Outcomes after cryoballoon ablation of paroxysmal atrial fibrillation with the PolarX or the Arctic Front Advance Pro: a prospective multicentre experience.

AIMS: The aim of this study was to compare procedural efficacy and safety, including 1-year freedom from AF recurrence, between the novel cryoballoon system PolarX (Boston Scientific) and the Arctic Front Advance Pro (AFA-Pro) (Medtronic), in patients with paroxysmal AF undergoing PVI. METHODS AND RESULTS: This multicentre prospective observational study included 267 consecutive patients undergoing a first cryoablation procedure for paroxysmal AF (137 PolarX, 130 AFA-Pro). Kaplan-Meier curves with the log-rank test was used to compare the 1-year freedom from AF recurrence between both groups. Multivariate Cox model was performed to evaluate whether the type of procedure (PolarX vs. AFA-Pro) had an impact on the occurrence of AF recurrences after adjustment on potentially confounding factors. The PolarX reaches lower temperatures than the AFA-Pro (LSPV 52 ± 5, vs. 59 ± 6; LIPV 49 ± 6 vs. 56 ± 6; right superior pulmonary vein: 49 ± 6 vs. 57 ± 7; right inferior pulmonary vein: 52 ± 6 vs. 59 ± 6; P < 0.0001). A higher rate of transient phrenic nerve palsy was found in patients treated with the PolarX system (15% vs. 7%, P = 0.05). After a mean follow-up of 15 ± 5 months, 20 patients (15%) had recurrences in AFA-Pro group and 27 patients (19%) in PolarX group (P = 0.35). Based on survival analysis, no significant difference was observed between both groups with a 12-month free of recurrence survival of 91.2% (85.1-95.4%) vs. 83.7% (76.0%-89.1%) (log-rank test P = 0.11). In multivariate Cox model hazard ratio of recurrence for PolarX vs. AFA-Pro was not significant [HR = 1.6 (0.9-2.8), P = 0.12]. CONCLUSION: PolarX and AFA-Pro have comparable efficacy and safety profiles for pulmonary veins isolation in paroxysmal atrial fibrillation.

EUropean real-world outcomes with Pulsed field ablatiOn in patients with symptomatic atRIAl fibrillation: lessons from the multi-centre EU-PORIA registry.

AIMS: Pulsed field ablation (PFA) is a new, non-thermal ablation modality for pulmonary vein (PV) isolation in patients with atrial fibrillation (AF). The multi-centre EUropean Real World Outcomes with Pulsed Field AblatiOn in Patients with Symptomatic AtRIAl Fibrillation (EU-PORIA) registry sought to determine the safety, efficacy, and learning curve characteristics for the pentaspline, multi-electrode PFA catheter. METHODS AND RESULTS: All-comer AF patients from seven high-volume centres were consecutively enrolled. Procedural and follow-up data were collected. Learning curve effects were analysed by operator ablation experience and primary ablation modality. In total, 1233 patients (61% male, mean age 66 ± 11years, 60% paroxysmal AF) were treated by 42 operators. In 169 patients (14%), additional lesions outside the PVs were performed, most commonly at the posterior wall (n = 127). Median procedure and fluoroscopy times were 58 (interquartile range: 40-87) and 14 (9-21) min, respectively, with no differences due to operator experience. Major complications occurred in 21/1233 procedures (1.7%) including pericardial tamponade (14; 1.1%) and transient ischaemic attack or stroke (n = 7; 0.6%), of which one was fatal. Prior cryoballoon users had less complication. At a median follow-up of 365 (323-386) days, the Kaplan-Meier estimate of arrhythmia-free survival was 74% (80% for paroxysmal and 66% for persistent AF). Freedom from arrhythmia was not influenced by operator experience. In 149 (12%) patients, a repeat procedure was performed due to AF recurrence and 418/584 (72%) PVs were durably isolated. CONCLUSION: The EU-PORIA registry demonstrates a high single-procedure success rate with an excellent safety profile and short procedure times in a real-world, all-comer AF patient population.

Pulsed electric field, cryoballoon, and radiofrequency for paroxysmal atrial fibrillation ablation: a propensity score-matched comparison.

