A machine learning model to predict liver-related outcomes after the functional cure of chronic hepatitis B.

BACKGROUND & AIMS: The risk of hepatocellular carcinoma (HCC) and hepatic decompensation persists after hepatitis B surface antigen (HBsAg) seroclearance. This study aimed to develop and validate a machine learning model to predict the risk of liver-related outcomes (LROs) following HBsAg seroclearance. METHODS: A total of 4,787 consecutive patients who achieved HBsAg seroclearance between 2000 and 2022 were enrolled from 6 centers in South Korea and a territory-wide database in Hong Kong, comprising the training (n=944), internal validation (n=1,102), and external validation (n=2,741) cohorts. Three machine learning-based models were developed and compared in each cohort. The primary outcome was the development of any LRO, including HCC, decompensation, and liver-related death. RESULTS: During a median follow-up of 55.2 (interquartile range=30.1-92.3) months, 123 LROs were confirmed (1.1%/person-year) in the Korean cohort. A model with the best predictive performance in the training cohort was selected as the final model (designated as PLAN-B-CURE), which was constructed using a gradient boosting algorithm and 7 variables (age, sex, diabetes, alcohol consumption, cirrhosis, albumin, and platelet count). Compared to previous HCC prediction models, PLAN-B-CURE showed significantly superior accuracy in the training cohort (c-index: 0.82 vs. 0.63-0.70, all P<0.001; area under the receiver operating characteristic curve: 0.86 vs. 0.62-0.72, all P<0.01; area under the precision-recall curve: 0.53 vs. 0.13-0.29, all P<0.01). PLAN-B-CURE showed a reliable calibration function (Hosmer-Lemeshow test P>0.05) and these results were reproduced in the internal and external validation cohorts. CONCLUSION: This novel machine learning model consisting of 7 variables provides reliable risk prediction of LRO after HBsAg seroclearance that can be used for personalized surveillance.

Prediction of intraoperative hypotension using deep learning models based on non-invasive monitoring devices.

PURPOSE: Intraoperative hypotension is associated with adverse outcomes. Predicting and proactively managing hypotension can reduce its incidence. Previously, hypotension prediction algorithms using artificial intelligence were developed for invasive arterial blood pressure monitors. This study tested whether routine non-invasive monitors could also predict intraoperative hypotension using deep learning algorithms. METHODS: An open-source database of non-cardiac surgery patients ( https://vitadb.net/dataset ) was used to develop the deep learning algorithm. The algorithm was validated using external data obtained from a tertiary Korean hospital. Intraoperative hypotension was defined as a systolic blood pressure less than 90 mmHg. The input data included five monitors: non-invasive blood pressure, electrocardiography, photoplethysmography, capnography, and bispectral index. The primary outcome was the performance of the deep learning model as assessed by the area under the receiver operating characteristic curve (AUROC). RESULTS: Data from 4754 and 421 patients were used for algorithm development and external validation, respectively. The fully connected model of Multi-head Attention architecture and the Globally Attentive Locally Recurrent model with Focal Loss function were able to predict intraoperative hypotension 5 min before its occurrence. The AUROC of the algorithm was 0.917 (95% confidence interval [CI], 0.915-0.918) for the original data and 0.833 (95% CI, 0.830-0.836) for the external validation data. Attention map, which quantified the contributions of each monitor, showed that our algorithm utilized data from each monitor with weights ranging from 8 to 22% for determining hypotension. CONCLUSIONS: A deep learning model utilizing multi-channel non-invasive monitors could predict intraoperative hypotension with high accuracy. Future prospective studies are needed to determine whether this model can assist clinicians in preventing hypotension in patients undergoing surgery with non-invasive monitoring.

Machine-learning model to predict the tacrolimus concentration and suggest optimal dose in liver transplantation recipients: a multicenter retrospective cohort study.

