LDCNN: A new arrhythmia detection technique with ECG signals using a linear deep convolutional neural network.

The electrocardiogram (ECG) is a fundamental and widely used tool for diagnosing cardiovascular diseases. It involves recording cardiac electrical signals using electrodes, which illustrate the functioning of cardiac muscles during contraction and relaxation phases. ECG is instrumental in identifying abnormal cardiac activity, heart attacks, and various cardiac conditions. Arrhythmia detection, a critical aspect of ECG analysis, entails accurately classifying heartbeats. However, ECG signal analysis demands a high level of expertise, introducing the possibility of human errors in interpretation. Hence, there is a clear need for robust automated detection techniques. Recently, numerous methods have emerged for arrhythmia detection from ECG signals. In our research, we developed a novel one-dimensional deep neural network technique called linear deep convolutional neural network (LDCNN) to identify arrhythmias from ECG signals. We compare our suggested method with several state-of-the-art algorithms for arrhythmia detection. We evaluate our methodology using benchmark datasets, including the PTB Diagnostic ECG and MIT-BIH Arrhythmia databases. Our proposed method achieves high accuracy rates of 99.24% on the PTB Diagnostic ECG dataset and 99.38% on the MIT-BIH Arrhythmia dataset.

[Detection model of atrial fibrillation based on multi-branch and multi-scale convolutional networks].

Atrial fibrillation (AF) is a life-threatening heart condition, and its early detection and treatment have garnered significant attention from physicians in recent years. Traditional methods of detecting AF heavily rely on doctor's diagnosis based on electrocardiograms (ECGs), but prolonged analysis of ECG signals is very time-consuming. This paper designs an AF detection model based on the Inception module, constructing multi-branch detection channels to process raw ECG signals, gradient signals, and frequency signals during AF. The model efficiently extracted QRS complex and RR interval features using gradient signals, extracted P-wave and f-wave features using frequency signals, and used raw signals to supplement missing information. The multi-scale convolutional kernels in the Inception module provided various receptive fields and performed comprehensive analysis of the multi-branch results, enabling early AF detection. Compared to current machine learning algorithms that use only RR interval and heart rate variability features, the proposed algorithm additionally employed frequency features, making fuller use of the information within the signals. For deep learning methods using raw and frequency signals, this paper introduced an enhanced method for the QRS complex, allowing the network to extract features more effectively. By using a multi-branch input mode, the model comprehensively considered irregular RR intervals and P-wave and f-wave features in AF. Testing on the MIT-BIH AF database showed that the inter-patient detection accuracy was 96.89%, sensitivity was 97.72%, and specificity was 95.88%. The proposed model demonstrates excellent performance and can achieve automatic AF detection.

[Early classification and recognition algorithm for sudden cardiac arrest based on limited electrocardiogram data trained with a two-stages convolutional neural network].

Sudden cardiac arrest (SCA) is a lethal cardiac arrhythmia that poses a serious threat to human life and health. However, clinical records of sudden cardiac death (SCD) electrocardiogram (ECG) data are extremely limited. This paper proposes an early prediction and classification algorithm for SCA based on deep transfer learning. With limited ECG data, it extracts heart rate variability features before the onset of SCA and utilizes a lightweight convolutional neural network model for pre-training and fine-tuning in two stages of deep transfer learning. This achieves early classification, recognition and prediction of high-risk ECG signals for SCA by neural network models. Based on 16 788 30-second heart rate feature segments from 20 SCA patients and 18 sinus rhythm patients in the international publicly available ECG database, the algorithm performance evaluation through ten-fold cross-validation shows that the average accuracy (Acc), sensitivity (Sen), and specificity (Spe) for predicting the onset of SCA in the 30 minutes prior to the event are 91.79%, 87.00%, and 96.63%, respectively. The average estimation accuracy for different patients reaches 96.58%. Compared to traditional machine learning algorithms reported in existing literatures, the method proposed in this paper helps address the requirement of large training datasets for deep learning models and enables early and accurate detection and identification of high-risk ECG signs before the onset of SCA.

