Design and Implementation of an Intensive Care Unit Command Center for Medical Data Fusion.

The rapid advancements in Artificial Intelligence of Things (AIoT) are pivotal for the healthcare sector, especially as the world approaches an aging society which will be reached by 2050. This paper presents an innovative AIoT-enabled data fusion system implemented at the CMUH Respiratory Intensive Care Unit (RICU) to address the high incidence of medical errors in ICUs, which are among the top three causes of mortality in healthcare facilities. ICU patients are particularly vulnerable to medical errors due to the complexity of their conditions and the critical nature of their care. We introduce a four-layer AIoT architecture designed to manage and deliver both real-time and non-real-time medical data within the CMUH-RICU. Our system demonstrates the capability to handle 22 TB of medical data annually with an average delay of 1.72 ms and a bandwidth of 65.66 Mbps. Additionally, we ensure the uninterrupted operation of the CMUH-RICU with a three-node streaming cluster (called Kafka), provided a failed node is repaired within 9 h, assuming a one-year node lifespan. A case study is presented where the AI application of acute respiratory distress syndrome (ARDS), leveraging our AIoT data fusion approach, significantly improved the medical diagnosis rate from 52.2% to 93.3% and reduced mortality from 56.5% to 39.5%. The results underscore the potential of AIoT in enhancing patient outcomes and operational efficiency in the ICU setting.

Enhancement of bicycle exercise capacity in patients with chronotropic incompetence through closed-loop stimulation: a randomized crossover trial.

AIMS: This study aimed to investigate the effectiveness of closed-loop stimulation (CLS) pacing compared with the traditional DDD mode in patients with chronotropic incompetence (CI) using bicycle-based cardiopulmonary exercise testing (CPET). METHODS AND RESULTS: This single-centre, randomized crossover trial involved 40 patients with CI. Patients were randomized to receive either DDD-CLS or DDD mode pacing for 2 months, followed by a crossover to the alternative mode for an additional 2 months. Bicycling-based CPET was conducted at the 3- and 5-month follow-up visits to assess exercise capacity. Other cardiopulmonary exercise outcome measures and health-related quality of life (QoL) were also assessed. DDD-CLS mode pacing significantly improved exercise capacity, resulting in a peak oxygen uptake (14.8 ± 4.0 vs. 12.0 ± 3.6 mL/kg/min, P < 0.001) and oxygen uptake at the ventilatory threshold (10.0 ± 2.2 vs. 8.7 ± 1.8 mL/kg/min, P < 0.001) higher than those of the DDD mode. However, there were no significant differences in other cardiopulmonary exercise outcome measures such as ventilatory efficiency of carbon dioxide production slope, oxygen uptake efficiency slope, and end-tidal carbon dioxide between the two modes. Patients in the DDD-CLS group reported a better QoL, and 97.5% expressed a preference for the DDD-CLS mode. CONCLUSION: DDD-CLS mode pacing demonstrated improved exercise capacity and QoL in patients with CI, highlighting its potential as an effective pacing strategy for this patient population.

Comparison of efficacy of pulmonary vein isolation between cryoballoon ablation and high-power short-duration ablation.

BACKGROUND: High-power short-duration (HPSD) and cryoballoon ablation (CBA) has been used for pulmonary vein isolation (PVI). OBJECTIVE: We aimed to compare the efficacy of PVI between CBA and HPSD ablation in patients with paroxysmal atrial fibrillation (PAF). METHODS: We retrospectively analyzed 251 consecutive PAF patients from January 2018 to July 2020. Of them, 124 patients (mean age 57.2 ± 10.1 year) received HPSD and 127 patients (mean age 59.6 ± 9.4 year) received CBA. In HPSD group, the radiofrequency energy was set as 50 W/10 s at anterior wall and 40 W/10 s at posterior wall. In CBA group, 28 mm s generation cryoballoon was used for PVI according the guidelines. RESULTS: There was no significant difference in baseline characteristics between these 2 groups. The time to achieve PVI was significantly shorter in cryoballoon ablation group than in HPSD group (20.6 ± 1.7 min vs 51.8 ± 36.3, P = 0.001). The 6-month overall recurrence for atrial tachyarrhythmias was not significantly different between the two groups (HPSD:14.50% vs CBA:11.0%, P = 0.40). There were different types of recurrent atrial tachyarrhythmia between these 2 groups. Recurrence as atrial flutter was significantly more common in CBA group compared to HPSD group (57.1% vs 12.5%, P = 0.04). CONCLUSION: In PAF patients, CBA and HPSD had a favourable and comparable outcome. The recurrence pattern was different between CBA and HPSD groups.