AIMS: Pulsed field ablation (PFA) has emerged as a novel, non-thermal energy source to selectively ablate cardiac tissue. We describe a multicentre experience on pulmonary vein isolation (PVI) via the pentaspline Farapulse™ PFA system vs. thermal-based technologies in a propensity score-matched population of paroxysmal atrial fibrillation (PAF) patients. METHODS AND RESULTS: Propensity score matching was adopted to compare PVI-only ablation outcomes via the Farawave™ system (Group PFA), cryoballoon (Group CRYO), or focal radiofrequency (Group RF) (PFA:CRYO:RF ratio = 1:2:2). Among 1572 (mean age: 62.4 ± 11.3 years; 42.5% females) PAF patients undergoing first time PVI with either PFA (n = 174), CRYO (n = 655), or RF (n = 743), propensity score matching yielded 174 PFA, 348 CRYO, and 348 RF patients. First-pass isolation was achieved in 98.8% of pulmonary veins (PVs) with PFA, 81.5% with CRYO, and 73.1% with RF (P < 0.001). Procedural and dwell times were significantly shorter with PFA, whereas the availability of a 3D mapping system led to a significant reduction in X-ray exposure with RF. Overall complication rates were 3.4% (n = 6) with PFA, 8.6% (n = 30) with CRYO, and 5.5% (n = 19) with RF (P = 0.052). The 1-year Kaplan-Meier estimated freedom from any atrial tachyarrhythmia was 79.3% with PFA, 74.7% with CRYO, and 72.4% with RF (log-rank P-value: 0.24). Among 145 repeat ablation procedures, PV reconnection rate was 19.1% after PFA, 27.5% after CRYO, and 34.8% after RF (P = 0.01). CONCLUSION: Pulsed field ablation contributed to significantly shorter procedural times. Follow-up data showed a similar arrhythmia freedom, although a higher rate of PV reconnection was documented in post-CRYO and post-RF redo procedures.

Machine Learning and Syncope Management in the ED: The Future Is Coming.

In recent years, machine learning (ML) has been promisingly applied in many fields of clinical medicine, both for diagnosis and prognosis prediction. Aims of this narrative review were to summarize the basic concepts of ML applied to clinical medicine and explore its main applications in the emergency department (ED) setting, with a particular focus on syncope management. Through an extensive literature search in PubMed and Embase, we found increasing evidence suggesting that the use of ML algorithms can improve ED triage, diagnosis, and risk stratification of many diseases. However, the lacks of external validation and reliable diagnostic standards currently limit their implementation in clinical practice. Syncope represents a challenging problem for the emergency physician both because its diagnosis is not supported by specific tests and the available prognostic tools proved to be inefficient. ML algorithms have the potential to overcome these limitations and, in the future, they could support the clinician in managing syncope patients more efficiently. However, at present only few studies have addressed this issue, albeit with encouraging results.

A large language model-based clinical decision support system for syncope recognition in the emergency department: A framework for clinical workflow integration.

Differentiation of syncope from transient loss of consciousness can be challenging in the emergency department (ED). Natural Language Processing (NLP) enables the analysis of free text in the electronic medical records (EMR). The present paper aimed to develop a large language models (LLM) for syncope recognition in the ED and proposed a framework for model integration within the clinical workflow. Two models, based on both the Italian and Multilingual Bidirectional Encoder Representations from Transformers (BERT) language model, were developed using consecutive EMRs. The "triage" model was only based on notes contained in the "triage" section of the EMR. The "anamnesis" model added data contained in the "medical history" section. Interpretation and calibration plots were generated. The Italian and Multi BERT models were developed and tested on both 15,098 and 15,222 EMRs, respectively. The triage model had an AUC of 0·95 for the Italian BERT and 0·94 for the Multi BERT. The anamnesis model had an AUC of 0·98 for the Italian BERT and 0·97 for Multi BERT. The LLM identified syncope when not explicitly mentioned in the EMR and also recognized common prodromal symptoms preceding syncope. Both models identified syncope patients in the ED with a high discriminative capability from nurses and doctors' notes, thus potentially acting as a tool helping physicians to differentiate syncope from others transient loss of consciousness.

Will Artificial Intelligence Be "Better" Than Humans in the Management of Syncope?

Clinical decision-making regarding syncope poses challenges, with risk of physician error due to the elusive nature of syncope pathophysiology, diverse presentations, heterogeneity of risk factors, and limited therapeutic options. Artificial intelligence (AI)-based techniques, including machine learning (ML), deep learning (DL), and natural language processing (NLP), can uncover hidden and nonlinear connections among syncope risk factors, disease features, and clinical outcomes. ML, DL, and NLP models can analyze vast amounts of data effectively and assist physicians to help distinguish true syncope from other types of transient loss of consciousness. Additionally, short-term adverse events and length of hospital stay can be predicted by these models. In syncope research, AI-based models shift the focus from causality to correlation analysis between entities. This prompts the search for patterns rather than defining a hypothesis to be tested a priori. Furthermore, education of students, doctors, and health care providers engaged in continuing medical education may benefit from clinical cases of syncope interacting with NLP-based virtual patient simulators. Education may be of benefit to patients. This article explores potential strengths, weaknesses, and proposed solutions associated with utilization of ML and DL in syncope diagnosis and management. Three main topics regarding syncope are addressed: 1) clinical decision-making; 2) clinical research; and 3) education. Within each domain, we question whether "AI will be better than humans," seeking evidence to support our objective inquiry.