Titrating tacrolimus concentration in liver transplantation recipients remains a challenge in the early post-transplant period. This multicenter retrospective cohort study aimed to develop and validate a machine-learning algorithm to predict tacrolimus concentration. Data from 443 patients undergoing liver transplantation between 2017 and 2020 at an academic hospital in South Korea were collected to train machine-learning models. Long short-term memory (LSTM) and gradient-boosted regression tree (GBRT) models were developed using time-series doses and concentrations of tacrolimus with covariates of age, sex, weight, height, liver enzymes, total bilirubin, international normalized ratio, albumin, serum creatinine, and hematocrit. We conducted performance comparisons with linear regression and populational pharmacokinetic models, followed by external validation using the eICU Collaborative Research Database collected in the United States between 2014 and 2015. In the external validation, the LSTM outperformed the GBRT, linear regression, and populational pharmacokinetic models with median performance error (8.8%, 25.3%, 13.9%, and - 11.4%, respectively; P < 0.001) and median absolute performance error (22.3%, 33.1%, 26.8%, and 23.4%, respectively; P < 0.001). Dosing based on the LSTM model's suggestions achieved therapeutic concentrations more frequently on the chi-square test (P < 0.001). Patients who received doses outside the suggested range were associated with longer ICU stays by an average of 2.5 days (P = 0.042). In conclusion, machine learning models showed excellent performance in predicting tacrolimus concentration in liver transplantation recipients and can be useful for concentration titration in these patients.

Prediction for Perioperative Stroke Using Intraoperative Parameters.

BACKGROUND: Perioperative stroke is a severe complication following surgery. To identify patients at risk for perioperative stroke, several prediction models based on the preoperative factors were suggested. Prediction models often focus on preoperative patient characteristics to assess stroke risk. However, most existing models primarily base their predictions on the patient's baseline characteristics before surgery. We aimed to develop a machine-learning model incorporating both pre- and intraoperative variables to predict perioperative stroke. METHODS AND RESULTS: This study included patients who underwent noncardiac surgery at 2 hospitals with the data of 15 752 patients from Seoul National University Hospital used for development and temporal internal validation, and the data of 449 patients from Boramae Medical Center used for external validation. Perioperative stroke was defined as a newly developed ischemic lesion on diffusion-weighted imaging within 30 days of surgery. We developed a prediction model composed of pre- and intraoperative factors (integrated model) and compared it with a model consisting of preoperative features alone (preoperative model). Perioperative stroke developed in 109 (0.69%) patients in the Seoul National University Hospital group and 11 patients (2.45%) in the Boramae Medical Center group. The integrated model demonstrated superior predictive performance with area under the curve values of 0.824 (95% CI, 0.762-0.880) versus 0.584 (95% CI, 0.499-0.667; P<0.001) in the internal validation; and 0.716 (95% CI, 0.560-0.859) versus 0.505 (95% CI, 0.343-0.654; P=0.018) in the external validation, compared to the preoperative model. CONCLUSIONS: We suggest that incorporating intraoperative factors into perioperative stroke prediction models can improve their accuracy.

Risk factor analysis and clinical experience of treating capsular contracture after prepectoral implant-based breast reconstruction.

Background: Capsular contracture is one of the most common and severe complications after implant-based breast reconstruction. Recently, prepectoral implant-based breast reconstruction using acellular dermal matrix (ADM) has become an alternative to subpectoral implant-based reconstruction. However, risk factors for capsular contracture associated with recent prepectoral reconstruction trends are not well refined yet. Thus, the aim of this study was to determine risk factors for capsular contracture, and share our experience of treating capsular contracture in prepectoral reconstruction. Methods: This retrospective comparative study focused on 110 patients who underwent prepectoral implant-based breast reconstruction with ADM. Risk factors of capsular contracture were analyzed by comparing a capsular contracture group (27 cases) and a non-capsular contracture group (83 cases). Secondary treatment after capsular contracture development was analyzed in capsular contracture group. Results: According to univariate and multivariate analyses of risk factors for capsular contracture, single staged implant-based reconstruction (direct-to-implant), infection, and postoperative radiotherapy were significantly related to the development of capsular contracture. Also, surgical intervention including capsulectomy and capsulotomy with implant change showed a significant higher remission rate than other groups. Conclusions: Our study provides insights into risk factors and treatment choices for capsular contracture after prepectoral implant-based breast reconstruction with ADM. These findings can aid selection of patients, postoperative care and preventative treatment before reconstruction.

Antagonistic Efficacy of Symbiotic Bacterium Xenorhabdus sp. SCG against Meloidogyne spp.