Prediction of Recurrent Atrial Tachyarrhythmia After Receiving Atrial Flutter Ablation in Patients With Prior Cardiac Surgery for Valvular Heart Disease.

Background: Surgical scars cause an intra-atrial conduction delay and anatomical obstacles that facilitate the perpetuation of atrial flutter (AFL). This study aimed to investigate the outcome and predictor of recurrent atrial tachyarrhythmia after catheter ablation in patients with prior cardiac surgery for valvular heart disease (VHD) who presented with AFL. Methods: Seventy-two patients with prior cardiac surgery for VHD who underwent AFL ablation were included. The patients were categorized into a typical AFL group (n = 45) and an atypical AFL group (n = 27). The endpoint was the recurrence of atrial tachyarrhythmia during follow-up. A multivariate analysis was performed to determine the predictor of recurrence. Results: No significant difference was found in the recurrence rate of atrial tachyarrhythmia between the two groups. Patients with concomitant atrial fibrillation (AF) had a higher recurrence of typical AFL compared with those without AF (13 vs. 0%, P = 0.012). In subgroup analysis, typical AFL patients with concomitant AF had a higher incidence of recurrent atrial tachyarrhythmia than those without it (53 vs. 14%, P = 0.006). Regarding patients without AF, the typical AFL group had a lower recurrence rate of atrial tachyarrhythmia than the atypical AFL group (14 vs. 40%, P = 0.043). Multivariate analysis showed that chronic kidney disease (CKD) and left atrial diameter (LAD) were independent predictors of recurrence. Conclusions: In our study cohort, concomitant AF was associated with recurrence of atrial tachyarrhythmia. CKD and LAD independently predicted recurrence after AFL ablation in patients who have undergone cardiac surgery for VHD.

Spatiotemporal differences in precordial electrocardiographic amplitude before and after flecainide provocation are associated with a history of unstable ventricular arrhythmia in Brugada syndrome.

INTRODUCTION: A drug provocation test (DPT) is important for the diagnosis of Brugada syndrome (BrS). The link, however, between dynamic changes of electrocardiography (ECG) features after DPT and unstable ventricular arrhythmia (VA) in BrS remains unknown. METHODS: Between 2014 and 2019, we assessed 27 patients with BrS (median age: 37.0 [interquartile range, IQR: 22.0-51.0] years; 25 men), including 9 (33.3%) with a history of unstable VA and 18 (66.7%) without. All patients in the study presented with Brugada-like ECG features before DPT. The ECG parameters and dynamic changes (∆) in 12-lead ECGs recorded from the second, third, and fourth intercostal spaces (ICS) before and at 1, 6, 12, 18, and 24 h after DPT (oral flecainide 400 mg) were analyzed. RESULTS: The total amplitude of V1 at the third ICS 18 and 24 h after DPT was significantly lower in patients with a history of unstable VA than in those without. Patients with BrS and unstable VAs had a significantly larger ∆ amplitude of V1 at the second ICS 12 h after DPT than in those without unstable VAs (0.28 [0.18-0.41] mV vs. 0.08 [0.01-0.15] mV, p = .01). A multivariate analysis revealed that the amplitude of V1 at the third ICS 18 and 24 h after DPT and the ∆ amplitude of V1 at the second ICS 12 h after DPT were associated with a history of unstable VA. CONCLUSION: Nonuniform changes and spatiotemporal differences in precordial ECG features after DPT were observed in patients with BrS and these may be surrogate markers for risk stratification.

Applying the CHA2DS2-VASc score to predict the risk of future acute coronary syndrome in patients receiving catheter ablation for atrial fibrillation.

OBJECTIVE: It remains unknown whether catheter ablation for atrial fibrillation (AF) reduces future acute coronary syndrome (ACS) risk or whether the CHA2DS2-VASc score has a role in predicting this risk. We aimed to compare very long-term risk of ACS between patients who received catheter ablation to AF or antiarrhythmic medications and controls without AF. METHODS: Propensity scores were calculated for each patient and used to assemble a cohort of 787 patients undergoing AF ablation in 2003-2012. Patients were compared to an equal number of AF patients treated with antiarrhythmic medications and a control group without AF. Patients with previous coronary events were excluded. The primary endpoint was ACS occurrence. RESULTS: Baseline clinical characteristics were comparable. After a mean 9.1 ± 3.2-year follow-up, the ablation group had lower incidence of new onset ACS than the medication and non-AF control groups (annual incidence: 0.15%. 0.78%, and 0.35%; with 2.67, 4.16, and 10.44 cases/1000 person-years, respectively; P < 0.001). After adjusting for multiple confounders, the ablation group had lower future ACS risk than the medication (hazard ratio [HR]: 0.20, 95% confidence interval [CI]: 0.13-0.30) and control groups (HR: 0.30, 95% CI: 0.20-0.45). The CHA2DS2-VASc score was a strong predictor of ACS (HR: 1.61, 95% CI: 1.47-1.76; AUC: 85.9%, 95% CI: 78.5-93.2%). A baseline CHA2DS2-VASc score ≥ 4 predicted future ACS (positive predictive rate: 14.3%). CONCLUSIONS: This study suggested that catheter ablation for AF may be beneficial to reduce future ACS risk in AF patients, and a high baseline CHA2DS2-VASc score can predict future acute coronary events.