Type I Kounis syndrome in a young woman without chest pain: a case report.

BACKGROUND: Kounis syndrome is defined as the concurrence of acute coronary syndromes in the setting of allergic or anaphylactic reactions. It primarily affects men aged 40-70 years and is often associated with chest pain. This syndrome is often unrecognized and undiagnosed in clinical practice due to a low level of awareness. Herein, we present a case of type I Kounis syndrome in a young woman without chest pain. CASE PRESENTATION: A 28-year-old Japanese woman with a history of atopic dermatitis received a glycyrrhizin, glutathione, and neurotropin preparation (a preparation of inflamed skin extract from rabbits inoculated with vaccinia virus) at a dermatology clinic to treat pruritus caused by atopic dermatitis. Immediately after the administration, the patient developed abdominal pain and generalized body wheals. The patient was diagnosed with anaphylaxis and was transported to our hospital. She had no chest pain on arrival at our hospital; however, a 12-lead electrocardiogram showed ST elevation in leads I, aVL, V2, and V3, and an echocardiogram showed decreased wall motion in the anterior and lateral walls of the left ventricle. Sublingual nitroglycerin administration improved ST-segment elevation and left ventricular wall motion abnormalities. The patient underwent emergency coronary angiography, which revealed no significant stenosis, and was diagnosed with type I Kounis syndrome. CONCLUSION: Kounis syndrome without chest pain is rare in young women. Since it can be fatal in cases with severe allergic symptoms such as anaphylaxis, the possibility of concurrent acute coronary syndrome should be considered when treating systemic allergic reactions, regardless of age, sex, or the presence or absence of chest symptoms.

Cough-induced severe bradycardia and syncope in a dog.

A 10-year-old spayed female shih tzu dog was brought to the hospital because of recurring syncope that occurred simultaneously with a cough. Physical examination did not reveal an abnormal heart rhythm or abnormal heart sounds. Electrocardiography revealed sinus arrest of 4.7 s with intermittent escape beats during coughing. Additional examinations, including thoracic radiography, clinical pathology, and echocardiography, revealed no abnormalities of concern. Forty-eight-hour Holter monitoring captured 1 syncopal episode following severe coughing, during which the longest sinus arrest lasted 16 s with intermittent escape beats. This observation confirmed our strong suspicion that coughing was the cause of varying degrees of sinus arrest in this dog. Theophylline, codeine, and short-term prednisolone were prescribed to treat the dog's cough. The daily episodes of syncope ceased and coughing decreased. Subsequent 48-hour Holter monitoring revealed no abnormal pauses, and the owner did not report syncope. Theophylline and codeine were continued for 5 mo, during which time no syncope occurred. To our knowledge, this case provides the first clear evidence of a correlation between cough-induced sinus arrest and syncope in a veterinary patient, as confirmed by Holter monitoring and electrocardiography. Key clinical message: Cough-induced severe bradycardia and syncope were identified in a shih tzu dog. After the antitussive medication was adjusted, the signs resolved.