The inhabitation and parasitism of root-knot nematodes (RKNs) can be difficult to control, as its symptoms can be easily confused with other plant diseases; hence, identifying and controlling the occurrence of RKNs in plants remains an ongoing challenge. Moreover, there are only a few biological agents for controlling these harmful nematodes. In this study, Xenorhabdus sp. SCG isolated from entomopathogenic nematodes of genus Steinernema was evaluated for nematicidal effects under in vitro and greenhouse conditions. The cell-free filtrates of strain SCG showed nematicidal activity against Meloidogyne species J2s, with mortalities of > 88% at a final concentration of 10%, as well as significant nematicidal activity against the three other genera of plant-parasitic nematodes in a dose-dependent manner. Thymine was isolated as active compounds by assay-guided fractionation and showed high nematicidal activity against M. incognita. Greenhouse experiments suggested that cell-free filtrates of strain SCG efficiently controlled the nematode population in M. incognita-infested tomatoes (Solanum lycopersicum L., cv. Rutgers). In addition, a significant increase in host plant growth was observed after 45 days of treatment. To our knowledge, this is the first to demonstrate the nematicidal activity spectrum of isolated Xenorhabdus species and their application to S. lycopersicum L., cv. Rutgers under greenhouse conditions. Xenorhabdus sp. SCG could be a promising biological nematicidal agent with plant growth-enhancing properties.

INSPIRE, a publicly available research dataset for perioperative medicine.

We present the INSPIRE dataset, a publicly available research dataset in perioperative medicine, which includes approximately 130,000 surgical operations at an academic institution in South Korea over a ten-year period between 2011 and 2020. This comprehensive dataset includes patient characteristics such as age, sex, American Society of Anesthesiologists physical status classification, diagnosis, surgical procedure code, department, and type of anaesthesia. The dataset also includes vital signs in the operating theatre, general wards, and intensive care units (ICUs), laboratory results from six months before admission to six months after discharge, and medication during hospitalisation. Complications include total hospital and ICU length of stay and in-hospital death. We hope this dataset will inspire collaborative research and development in perioperative medicine and serve as a reproducible external validation dataset to improve surgical outcomes.

Non-invasive prediction of massive transfusion during surgery using intraoperative hemodynamic monitoring data.

OBJECTIVE: Failure to receive prompt blood transfusion leads to severe complications if massive bleeding occurs during surgery. For the timely preparation of blood products, predicting the possibility of massive transfusion (MT) is essential to decrease morbidity and mortality. This study aimed to develop a model for predicting MT 10 min in advance using non-invasive bio-signal waveforms that change in real-time. METHODS: In this retrospective study, we developed a deep learning-based algorithm (DLA) to predict intraoperative MT within 10 min. MT was defined as the transfusion of 3 or more units of red blood cells within an hour. The datasets consisted of 18,135 patients who underwent surgery at Seoul National University Hospital (SNUH) for model development and internal validation and 621 patients who underwent surgery at the Boramae Medical Center (BMC) for external validation. We constructed the DLA by using features extracted from plethysmography (collected at 500 Hz) and hematocrit measured during surgery. RESULTS: Among 18,135 patients in SNUH and 621 patients in BMC, 265 patients (1.46%) and 14 patients (2.25%) received MT during surgery, respectively. The area under the receiver operating characteristic curve (AUROC) of DLA predicting intraoperative MT before 10 min was 0.962 (95% confidence interval [CI], 0.948-0.974) in internal validation and 0.922 (95% CI, 0.882-0.959) in external validation, respectively. CONCLUSION: The DLA can successfully predict intraoperative MT using non-invasive bio-signal waveforms.

Association Between High Preoperative White Blood Cell-to-Hemoglobin Ratio and Postoperative Symptomatic Cerebral Vasospasm in Patients With Aneurysmal Subarachnoid Hemorrhage.