Compressed-Domain ECG-Based Biometric User Identification Using Compressive Analysis.

Nowadays, user identification plays a more and more important role for authorized machine access and remote personal data usage. For reasons of privacy and convenience, biometrics-based user identification, such as iris, fingerprint, and face ID, has become mainstream methods in our daily lives. However, most of the biometric methods can be easily imitated or artificially cracked. New types of biometrics, such as electrocardiography (ECG), are based on physiological signals rather than traditional biological traits. Recently, compressive sensing (CS) technology that combines both sampling and compression has been widely applied to reduce the power of data acquisition and transmission. However, prior CS-based frameworks suffer from high reconstruction overhead and cannot directly align compressed ECG signals. In this paper, in order to solve the above two problems, we propose a compressed alignment-aided compressive analysis (CA-CA) algorithm for ECG-based biometric user identification. With CA-CA, it can avoid reconstruction and extract information directly from CS-based compressed ECG signals to reduce overall complexity and power. Besides, CA-CA can also align the compressed ECG signals in the eigenspace-domain, which can further enhance the precision of identifications and reduce the total training time. The experimental result shows that our proposed algorithm has a 94.16% accuracy based on a public database of 22 people.

Identification of critical isthmus using coherent mapping in patients with scar-related atrial tachycardia.

INTRODUCTION: Accurate identification of slow conducting regions in patients with scar-related atrial tachycardia (AT) is difficult using conventional electrogram annotation for cardiac electroanatomic mapping (EAM). Estimating delays between neighboring mapping sites is a potential option for activation map computation. We describe our initial experience with CARTO 3 Coherent Mapping (Biosense Webster Inc,) in the ablation of complex ATs. METHODS: Twenty patients (58 ± 10 y/o, 15 males) with complex ATs were included. We created three-dimensional EAMs using CARTO 3 system with CONFIDENSE and a high-resolution mapping catheter (Biosense Webster Inc). Local activation time and coherent maps were used to aid in the identification of conduction isthmus (CI) and focal origin sites. System-defined slow or nonconducting zones and CI, defined by concealed entrainment (postpacing interval < 20 ms), CV < 0.3 m/s and local fractionated electrograms were evaluated. RESULTS: Twenty-six complex ATs were mapped (mean: 1.3 ± 0.7 maps/pt; 4 focal, 22 isthmus-dependent). Coherent mapping was better in identifying CI/breakout sites where ablation terminated the tachycardia (96.2% vs 69.2%; P = .010) and identified significantly more CI (mean/chamber 2.0 ± 1.1 vs 1.0 ± 0.7; P < .001) with narrower width (19.8 ± 10.5 vs 43.0 ± 23.9 mm; P < .001) than conventional mapping. Ablation at origin and CI sites was successful in 25 (96.2%) with long-term recurrence in 25%. CONCLUSIONS: Coherent mapping with conduction velocity vectors derived from adjacent mapping sites significantly improved the identification of CI sites in scar-related ATs with isthmus-dependent re-entry better than conventional mapping. It may be used in conjunction with conventional mapping strategies to facilitate recognition of slow conduction areas and critical sites that are important targets of ablation.

Application of noninvasive signal-averaged electrocardiogram analysis in predicting the requirement of epicardial ablation in patients with arrhythmogenic right ventricular cardiomyopathy.