Bradycardie sévère et syncope provoquées par la toux chez un chienUne chienne shih tzu stérilisée âgée de 10 ans a été amenée à l'hôpital en raison d'une syncope récurrente survenue simultanément avec une toux. L'examen physique n'a révélé aucun rythme cardiaque anormal ni bruits cardiaques anormaux. L'électrocardiographie a révélé un arrêt sinusal de 4,7 s avec des battements d'échappements intermittents lors de la toux. Des examens complémentaires, notamment une radiographie thoracique, des analyses en pathologie clinique et une échocardiographie, n'ont révélé aucune anomalie préoccupante. Une surveillance Holter de 48 heures a capturé 1 épisode syncopal à la suite d'une toux sévère, au cours duquel l'arrêt sinusal le plus long a duré 16 s avec des battements d'échappements intermittents. Cette observation a confirmé nos fortes suspicions selon lesquelles la toux était la cause de divers degrés d'arrêt sinusal chez ce chien. De la théophylline, de la codéine et de la prednisolone de courte durée ont été prescrites pour traiter la toux du chien. Les épisodes quotidiens de syncope ont cessé et la toux a diminué. Une surveillance Holter ultérieure de 48 heures n'a révélé aucune pause anormale et le propriétaire n'a pas signalé de syncope. La théophylline et la codéine ont été poursuivies pendant 5 mois, période pendant laquelle aucune syncope ne s'est produite. À notre connaissance, ce cas constitue la première preuve claire d'une corrélation entre l'arrêt sinusal induit par la toux et la syncope chez un patient vétérinaire, comme le confirme la surveillance Holter et l'électrocardiographie.Message clinique clé :Une bradycardie et une syncope sévères induites par la toux ont été identifiées chez un chien shih tzu. Après ajustement du traitement antitussif, les signes ont disparu.(Traduit par Dr Serge Messier).

Zebra II as A Novel System to Record Electrophysiological Signals in Zebrafish.

Zebrafish and their mutant lines have been extensively used in biomedical investigations, cardiovascular studies, and drug screening. In the current study, the commercial version of the novel system, Zebra II, is presented. The protocol demonstrates electrocardiogram (ECG) acquisition and analysis from multiple zebrafish within controllable working environments. The device is composed of an external and independent perfusion system, a 4-point electrode, temperature sensors, and an embedded electronic system. In previous studies, the device prototype underwent validation against the established iWORX system through several tests, demonstrating similar data quality and ECG response to drug interventions. Following this, the study delved into examining the impact of anesthetic drugs and temperature fluctuations on zebrafish ECG, necessitating instant data evaluation. Thanks to the apparatus's capacity for consistent delivery of anesthetics and drugs, it was possible to extend ECG data collection up to 1 h, markedly longer than the 5 min duration supported by current systems. This paper introduces a pioneering, cloud-based, automated analysis utilizing data from four zebrafish, offering an efficient method for conducting combination experiments and significantly reducing time and effort. The system proved effective in capturing and analyzing ECG, especially in detecting drug-induced arrhythmias in wild-type zebrafish. Additionally, the capability to gather data across multiple channels facilitated the execution of randomized controlled trials with zebrafish models. The developed ECG system overcomes existing limitations, showing the potential to greatly expedite drug discovery and cardiovascular research involving zebrafish.

In Vivo Thoracic Dorsal Root Ganglia (DRG) Calcium Imaging and ECG Recording for Studying Peripheral Nerve Stimulation.

The dorsal root ganglia (DRG), housing primary sensory neurons, transmit somatosensory and visceral afferent inputs to the dorsal horn of the spinal cord. They play a pivotal role in both physiological and pathological states, including neuropathic and visceral pain. In vivo calcium imaging of DRG enables real-time observation of calcium transients in single units or neuron ensembles. Accumulating evidence indicates that DRG neuronal activities induced by somatic stimulation significantly affect autonomic and visceral functions. While lumbar DRG calcium imaging has been extensively studied, thoracic segment DRG calcium imaging has been less explored due to surgical exposure and stereotaxic fixation challenges. Here, we utilized in vivo calcium imaging at the thoracic1 dorsal root ganglion (T1-DRG) to investigate changes in neuronal activity resulting from somatic stimulations of the forelimb. This approach is crucial for understanding the somato-cardiac reflex triggered by peripheral nerve stimulations (PENS), such as acupuncture. Notably, synchronization of cardiac function was observed and measured by electrocardiogram (ECG), with T-DRG neuronal activities, potentially establishing a novel paradigm for somato-visceral reflex in the thoracic segments.

Takotsubo Cardiomyopathy Mimicking Obstructive Hypertrophic Cardiomyopathy.