BACKGROUND: Cerebral vasospasm after aneurysmal subarachnoid hemorrhage (ASAH) is a serious complication and has a strong relationship with systemic inflammatory responses. Given previously reported relationships between leukocytosis and anemia with ASAH-related cerebral vasospasm, this study examined the association between the preoperative white blood cell-to-hemoglobin ratio (WHR) and postoperative symptomatic cerebral vasospasm (SCV) in patients with ASAH. METHODS: Demographic, preoperative (comorbidities, ASAH characteristics, laboratory findings), intraoperative (operation and anesthesia), and postoperative (SCV, other neurological complications, clinical course) data were retrospectively analyzed in patients with ASAH who underwent surgical or endovascular treatment of the culprit aneurysm. Patients were divided into high-WHR (n=286) and low-WHR (n=257) groups based on the optimal cutoff value of preoperative WHR (0.74), and stabilized inverse probability weighting was performed between the 2 groups. The predictive power of the WHR and other preoperative systemic inflammatory indices (neutrophil-to-albumin, neutrophil-to-lymphocyte, platelet-to-lymphocyte, platelet-to-neutrophil, platelet-to-white blood cell ratios, and systemic immune-inflammation index) for postoperative SCV was evaluated. RESULTS: Postoperative SCV was more frequent in the high-WHR group than in the low-WHR group before (33.2% vs. 12.8%; P<0.001) and after (29.4% vs. 19.1%; P=0.005) inverse probability weighting. Before weighting, the predictive power for postoperative SCV was the highest for the WHR among the preoperative systematic inflammatory indices investigated (area under receiver operating characteristics curve 0.66, P<0.001). After weighting, preoperative WHR ≥0.74 was independently associated with postoperative SCV (odds ratio 1.76; P=0.006). CONCLUSIONS: High preoperative WHR was an independent predictor of postoperative SCV in patients with ASAH.

Effect of hyperbilirubinemia on the accuracy of continuous non-invasive hemoglobin measurements in liver transplantation recipients.

This study evaluated the effect of hyperbilirubinemia on the accuracy of continuous non-invasive hemoglobin (SpHb) measurements in liver transplantation recipients. Overall, 1465 SpHb and laboratory hemoglobin (Hb) measurement pairs (n = 296 patients) were analyzed. Patients were grouped into normal (< 1.2 mg/dL), mild-to-moderate (1.2-3.0 mg/dL), and severe (> 3.0 mg/dL) hyperbilirubinemia groups based on the preoperative serum total bilirubin levels. Bland-Altman analysis showed a bias of 0.20 (95% limit of agreement, LoA: - 2.59 to 3.00) g/dL, 0.98 (95% LoA: - 1.38 to 3.35) g/dL, and 1.23 (95% LoA: - 1.16 to 3.63) g/dL for the normal, mild-to-moderate, and severe groups, respectively. The four-quadrant plot showed reliable trending ability in all groups (concordance rate > 92%). The rates of possible missed transfusion (SpHb > 7.0 g/dL for Hb < 7.0 g/dL) were higher in the hyperbilirubinemia groups (2%, 7%, and 12% for the normal, mild-to-moderate, and severe group, respectively. all P < 0.001). The possible over-transfusion rate was less than 1% in all groups. In conclusion, the use of SpHb in liver transplantation recipients with preoperative hyperbilirubinemia requires caution due to the positive bias and high risk of missed transfusion. However, the reliable trending ability indicated its potential use in clinical settings.

Real-time machine learning model to predict short-term mortality in critically ill patients: development and international validation.

BACKGROUND: A real-time model for predicting short-term mortality in critically ill patients is needed to identify patients at imminent risk. However, the performance of the model needs to be validated in various clinical settings and ethnicities before its clinical application. In this study, we aim to develop an ensemble machine learning model using routinely measured clinical variables at a single academic institution in South Korea. METHODS: We developed an ensemble model using deep learning and light gradient boosting machine models. Internal validation was performed using the last two years of the internal cohort dataset, collected from a single academic hospital in South Korea between 2007 and 2021. External validation was performed using the full Medical Information Mart for Intensive Care (MIMIC), eICU-Collaborative Research Database (eICU-CRD), and Amsterdam University Medical Center database (AmsterdamUMCdb) data. The area under the receiver operating characteristic curve (AUROC) was calculated and compared to that for the National Early Warning Score (NEWS). RESULTS: The developed model (iMORS) demonstrated high predictive performance with an internal AUROC of 0.964 (95% confidence interval [CI] 0.963-0.965) and external AUROCs of 0.890 (95% CI 0.889-0.891) for MIMIC, 0.886 (95% CI 0.885-0.887) for eICU-CRD, and 0.870 (95% CI 0.868-0.873) for AmsterdamUMCdb. The model outperformed the NEWS with higher AUROCs in the internal and external validation (0.866 for the internal, 0.746 for MIMIC, 0.798 for eICU-CRD, and 0.819 for AmsterdamUMCdb; p < 0.001). CONCLUSIONS: Our real-time machine learning model to predict short-term mortality in critically ill patients showed excellent performance in both internal and external validations. This model could be a useful decision-support tool in the intensive care units to assist clinicians.

