, France
| | - Lukas Dekker
- Catharina Ziekenhuis Eindhoven, Eindhoven, Netherlands
| | - Elaine Wan
- Cardiology and Cardiac Electrophysiology, Columbia University, New York, NY, USA
| | - Pramesh Kavoor
- Cardiology Department, Westmead Hospital, Westmead, New South Wales, Australia
| | | | - Moritz Sinner
- Univ. Hospital Munich, Campus Grosshadern, Munich, Germany
| |
Collapse
|
4
|
He Z, Du L, Du S, Wu B, Fan Z, Xin B, Chen X, Fang Z, Liu J. Machine learning for the early prediction of head-up tilt testing outcome. Biomed Signal Process Control 2021. [DOI: 10.1016/j.bspc.2021.102904] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/21/2022]
|
5
|
Myrovali E, Fragakis N, Vassilikos V, Hadjileontiadis LJ. Efficient syncope prediction from resting state clinical data using wavelet bispectrum and multilayer perceptron neural network. Med Biol Eng Comput 2021; 59:1311-1324. [PMID: 33959855 DOI: 10.1007/s11517-021-02353-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/24/2020] [Accepted: 03/19/2021] [Indexed: 11/28/2022]
Abstract
Neurally mediated syncope (NMS) is the most common type of syncope, and head up tilt test (HUTT) is, so far, the most appropriate tool to identify NMS. In this work, an effort to predict the NMS before performing the HUTT is attempted. To achieve this, the heart rate variability (HRV) at rest and during the first minutes of tilting position during HUTT was analyzed using both time and frequency domains. Various features from HRV regularity and complexity, along with wavelet higher-order spectrum (WHOS) analysis in low-frequency (LF) and high-frequency (HF) bands were examined. The experimental results from 26 patients with history of NMS have shown that at rest, a time domain entropy measure and WHOS-based features in LF band exhibit significant differences between positive and negative HUTT as well as among 10 healthy subjects and NMS patients. The best performance of multilayer perceptron neural network (MPNN) was achieved by using an input vector consisted of WHOS-based HRV features in the LF zone and systolic blood pressure from the resting period, yielding an accuracy of 89.7%, assessed by 5-fold cross-validation. The promising results presented here pave the way for an early prediction of the HUTT outcome from resting state, contributing to the identification of patients at higher risk NMS. The HRV analysis along with systolic blood pressure at rest predict NMS using a multilayer perceptron neural network.
Collapse
Affiliation(s)
- Evangelia Myrovali
- Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki, GR 54645, Thessaloniki, Greece.
| | - Nikolaos Fragakis
- 3rd Cardiology Department, Aristotle University of Thessaloniki, Hippokration General Hospital, 49 Konstantinoupoleos str, 54642, Thessaloniki, Greece
| | - Vassilios Vassilikos
- 3rd Cardiology Department, Aristotle University of Thessaloniki, Hippokration General Hospital, 49 Konstantinoupoleos str, 54642, Thessaloniki, Greece
| | - Leontios J Hadjileontiadis
- Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki, GR 54645, Thessaloniki, Greece.,Department of Electrical and Computer Engineering, Khalifa University of Science and Technology, PO BOX 127788, Abu Dhabi, UAE
| |
Collapse
|
6
|
Chang IS, Armanfard N, Javaid AQ, Boger J, Mihailidis A. Unobtrusive Detection of Simulated Orthostatic Hypotension and Supine Hypertension Using Ballistocardiogram and Electrocardiogram of Healthy Adults. IEEE JOURNAL OF TRANSLATIONAL ENGINEERING IN HEALTH AND MEDICINE-JTEHM 2018; 6:2700613. [PMID: 30345183 PMCID: PMC6193524 DOI: 10.1109/jtehm.2018.2864738] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 04/02/2018] [Revised: 06/15/2018] [Accepted: 07/18/2018] [Indexed: 12/11/2022]
Abstract
Effective management of neurogenic orthostatic hypotension and supine hypertension (SH-OH) due autonomic failure requires a frequent and timely adjustment of medication throughout the day to maintain the blood pressure (BP) within the normal range, i.e., an accurate depiction of BP is a key prerequisite of effective management. One of the emerging technologies that provide one’s circadian and long-term physiological status with increased usability is unobtrusive zero-effort monitoring. In this paper, a zero-effort device, a floor tile, was used to develop an unobtrusive BP monitoring technique. Namely, RJ-interval, the time between the J-peak of a ballistocardiogram and the R-peak of an electrocardiogram, was used to develop a classifier that can detect changes in systolic BP (SBP) induced by the Valsalva maneuver on healthy adults (i.e., a simulated SH-OH). A t-test was used to show statistical differences between the mean RJ-intervals of decreased SBP, baseline, and increased SBP. Following the t-test, a classifier that detected a change in SBP was developed based on a naïve Bayes classifier (NBC). The t-test showed a clear statistical difference between the mean RJ-intervals of the increased SBP, baseline, and decreased SBP. The NBC-based classifier was able to detect increased SBP with 89.3% true positive rate (TPR), 100% true negative rate (TNR), and 94% accuracy and detect decreased SBP with 92.3% TPR, 100% TNR, and 95% accuracy. The analysis showed strong potential in using the developed classifier to assist monitoring of people with SH-OH; the algorithm may be used clinically to detect a long-term trend of symptoms of SH-OH.
Collapse
Affiliation(s)
- Isaac S Chang
- Institute of Biomaterials and Biomedical Engineering, University of TorontoTorontoONM5S 3G9Canada
| | - Narges Armanfard
- Toronto-RehabUniversity Health Network, University of TorontoTorontoONM5G 2A2Canada
| | - Abdul Q Javaid
- Department of Occupational Science and Occupational TherapyUniversity of TorontoTorontoONM5S 3G9Canada
| | - Jennifer Boger
- Systems Design EngineeringUniversity of WaterlooWaterlooONN2L 3G1Canada.,Research Institute for AgingWaterlooONN2J 0E2Canada
| | - Alex Mihailidis
- Department of Occupational Science and Occupational TherapyUniversity of TorontoTorontoONM5S 3G9Canada
| |
Collapse
|
7
|
Silva B, Muehlsteff J, Couceiro R, Henriques J, Carvalho P. Artifact detection in accelerometer signals acquired from the carotid. ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. ANNUAL INTERNATIONAL CONFERENCE 2017; 2017:135-138. [PMID: 29059828 DOI: 10.1109/embc.2017.8036780] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/07/2022]
Abstract
Manual pulse palpation is the common procedure to assess pulse in unconscious patients. This is an error prone procedure during cardiopulmonary resuscitation and therefore automatic pulse detection techniques are being investigated. Accelerometry is an interesting sensing modality for this type of applications. However, accelerometers are highly prone to movement artifacts. Hence, one challenge in designing a solution using accelerometers is to handle motion artifacts properly. In this paper we investigate computationally simple features and classifier to capture movement artifacts in accelerometer signals acquired from the carotid. In particular, based on data obtained from health subjects we show that it is possible to use simple features to achieve an artifact detection sensitivity and specificity higher than 90%.
Collapse
|