Takotsubo cardiomyopathy (TCM) is characterized by transient left ventricular dysfunction, diagnosed via echocardiography or left ventriculography. In most cases, TCM involves an emotional, physical, or combined trigger. Acute coronary syndrome is one of the most frequent misdiagnoses in TCM patients due to electrocardiogram (ECG) abnormalities and elevated cardiac biomarkers. Typically, coronary angiography reveals no stenosis or occlusion of the coronary arteries. Hypertrophic cardiomyopathy (HCM) is a distinct pathology characterized by a hypertrophied left ventricle with various phenotypes. However, some reports have described TCM cases mimicking obstructive-type HCM in some patients. We present a case of a female patient diagnosed with TCM based on clinical, laboratory, and imaging tests. Differentiating TCM from HCM was challenging due to ventriculography and echocardiography findings, as hyperdynamic contraction of the basal segments of the left ventricle caused an increased left ventricular outflow tract (LVOT) gradient and severe mitral valve regurgitation. Detailed evaluation and close echocardiographic follow-up are essential in such rare cases.

Does IV fentanyl, frequently used in emergency departments, change QTC value? A prospective observational study.

BACKGROUND: Fentanyl is an opioid analgesic frequently used in the emergency department (ED) and is usually administered without knowing the QTC values of the patients or being monitored. However, the effect of fentanyl on QTC, prolongation or shortening, has not been elucidated. This study aimed to determine the effect of fentanyl on QTC. METHODS: This is a prospective observational study in the ED of a tertiary hospital on patients who received intravenous fentanyl for procedures other than intubation. ECG was performed before and at 1, 5, 15, 30, and 60 min after the initiation of fentanyl administration, and QTC value was calculated. Primary outcomes were QTC prolongation, defined as an increase in the QTC to ≥ 500 ms or any increase in QTC by ≥ 60 ms. RESULTS: The study included 109 patients. Of these, 60 patients were male, and the median age was 40. Compared with the baseline QTC value, statistically significant prolongation was detected at the 5th, 15th, 30th, and 60th minutes, with the maximum prolongation at 30 min, and the median was 13.08 ms. Most patients with QTC prolongation were female and over 40 years of age. Clinically, none of these patients developed malignant arrhythmias during the 60-minute monitored observation period. CONCLUSION: Fentanyl prolonged the QTC value statistically significantly. Although no patient developed malignant arrhythmia clinically, our results suggest that this QTC-prolonging effect should be considered when using fentanyl in patients at risk of torsades.

Monitoring of myocardial injury by serial measurements of QRS area and T area: The MaastrICCht cohort.

BACKGROUND: Manually derived electrocardiographic (ECG) parameters were not associated with mortality in mechanically ventilated COVID-19 patients in earlier studies, while increased high-sensitivity cardiac troponin-T (hs-cTnT) and N-terminal pro-B-type natriuretic peptide (NT-proBNP) were. To provide evidence for vectorcardiography (VCG) measures as potential cardiac monitoring tool, we investigated VCG trajectories during critical illness. METHODS: All mechanically ventilated COVID-19 patients were included in the Maastricht Intensive Care Covid Cohort between March 2020 and October 2021. Serum hs-cTnT and NT-proBNP concentrations were measured daily. Conversion of daily 12-lead ECGs to VCGs by a MATLAB-based script provided QRS area, T area, maximal QRS amplitude, and QRS duration. Linear mixed-effect models investigated trajectories in serum and VCG markers over time between non-survivors and survivors, adjusted for confounders. RESULTS: In 322 patients, 5461 hs-cTnT, 5435 NT-proBNP concentrations and 3280 ECGs and VCGs were analyzed. Non-survivors had higher hs-cTnT concentrations at intubation and both hs-cTnT and NT-proBNP significantly increased compared with survivors. In non-survivors, the following VCG parameters decreased more when compared to survivors: QRS area (-0.27 (95% CI) (-0.37 to -0.16, p < .01) µVs per day), T area (-0.39 (-0.62 to -0.16, p < .01) µVs per day), and maximal QRS amplitude (-0.01 (-0.01 to -0.01, p < .01) mV per day). QRS duration did not differ. CONCLUSION: VCG-derived QRS area and T area decreased in non-survivors compared with survivors, suggesting that an increase in myocardial damage and tissue loss play a role in the course of critical illness and may drive mortality. These VCG markers may be used to monitor critically ill patients.