Regulation of long-range BMP gradients and embryonic polarity by propagation of local calcium-firing activity.

Many amniote vertebrate species including humans can form identical twins from a single embryo, but this only occurs rarely. It has been suggested that the primitive-streak-forming embryonic region emits signals that inhibit streak formation elsewhere but the signals involved, how they are transmitted and how they act has not been elucidated. Here we show that short tracks of calcium firing activity propagate through extraembryonic tissue via gap junctions and prevent ectopic primitive streak formation in chick embryos. Cross-regulation of calcium activity and an inhibitor of primitive streak formation (Bone Morphogenetic Protein, BMP) via NF-κB and NFAT establishes a long-range BMP gradient spanning the embryo. This mechanism explains how embryos of widely different sizes can maintain positional information that determines embryo polarity. We provide evidence for similar mechanisms in two different human embryo models and in Drosophila, suggesting an ancient evolutionary origin.

Multicentre validation of a machine learning model for predicting respiratory failure after noncardiac surgery.

BACKGROUND: Postoperative respiratory failure is a serious complication that could benefit from early accurate identification of high-risk patients. We developed and validated a machine learning model to predict postoperative respiratory failure, defined as prolonged (>48 h) mechanical ventilation or reintubation after surgery. METHODS: Easily extractable electronic health record (EHR) variables that do not require subjective assessment by clinicians were used. From EHR data of 307,333 noncardiac surgical cases, the model, trained with a gradient boosting algorithm, utilised a derivation cohort of 99,025 cases from Seoul National University Hospital (2013-9). External validation was performed using three separate cohorts A-C from different hospitals comprising 208,308 cases. Model performance was assessed by area under the receiver operating characteristic (AUROC) curve and area under the precision-recall curve (AUPRC), a measure of sensitivity and precision at different thresholds. RESULTS: The model included eight variables: serum albumin, age, duration of anaesthesia, serum glucose, prothrombin time, serum creatinine, white blood cell count, and body mass index. Internally, the model achieved an AUROC of 0.912 (95% confidence interval [CI], 0.908-0.915) and AUPRC of 0.113. In external validation cohorts A, B, and C, the model achieved AUROCs of 0.879 (95% CI, 0.876-0.882), 0.872 (95% CI, 0.870-0.874), and 0.931 (95% CI, 0.925-0.936), and AUPRCs of 0.029, 0.083, and 0.124, respectively. CONCLUSIONS: Utilising just eight easily extractable variables, this machine learning model demonstrated excellent discrimination in both internal and external validation for predicting postoperative respiratory failure. The model enables personalised risk stratification and facilitates data-driven clinical decision-making.

Microbiologic Profile of Nipple Swab Culture and its Association With Postoperative Complications in Prosthetic Breast Reconstruction.

BACKGROUND: In breast surgeries, a lactiferous duct leading to lactic glands of breast parenchyma allows direct contamination by normal bacterial flora of the nipple-areola complex. Complete blockage of nipple flora from the intraoperative field is almost impossible. OBJECTIVES: We aimed to analyze the microbiological profile of nipple flora of breast cancer patients who underwent an implant-based immediate breast reconstruction after a total mastectomy, and to evaluate the association of nipple bacterial flora with postoperative complications. METHODS: A retrospective chart review was performed of patients who underwent an implant-based immediate breast reconstruction after a total mastectomy. A nipple swab culture was performed preoperatively. Patient demographics, surgical characteristics, and complications were compared between positive and negative nipple swab culture groups. Microbiological profile data including antibacterial resistance were collected. RESULTS: Among 128 breasts, 60 cases (46.9%) had positive preoperative nipple swab culture results. Staphylococcus epidermidis accounted for 41.4% of microorganisms isolated. A multivariate logistic regression analysis of postoperative complications revealed that the presence of nipple bacterial flora was a risk factor for capsular contracture. Seven cases of postoperative infection were analyzed. In 2 cases (40% of pathogen-proven infection), the causative pathogen matched the patient's nipple bacterial flora, which was methicillin-resistant S. epidermidis in both cases. CONCLUSIONS: Nipple bacterial flora was associated with an increased risk of capsular contracture. Preoperative analysis of nipple bacterial flora can be an informative source for treating clinically diagnosed postoperative infections. More studies are needed to determine the effectiveness of active antibiotic decolonization of the nipple.

