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: 2] [Impact Index Per Article: 2.0] [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
|
Quantitative Complexity Theory Used in the Prediction of Head-Up Tilt Testing Outcome. Cardiol Res Pract 2021; 2021:8882498. [PMID: 34603772 PMCID: PMC8486546 DOI: 10.1155/2021/8882498] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 08/25/2020] [Revised: 05/29/2021] [Accepted: 09/08/2021] [Indexed: 11/17/2022] Open
Abstract
Background Head-up tilt testing (HUTT), a well-established tool in the diagnosis of vasovagal syncope, is time-consuming, and every provoked vasovagal reaction may result in consolidating the reflex mechanism. Therefore, identification of parameters that could shorten the duration of HUTT and prevent fainting is desirable. Quantitative complexity theory (QCT) may provide holistic information on the cardiovascular reaction in HUTT. The aim of the present article was to evaluate the prognostic value of complexity in comparison with traditional haemodynamic parameters (HR and BP) in predicting the HUTT outcome. Methods Eighty-one healthy volunteers (74 men; mean age: 37.8 years) were included in this retrospective analysis of data collected within the project realized in Department of Cardiology and Internal Diseases, Military Institute of Medicine between January 2012 and October 2014. The subjects underwent HUTT, with beat-to-beat haemodynamic monitoring with a Niccomo™. The chosen haemodynamic parameters (including BP, HR, stroke volume, cardiac output, systemic vascular resistance) have been used in complexity analysis. Results HUTT was positive in 54 (66.7%) study participants. The values of complexity were already higher in fainting subjects than those were in nonfainting ones 300 s before HUTT termination (HUTT_end), with a significant upward trend starting 150 s before (pre)syncope. An area under the curve (AUC) over 0.700 was observed for complexity from 120 s before HUTT_end, with a sensitivity of 63% and specificity of 78% at this time point. The prognostic value of complexity was superior to that of the HR and mean arterial pressure (MAP). Conclusions Complexity has been shown to be a sensitive marker of cardiovascular haemodynamic response to orthostatic stress and proved to be superior over HR and BP in predicting HUTT outcomes.
Collapse
|
4
|
Buszko K, Piątkowska A, Koźluk E, Fabiszak T, Opolski G. Transfer Information Assessment in Diagnosis of Vasovagal Syncope Using Transfer Entropy. ENTROPY (BASEL, SWITZERLAND) 2019; 21:e21040347. [PMID: 33267061 PMCID: PMC7514832 DOI: 10.3390/e21040347] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 02/08/2019] [Revised: 03/25/2019] [Accepted: 03/26/2019] [Indexed: 06/12/2023]
Abstract
The paper presents an application of Transfer Entropy (TE) to the analysis of information transfer between biosignals (heart rate expressed as R-R intervals (RRI), blood pressure (sBP, dBP) and stroke volume (SV)) measured during head up tilt testing (HUTT) in patients with suspected vasovagal syndrome. The study group comprised of 80 patients who were divided into two groups: the HUTT(+) group consisting of 57 patients who developed syncope during the passive phase of the test and HUTT(-) group consisting of 23 patients who had a negative result of the passive phase and experienced syncope after provocation with nitroglycerin. In both groups the information transfer depends on the phase of the tilt test. In supine position the highest transfer occurred between driver RRI and other components. In upright position it is the driver sBP that plays the crucial role. The pre-syncope phase features the highest information transfer from driver SV to blood pressure components. In each group the comparisons of TE between different phases of HUT test showed significant differences for RRI and SV as drivers.