Complex mechanism of brugada phenocopy: moderate hyponatremia and right ventricular compression by liver metastatic tumor - case report.

Brugada phenocopy (BrP) occurs in various clinical conditions and manifests as a Brugada-like ECG pattern with coved (type 1) or saddle-back (type 2) ST-segment elevation in the right precordial leads. Unlike Brugada syndrome (BrS), which is an inherited channelopathy, BrP is not associated with an increased risk of malignant arrhythmia. BrP has been reported in severe metabolic disturbances (significant hyponatremia, hypokalemia or hyperkalemia), mechanical heart compression, coronary artery disease, pulmonary embolism and myocarditis/pericarditis. The authors described a case of a 69-year-old female whose Brugada-like ECG was atypically associated with only moderate hyponatremia (127 mmol/l). She was admitted due to a skin and subcutaneous tissue infection of the left shank and coexistent urinary tract infection (without a fever). She had the history of advanced melanoma with multiple liver metastases. Her cardiac history was negative, especially the patient has never suffered from ventricular arrhythmias. ECG on admission showed saddle-back ST-segment elevation in the right precordial leads; however, the patient did not report any chest pain. Troponin I level and left ventricular function in echocardiography were normal while regional longitudinal strain in RV apex was decreased and showed post-systolic shortening. The substernal view revealed compression of the right ventricle (RV) by liver metastatic tumor. ECG changes disappeared quickly during natrium chloride supplementation and did not recur during hospitalization. This case illustrates that even moderate hyponatremia may be a reversible cause of BrP when other predisposing conditions (e.g. heart compression by tumor) coexist.

Interoceptive signals shape the earliest markers and neural pathway to awareness at the visual threshold.

Variations in interoceptive signals from the baroreceptors (BRs) across the cardiac and respiratory cycle can modulate cortical excitability and so affect awareness. It remains debated at what stages of processing they affect awareness-related event-related potentials (ERPs) in different sensory modalities. We investigated the influence of the cardiac (systole/diastole) and the respiratory (inhalation/exhalation) phase on awareness-related ERPs. Subjects discriminated visual threshold stimuli while their electroencephalogram, electrocardiogram, and respiration were simultaneously recorded. We compared ERPs and their intracranial generators for stimuli classified correctly with and without awareness as a function of the cardiac and respiratory phase. Cyclic variations of interoceptive signals from the BRs modulated both the earliest electrophysiological markers and the trajectory of brain activity when subjects became aware of the stimuli: an early sensory component (P1) was the earliest marker of awareness for low (diastole/inhalation) and a perceptual component (visual awareness negativity) for high (systole/exhalation) BR activity, indicating that BR signals interfere with the sensory processing of the visual input. Likewise, activity spread from the primary visceral cortex (posterior insula) to posterior parietal cortices during high and from associative interoceptive centers (anterior insula) to the prefrontal cortex during low BR activity. Consciousness is thereby resolved in cognitive/associative regions when BR is low and in perceptual centers when it is high. Our results suggest that cyclic fluctuations of BR signaling affect both the earliest markers of awareness and the brain processes underlying conscious awareness.

Cardiovascular risk and its influencing factors during exercise in apparently healthy Chinese adult population.