High Intraoperative Serum Lactate Level is Associated with Acute Kidney Injury after Brain Tumor Resection.

BACKGROUND: Postoperative acute kidney injury (AKI) is associated with poor clinical outcomes. Identification of risk factors for postoperative AKI is clinically important. Serum lactate can increase in situations of inadequate oxygen delivery and is widely used to assess a patient's clinical course. We investigated the association between intraoperative serum lactate levels and AKI after brain tumor resection. METHODS: Demographics, medical and surgical history, tumor characteristics, surgery, anesthesia, preoperative and intraoperative blood test results, and postoperative clinical outcomes were retrospectively collected from 4131 patients who had undergone brain tumor resection. Patients were divided into high (n=1078) and low (n=3053) lactate groups based on an intraoperative maximum serum lactate level of 3.35 mmol/L. After propensity score matching, 1005 patients were included per group. AKI was diagnosed using the Kidney Disease Improving Global Outcomes criteria, based on serum creatinine levels within 7 days after surgery. RESULTS: Postoperative AKI was observed in 53 (1.3%) patients and was more frequent in those with high lactate both before (3.2% [n=35] vs. 0.6% [n=18]; P < 0.001) and after (3.3% [n=33] vs. 0.6% [n=6]; P < 0.001) propensity score matching. Intraoperative predictors of postoperative AKI were maximum serum lactate levels > 3.35 mmol/L (odds ratio [95% confidence interval], 3.57 [1.45-8.74], P = 0.005), minimum blood pH (odds ratio per 1 unit, 0.01 [0.00-0.24], P = 0.004), minimum hematocrit (odds ratio per 1%, 0.91 [0.84-1.00], P = 0.037), and mean serum glucose levels > 200 mg/dL (odds ratio, 6.22 [1.75-22.16], P = 0.005). CONCLUSION: High intraoperative serum lactate levels were associated with AKI after brain tumor resection.

Continuous cardiac output estimation using a new modified Fick method during off-pump coronary artery bypass graft surgery: a retrospective observational study.

PURPOSE: Several technical aspects of the Fick method limit its use intraoperatively. A data-driven modification of the Fick method may enable its use in intraoperative settings. METHODS: This two-center retrospective observational study included 57 (28 and 29 in each center) patients who underwent off-pump coronary artery bypass graft (OPCAB) surgery. Intraoperative recordings of physiological data were obtained and divided into training and test datasets. The Fick equation was used to calculate cardiac output (CO-Fick) using ventilator-determined variables, intraoperative hemoglobin level, and SvO2, with continuous thermodilution cardiac output (CCO) used as a reference. A modification CO-Fick was derived and validated: CO-Fick-AD, which adjusts the denominator of the original equation. RESULTS: Increased deviation between CO-Fick and CCO was observed when oxygen extraction was low. The root mean square error of CO-Fick was decreased from 6.07 L/min to 0.70 L/min after the modification. CO-Fick-AD showed a mean bias of 0.17 (95% CI 0.00-0.34) L/min, with a 36.4% (95% CI 30.6-44.4%) error. The concordance rates of CO-Fick-AD ranged from 73.3 to 87.1% depending on the time interval and exclusion zone. CONCLUSIONS: The original Fick method is not reliable when oxygen extraction is low, but a modification using data-driven approach could enable continuous estimation of cardiac output during the dynamic intraoperative period with minimal bias. However, further improvements in precision and trending ability are needed.

Comparison of invasive and non-invasive measurements of cardiac index and systemic vascular resistance in living-donor liver transplantation: a prospective, observational study.