Collapse
Affiliation(s)
- Katarzyna Buszko
- Department of Theoretical Foundations of Bio-Medical Science and Medical Informatics, Collegium Medicum, Nicolaus Copernicus University, 85-067 Bydgoszcz, Poland
| | - Agnieszka Piątkowska
- Department of Emergency Medicine, Wroclaw Medical University, 50-556 Wroclaw, Poland
- 1st Chair and Department of Cardiology, Medical University of Warsaw, 02-097 Warsaw, Poland
| | - Edward Koźluk
- 1st Chair and Department of Cardiology, Medical University of Warsaw, 02-097 Warsaw, Poland
| | - Tomasz Fabiszak
- Department of Cardiology and Internal Diseases, Collegium Medicum, Nicolaus Copernicus University, 85-067 Bydgoszcz, Poland
| | - Grzegorz Opolski
- 1st Chair and Department of Cardiology, Medical University of Warsaw, 02-097 Warsaw, Poland
| |
Collapse
|
5
|
Buszko K, Kujawski S, Newton JL, Zalewski P. Hemodynamic Response to the Head-Up Tilt Test in Patients With Syncope as a Predictor of the Test Outcome: A Meta-Analysis Approach. Front Physiol 2019; 10:184. [PMID: 30899228 PMCID: PMC6416221 DOI: 10.3389/fphys.2019.00184] [Citation(s) in RCA: 10] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/07/2018] [Accepted: 02/14/2019] [Indexed: 11/13/2022] Open
Abstract
Aim: The paper presents a meta-analysis of studies comparing hemodynamic parameters: heart rate (HR), systolic blood pressure (sBP), diastolic blood pressure (dBP), and stroke volume (SV) measured during head-up tilt table test (HUTT) in patients with positive and negative HUT test outcome. Methods: Pubmed and Clinical Key databases were searched for English-only articles presenting results of biosignals measurements during tilt test in patients suffering from syncope. From 3,289 articles 13 articles published between 1997 and 2015 investigating 892 patients (467 with positive HUTT outcome and 401 with negative one) were selected. Results: There were not statistically significant differences observed between the parameters measured in supine position in patients with positive and negative test outcome [HR (p = 0.86), sBP (p = 0.32), dBP (p = 0.21), SV (p = 0.71)]. In tilt position the parameters HR and SV were significantly different when compared between the two groups of patients [HR (p = 0.02), sBP (p = 0.10), dBP (p = 0.59), SV (p = 0.0004)]. Conclusions: Changes in HR and SV parameters in response to tilt test turned out to be statistically significant. In supine position the differences between patients with positive and negative test outcome were not significant, hence tilt test can be considered as necessary in the diagnosis of vasovagal syndrome.
Collapse
Affiliation(s)
- Katarzyna Buszko
- Department of Theoretical Foundations of Bio-Medical Science and Medical Informatics, Collegium Medicum, Nicolaus Copernicus University, Bydgoszcz, Poland
| | - Sławomir Kujawski
- Division of Ergonomics and Exercise Physiology, Department of Hygiene, Epidemiology and Ergonomics, Collegium Medicum in Bydgoszcz, Nicolaus Copernicus University in Toruń, Toruń, Poland
| | - Julia L Newton
- The Medical School, Institute for Ageing and Health, Newcastle University, Newcastle-upon-Tyne, United Kingdom
| | - Paweł Zalewski
- Division of Ergonomics and Exercise Physiology, Department of Hygiene, Epidemiology and Ergonomics, Collegium Medicum in Bydgoszcz, Nicolaus Copernicus University in Toruń, Toruń, Poland
| |
Collapse
|
6
|
Buszko K, Piątkowska A, Koźluk E, Fabiszak T, Opolski G. The complexity of hemodynamic response to the tilt test with and without nitroglycerine provocation in patients with vasovagal syncope. Sci Rep 2018; 8:14554. [PMID: 30266992 PMCID: PMC6162241 DOI: 10.1038/s41598-018-32718-2] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/24/2018] [Accepted: 09/10/2018] [Indexed: 11/09/2022] Open
Abstract
The paper presents a comparison of vasovagal syndrome occurrence in a head up tilt table test between patients with a positive result of passive tilt test and those with a positive result after pharmacological provocation. The study group consisted of 80 patients: 57 patients who experienced syncope in the passive phase of the test (43 women (aged: 35.6 ± 16.2) and 14 men (aged: 41.7 ± 15.6) and 23 patients who experienced syncope after pharmacological provocation (17 women (age: 32.3 ± 12) and 6 men (age: 43 ± 15). The main investigation was based on the assessment of monitored signals complexity: heart rate, blood pressure and stroke volume. The analysis of complexity in chosen measurement phases was performed with Sample Entropy. The investigation showed that the reactions of autonomic nervous system during tilt test and before syncope are similar for positive result of passive tilt test and positive result of tilt test with provocation. The differences in supine position occurred only in analysis based on impedance measurement (SV: p = 0.01). Significant differences were denoted for all signals just before the syncope (RRI, sBP, dBP: p = 0,00001 and SV: p = 0.01). In analysis of signals complexity the significant differences occurred just before the syncope for Sample Entropy of blood pressure (SampEn (sBP): p = 0.0008, SampEn (dBP): p = 0,0001).