BACKGROUND: There are few studies on the safety of sub-maximal exercise testing of aerobic exercise in apparently healthy Chinese populations. The purpose of this study was to explore the frequency of exercise electrocardiography (ECG) abnormalities and the corresponding exercise intensities, as well as the associated influencing factors, during a symptom-limited stepwise incremental cardiopulmonary exercise test (CPET) in an apparently healthy Chinese population. METHODS: A cross-sectional study was done in four communities, one urban and one rural in the North (Beijing) and in the South (Hezhou, Guangxi) of China from 1 January 2017 to 31 December 2018, respectively. Total of 1642 participants was recruited, 918 were eligible and completed demographic indicators, routine blood indicators, physical activity status, symptom-limited CPET and exercise ECG were included in the analysis. RESULTS: Of the exercise ECG outcomes, 10 (1.1%) were positive and occurred at exercise intensities ≥ 62.50% heart rate reserve (HRR); 44 (4.8%) were equivocal and 864 (94.1%) were normal. Individuals with Cardiovascular Disease Risk Factor (CVDRF) = 3-4 were 2.6 times more likely to have a equivocal and abnormal exercise ECG than those with CVDRF = 0-2. Exercise ECGs of individuals with CVDRF = 5-7 were 5.4 times more likely to be positive and abnormal than exercise ECGs of individuals with CVDRF = 0-2. CONCLUSIONS: The exercise intensity of 62.5% HRR can be used as a safe upper limit for safe participation in exercise in apparently healthy Chinese population; the greater the number of CVDRFs, the greater the likelihood of cardiovascular risk during exercise.

[Characteristics of electrocardiogram in fulminant myocarditis].

Objective: To investugate the unique electrocardiogram (ECG) characteristics of fulminant myocarditis (FM) patients and provide important clues for the diagnosis of FM. Methods: This was a retrospective study. Patients diagnosed with acute myocarditis at Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology from February 2017 to April 2022 were enrolled and divided into fulminant myocarditis group (FM) and non-fulminant myocarditis group (NFM) according to clinical diagnosis. A total of 246 healthy people who underwent physical examination in the Health examination Center of Tongji Hospital at the same period were selected as the control group. The clinical data and ECG characteristics of the above 3 groups were analyzed and compared. Logistic regression model was used to analyze the influence of ECG parameters on left ventricular ejection fraction in FM patients. Receiver operating curves were constructed to evaluate the predictive value of different ECG parameters for FM. Results: A total of 180 patients were included in this study (FM group: n=123; NFM group: n=57), with an age of (35.0±16.2) years and 106 males (58.89%). Compared with NFM group, ECG was significantly abnormal in FM group, with a higher incidence of sinus tachycardia, ventricular tachycardia or ventricular fibrillation, escape rhythm, right bundle branch block, third degree atrioventricular block, ST-segment elevation, low voltage, prolonged QTc interval, and widened QRS wave in the FM group (all P<0.05). The ECG parameters showed that the amplitude of the full lead QRS wave in FM group was lower than that in NFM group (P<0.01). The average heart rate and QTc interval of FM group were significantly higher than those of NFM and control groups (all P<0.05). Although ST-segment elevation had a higher incidence in the FM group, ECG parameters showed that except for leads Ⅲ and aVF, the ST segment levels in all leads in the FM group were lower than those in the control group (all P<0.05). There was a statistically significant difference in some ST segment changes between FM and NFM groups, while there was no statistical difference between the NFM and control groups. Multivariate regression analysis showed that widened QRS wave and increased heart rate were the influencing factors for left ventricular systolic dysfunction in FM patients (OR=16.914, 95%CI: 1.367-209.224, P=0.028; OR=1.026, 95%CI: 1.010-1.042, P=0.001). Receiver operating curve analysis showed that heart rate>86.90 beat/min, QTc>431.50 ms, and RV5+SV1<1.72 mV had certain predictive value for FM diagnosis. Conclusions: FM patients displayed marked and severe ECG abnormalities, and characteristic changes in ECG can provide important first clues for the diagnosis of FM.

[Development of an Intelligent Multi-Parameter Sleep Diagnosis and Analysis System].