BACKGROUND: Based on the controversy surrounding pulmonary artery catheterization (PAC) in surgical patients, we investigated the interchangeability of cardiac index (CI) and systemic vascular resistance (SVR) measurements between ClearSight™ and PAC during living-donor liver transplantation (LDLT). METHODS: This prospective study included consecutively selected LDLT patients. ClearSight™-based CI and SVR measurements were compared with those from PAC at seven LDLT-stage time points. ClearSight™-based systolic (SAP), mean (MAP), and diastolic (DAP) arterial pressures were also compared with those from femoral arterial catheterization (FAC). For the comparison and analysis of ClearSight™ and the reference method, Bland-Altman analysis was used to analyze accuracy while polar and four-quadrant plots were used to analyze the trending ability. RESULTS: From 27 patients, 189 pairs of ClearSight™ and reference values were analyzed. The CI and SVR performance errors (PEs) exhibited poor accuracy between the two methods (51.52 and 51.73%, respectively) in the Bland-Altman analysis. CI and SVR also exhibited unacceptable trending abilities in both the polar and four-quadrant plot analyses. SAP, MAP, and DAP PEs between the two methods displayed favorable accuracy (24.28, 21.18, and 26.26%, respectively). SAP and MAP exhibited acceptable trending ability in the four-quadrant plot between the two methods, but not in the polar plot analyses. CONCLUSIONS: During LDLT, CI and SVR demonstrated poor interchangeability, while SAP and MAP exhibited acceptable interchangeability between ClearSight™ and FAC.

Automatic segmentation of atrial fibrillation and flutter in single-lead electrocardiograms by self-supervised learning and Transformer architecture.

OBJECTIVES: Automatic detection of atrial fibrillation and flutter (AF/AFL) is a significant concern in preventing stroke and mitigating hemodynamic instability. Herein, we developed a Transformer-based deep learning model for AF/AFL segmentation in single-lead electrocardiograms (ECGs) by self-supervised learning with masked signal modeling (MSM). MATERIALS AND METHODS: We retrieved data from 11 open-source databases on PhysioNet; 7 of these databases included labeled ECGs, while the other 4 were without labels. Each database contained ECG recordings with durations of ≥30 s. A total of 24 intradialytic ECGs with paroxysmal AF/AFL during 4 h of hemodialysis sessions at Seoul National University Hospital were used for external validation. The model was pretrained by predicting masked areas of ECG signals and fine-tuned by predicting AF/AFL areas. Cross-database validation was used for evaluation, and the intersection over union (IOU) was used as a main performance metric in external database validation. RESULTS: In the 7 labeled databases, the areas marked as AF/AFL constituted 41.1% of the total ECG signals, ranging from 0.19% to 51.31%. In the evaluation per ECG segment, the model achieved IOU values of 0.9254 and 0.9477 for AF/AFL segmentation and other segmentation tasks, respectively. When applied to intradialytic ECGs with paroxysmal AF/AFL, the IOUs for the segmentation of AF/AFL and non-AF/AFL were 0.9896 and 0.9650, respectively. Model performance by different training procedure indicated that pretraining with MSM and the application of an appropriate masking ratio both contributed to the model performance. It also showed higher IOUs of AF/AFL labels than in previous studies when training and test databases were matched. CONCLUSION: The present model with self-supervised learning by MSM performs robustly in segmenting AF/AFL.

Real-time machine learning model to predict in-hospital cardiac arrest using heart rate variability in ICU.

Predicting in-hospital cardiac arrest in patients admitted to an intensive care unit (ICU) allows prompt interventions to improve patient outcomes. We developed and validated a machine learning-based real-time model for in-hospital cardiac arrest predictions using electrocardiogram (ECG)-based heart rate variability (HRV) measures. The HRV measures, including time/frequency domains and nonlinear measures, were calculated from 5 min epochs of ECG signals from ICU patients. A light gradient boosting machine (LGBM) algorithm was used to develop the proposed model for predicting in-hospital cardiac arrest within 0.5-24 h. The LGBM model using 33 HRV measures achieved an area under the receiver operating characteristic curve of 0.881 (95% CI: 0.875-0.887) and an area under the precision-recall curve of 0.104 (95% CI: 0.093-0.116). The most important feature was the baseline width of the triangular interpolation of the RR interval histogram. As our model uses only ECG data, it can be easily applied in clinical practice.