Collapse
Affiliation(s)
- Katarzyna Buszko
- Department of Theoretical Foundations of Bio-Medical Science and Medical Informatics, Collegium Medicum, Nicolaus Copernicus University, 85-067, Bydgoszcz, Poland.
| | - Agnieszka Piątkowska
- Department and Clinic of Emergency Medicine, Wroclaw Medical University, Wroclaw, 50-556, Poland
- 1st Chair and Department of Cardiology, Medical University of Warsaw, Warsaw, 02-091, Poland
| | - Edward Koźluk
- 1st Chair and Department of Cardiology, Medical University of Warsaw, Warsaw, 02-091, Poland
| | - Tomasz Fabiszak
- Department of Cardiology and Internal Diseases, Collegium Medicum, Nicolaus Copernicus University, 85-067, Bydgoszcz, Poland
| | - Grzegorz Opolski
- 1st Chair and Department of Cardiology, Medical University of Warsaw, Warsaw, 02-091, Poland
| |
Collapse
|
7
|
Jureczko L, Dobronska K, Kolacz M, Radziszewski P, Dobronski P. Non-invasive evaluation of cardiac index by impedance cardiography in patients undergoing percutaneous nephrolithotomy. Arch Med Sci 2018; 14:801-806. [PMID: 30002697 PMCID: PMC6040138 DOI: 10.5114/aoms.2016.63598] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 03/21/2016] [Accepted: 06/23/2016] [Indexed: 11/17/2022] Open
Abstract
INTRODUCTION Percutaneous nephrolithotomy (PNL) is an endoscopic treatment of renal lithiasis. It is usually two-staged: it begins in the lithotomy position for ureteral catheter placement and retrograde pyelography, and subsequently an optimal renal access is obtained in the prone position. Some patients under epidural anesthesia do not tolerate the prone position and the PNL cannot be continued. This may be related to changes occurring within the circulatory system. The aim of this prospective randomized double-blind study was to evaluate the changes of the cardiac index (CI) during PNL. MATERIAL AND METHODS In a group of 50 patients, with ASA physical status grade 1-2, epidural anesthesia with either 0.2% ropivacaine or 0.25% bupivacaine was performed and then the CI was evaluated by impedance cardiography. RESULTS Forty-five patients were included in the analysis; all tolerated the PNL well. After turning prone, a decrease in the CI was always recorded, a maximum after 10-15 min - 22.58 ±11.47%. There was significant variability of recorded values. The average CI dropped from 2.96 ±0.42 l/min/m2 to 2.28 ±0.39 l/min/m2. In 7 patients the decrease in the CI was greater than 35%. No correlation was observed with the arterial blood pressure or the heart rate. The decrease in the CI occurred irrespective of the type of local anesthetic used (p = 0.91). CONCLUSIONS A decrease in the CI was observed in every case, and it should be taken into consideration during qualification for PNL in the prone position.
Collapse
Affiliation(s)
- Lidia Jureczko
- st Department of Anesthesiology and Intensive Care, Medical University of Warsaw, Warsaw, Poland
| | - Karolina Dobronska
- st Department of Anesthesiology and Intensive Care, Medical University of Warsaw, Warsaw, Poland
| | - Marcin Kolacz
- st Department of Anesthesiology and Intensive Care, Medical University of Warsaw, Warsaw, Poland
| | | | - Piotr Dobronski
- Department of Urology, Medical University of Warsaw, Warsaw, Poland
| |
Collapse
|
8
|
Lemaitre F, Chowdhury T, Schaller B. The trigeminocardiac reflex - a comparison with the diving reflex in humans. Arch Med Sci 2015; 11:419-26. [PMID: 25995761 PMCID: PMC4424259 DOI: 10.5114/aoms.2015.50974] [Citation(s) in RCA: 37] [Impact Index Per Article: 4.1] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 03/19/2013] [Revised: 05/03/2013] [Accepted: 06/30/2013] [Indexed: 12/21/2022] Open
Abstract
The trigeminocardiac reflex (TCR) has previously been described in the literature as a reflexive response of bradycardia, hypotension, and gastric hypermotility seen upon mechanical stimulation in the distribution of the trigeminal nerve. The diving reflex (DR) in humans is characterized by breath-holding, slowing of the heart rate, reduction of limb blood flow and a gradual rise in the mean arterial blood pressure. Although the two reflexes share many similarities, their relationship and especially their functional purpose in humans have yet to be fully elucidated. In the present review, we have tried to integrate and elaborate these two phenomena into a unified physiological concept. Assuming that the TCR and the DR are closely linked functionally and phylogenetically, we have also highlighted the significance of these reflexes in humans.
Collapse
Affiliation(s)
- Frederic Lemaitre
- Faculty of Sports Sciences, University of Rouen, Mont-Saint-Aignan, France
| | - Tumul Chowdhury
- Department of Anesthesia and Perioperative Medicine, University of Manitoba, Winnipeg, Canada
| | - Bernhard Schaller
- Department of Neurosurgery, University Hospital Paris, Paris, France
| |
Collapse
|