Sleep disordered breathing (SDB) is a common sleep disorder with an increasing prevalence. The current gold standard for diagnosing SDB is polysomnography (PSG), but existing PSG techniques have some limitations, such as long manual interpretation times, a lack of data quality control, and insufficient monitoring of gas metabolism and hemodynamics. Therefore, there is an urgent need in China's sleep clinical applications to develop a new intelligent PSG system with data quality control, gas metabolism assessment, and hemodynamic monitoring capabilities. The new system, in terms of hardware, detects traditional parameters like nasal airflow, blood oxygen levels, electrocardiography (ECG), electroencephalography (EEG), electromyography (EMG), electrooculogram (EOG), and includes additional modules for gas metabolism assessment via end-tidal CO 2 and O 2 concentration, and hemodynamic function assessment through impedance cardiography. On the software side, deep learning methods are being employed to develop intelligent data quality control and diagnostic techniques. The goal is to provide detailed sleep quality assessments that effectively assist doctors in evaluating the sleep quality of SDB patients.

[Development of ECG Common Mode Rejection Ratio Automatic Test System].

To comprehensively meet the test requirements for the common mode rejection ratio (CMRR) across different ECG-particular standards, this paper presents the design of an ECG CMRR automatic test system. The hardware component primarily consists of a test signal generation module, an automatic control network (which includes a resistance-capacitance network control module and a polarization voltage control module), and a noise level switching module. The software portion enables automatic control and user interaction. Experimental results indicate that the system is stable, reliable, and highly automated, capable of satisfying the test requirements of various ECG-particular standards, thus demonstrating a promising application prospect.

Diagnosis of Brugada Syndrome With a Sodium-Channel-Blocker Test: Who Should Be Tested? Who Should Not?

Intravenous infusion of sodium-channel blockers (SCB) with either ajmaline, flecainide, procainamide, or pilsicainide to unmask the ECG of Brugada syndrome is the drug challenge most commonly used for diagnostic purposes when investigating cases possibly related to inherited arrhythmia syndromes. For a patient undergoing an SCB challenge, the impact of a positive result goes well beyond its diagnostic implications. It is, therefore, appropriate to question who should undergo a SCB test to diagnose or exclude Brugada syndrome and, perhaps more importantly, who should not. We present a critical review of the benefits and drawbacks of the SCB challenge when performed in cardiac arrest survivors, patients presenting with syncope, family members of probands with confirmed Brugada syndrome, and asymptomatic patients with suspicious ECG.

Continuous sepsis trajectory prediction using tensor-reduced physiological signals.

The quick Sequential Organ Failure Assessment (qSOFA) system identifies an individual's risk to progress to poor sepsis-related outcomes using minimal variables. We used Support Vector Machine, Learning Using Concave and Convex Kernels, and Random Forest to predict an increase in qSOFA score using electronic health record (EHR) data, electrocardiograms (ECG), and arterial line signals. We structured physiological signals data in a tensor format and used Canonical Polyadic/Parallel Factors (CP) decomposition for feature reduction. Random Forests trained on ECG data show improved performance after tensor decomposition for predictions in a 6-h time frame (AUROC 0.67 ± 0.06 compared to 0.57 ± 0.08, p = 0.01 ). Adding arterial line features can also improve performance (AUROC 0.69 ± 0.07, p < 0.01 ), and benefit from tensor decomposition (AUROC 0.71 ± 0.07, p = 0.01 ). Adding EHR data features to a tensor-reduced signal model further improves performance (AUROC 0.77 ± 0.06, p < 0.01 ). Despite reduction in performance going from an EHR data-informed model to a tensor-reduced waveform data model, the signals-informed model offers distinct advantages. The first is that predictions can be made on a continuous basis in real-time, and second is that these predictions are not limited by the availability of EHR data. Additionally, structuring the waveform features as a tensor conserves structural and temporal information that would otherwise be lost if the data were presented as flat vectors.