BACKGROUND: Signal-averaged electrocardiogram (SAECG) provides not only diagnostic information but also the prognostic implication of ablation in arrhythmogenic right ventricular cardiomyopathy (ARVC). OBJECTIVE: This study aimed to validate the role of SAECG in identifying arrhythmogenic substrates requiring an epicardial approach in ARVC. METHODS: Ninety-one patients with a definite diagnosis of ARVC who underwent successful ablation for drug-refractory ventricular arrhythmia were enrolled and classified into 2 groups: group 1 who underwent successful ablation at the endocardium only and group 2 who underwent successful ablation requiring an additional epicardial approach. The baseline characteristics of patients and SAECG parameters were obtained for analysis. RESULTS: Male predominance, worse right ventricular (RV) function, higher incidence of syncope, and depolarization abnormality were observed in group 2. Moreover, the number of abnormal SAECG criteria was higher in group 2 than in group 1. After a multivariate analysis, the independent predictors of the requirement of epicardial ablation included the number of abnormal SAECG criteria (odds ratio 2.8, 95% confidence interval 1.4-5.4; P = .003) and presence of syncope (odds ratio 11.7; 95% confidence interval 2.7-50.4; P = .001). In addition, ≥2 abnormal SAECG criteria were associated with larger RV endocardial unipolar low-voltage zone (P < .001), larger RV endocardial/epicardial bipolar low-voltage zone/scar (P < .05), and longer RV endocardial/epicardial total activation time (P < .001 and P = .004, respectively). CONCLUSION: The number of abnormal SAECG criteria was correlated with the extent of diseased epicardial substrates and could be a potential surrogate marker for predicting the requirement of epicardial ablation in patients with ARVC.

Clinical efficacy of open-irrigated electrode cooled with half-normal saline for initially failed radiofrequency ablation of idiopathic outflow tract ventricular arrhythmias.

BACKGROUND: Acute failure of radiofrequency ablation (RFA) of ventricular arrhythmias (VAs) occur in 10%-20% of patients and is partly attributed to inadequate lesion depth acquired with standard ablation protocols. Half-normal saline (HNS)-irrigation is a promising strategy to improve the success rate of VA ablation. OBJECTIVE: This study investigated the efficacy of HNS-irrigated ablation after a failed standard plain normal saline solution (PNSS)-irrigated ablation on idiopathic outflow tract ventricular arrhythmia (OT-VA). METHOD: This is a prospective observational study of consecutive patients undergoing RFA of idiopathic OT-VA comparing the efficacy of additional HNS-irrigated ablation for failed standard PNSS-irrigated ablation. Acute failure was defined as persistence of spontaneous VA or persistent inducibility of the clinical VA. RESULTS: Out of 160 OT-VA cases (51 ± 15-year-old, 62 males), 31 underwent HNS irrigation after a failed standard PNSS-irrigated ablation. The HNS group had a significantly longer procedure time (60.06 ± 43.83 vs 37.51 ± 33.40 minutes; P = .013) and higher radiation exposure (31.45 ± 20.24 vs 17.22 ± 15.25 minutes; P = .001) than the PNSS group but provided an additional acute success in 21 of 31 (67.7%) patients. Over a follow-up duration of 7.8 ± 4.6 months, 24 recurrences were identified, including 8 (25.8%) in the HNS and 16 (12.4%) in the PNSS group, with lower freedom from recurrence in the HNS group (log rank P = .009). No major complication was observed. CONCLUSION: HNS-irrigated ablation after failed standard PNSS-irrigated ablation is safe and additionally improves acute ablation success by 67.7% for idiopathic OT-VA but with a higher rate of recurrence on follow-up. Whether the application of HNS as initial irrigant could result in better outcome requires further investigation.

Low-Complexity Privacy-Preserving Compressive Analysis Using Subspace-Based Dictionary for ECG Telemonitoring System.

Compressive sensing (CS) is attractive in long-term electrocardiography (ECG) telemonitoring to extend life-time for resource-constrained wireless wearable sensors. However, the availability of transmitted personal information has posed great concerns for potential privacy leakage. Moreover, the traditional CS-based security frameworks focus on secured signal recovery instead of privacy-preserving data analytics; hence, they provide only computational secrecy and have impractically high complexities for decryption. In this paper, to protect privacy from an information-theoretic perspective while delivering the classification capability, we propose a low-complexity framework of Privacy-Preserving Compressive Analysis (PPCA) based on subspace-based representation. The subspace-based dictionary is used for both encrypting and decoding the CS measurements online, and it is built by dividing signal space into discriminative and complementary subspace offline. The encrypted signal is unreconstructable even if the eavesdropper cracks the measurement matrix and the dictionary. PPCA is implemented in ECG-based atrial fibrillation detection. It can reduce the mutual information by 1.98 bits via encrypting measurements with signal-dependent noise at 1 dB, while the classification accuracy remains 96.05% with the decoding matrix. Furthermore, by decoding via matrix-vector product, rather than sparse coding, this computational complexity of PPCA is 341 times fewer compared with the traditional CS-based security.