Durability of Pulmonary Vein Isolation Using Pulsed-Field Ablation: Results From the Multicenter EU-PORIA Registry.

BACKGROUND: Pulsed-field ablation (PFA) is a novel nonthermal ablation technology with high procedural safety and efficiency for pulmonary vein isolation (PVI). Premarket data showed high PVI durability during mandatory remapping studies. Data on lesion durability in real-world patients with clinically indicated redo procedures are scarce. OBJECTIVES: This study sought to report PVI durability rates in patients undergoing a clinically indicated redo procedure after an index PVI using PFA. METHODS: Patients from 7 European centers undergoing an index PVI using PFA were included the EU-PORIA (European Real-world Outcomes With Pulsed Field Ablation in Patients With Symptomatic Atrial Fibrillation) registry. In patients with subsequent left atrial redo procedures due to arrhythmia recurrence, 3-dimensional electroanatomical maps were acquired. PVI durability was assessed on a per-vein and per-patient level, and sites of reconnections and predictors of lesion durability were identified. RESULTS: Of 1,184 patients (62% paroxysmal atrial fibrillation) undergoing an index PVI using PFA, 272 (23%) had an arrhythmia recurrence. Of these, 144 (53%) underwent a left atrial redo procedure a median of 7 (Q1-Q3: 5-10) months after the first ablation. Three-dimensional electroanatomical maps identified 404 of 567 pulmonary veins (71%) with durable isolation. In 54 patients (38%), all pulmonary veins were durably isolated. Prior operator experience with cryoballoon ablation was associated with a higher PVI durability compared to operators with only point-by-point radiofrequency experience (76% vs 60%; P < 0.001). Neither the operators' cumulative experience in atrial fibrillation ablation (≤5 vs >5 years) nor the size of the PFA device used (31 mm vs 35 mm) had an impact on subsequent lesion durability (both P > 0.50). CONCLUSIONS: In 144 patients with arrhythmia recurrence after PFA PVI, durable isolation was observed in 71% of the pulmonary veins during the redo procedure, and 38% of all patients showed durable isolation of all veins.

Characterization and Clinical Significance of Hemolysis After Pulsed Field Ablation for Atrial Fibrillation: Results of a Multicenter Analysis.

BACKGROUND: Pulsed field ablation (PFA) is increasingly used in clinical practice for the treatment of atrial fibrillation. While the susceptibility of erythrocytes to electroporation is well established, the effect of cardiac PFA technologies on hemolysis has remained underreported. The aim of this study was to investigate the incidence, severity, and clinical impact of PFA-induced hemolysis. METHODS: We included n=145 patients undergoing atrial fibrillation catheter ablation with a pentaspline PFA catheter (biphasic, bipolar pulses of 2 kV) and n=70 patients receiving radiofrequency ablation (40-90 W) at 4 high-volume European centers. The lesion set comprised pulmonary vein isolation for paroxysmal atrial fibrillation and pulmonary vein isolation±additional lesions for persistent atrial fibrillation. Hemolysis and renal function biomarkers were analyzed in blood samples at baseline, at the end of ablation, and 24 hours after the procedure. RESULTS: Baseline characteristics were well balanced between groups (overall mean, 65.7±9.4 years; 69.3% male). The ablation procedures comprised a mean of 61.6±27.4 PFA deliveries and 26.3±15.0 minute RF duration. Hemolysis was detected in 94.3% versus 6.8% of patients after PFA versus radiofrequency ablation (P<0.001): PFA was associated with significantly lower haptoglobin levels (0.5±0.4 versus 1.0±0.4 g/L), while free plasma hemoglobin (592.8±330.6 versus 147.8±183.0 mg/L), bilirubin (21.3±11.3 versus 14.8±8.8 µmol/L), and LDH (lactate dehydrogenase, 352.7±115.7 versus 253.2±56.5 U/L) were significantly higher after PFA versus radiofrequency ablation (all P<0.001). Hemolysis correlated with the number of PFA deliveries (r=0.62 [95% CI, 0.33-0.80]; P<0.001), with the highest severity occurring ≥54 PFA deliveries. After PFA, hemoglobinuria occurred in 36.4%, while creatinine increase was higher in patients with baseline glomerular filtration rate <50 mL/min versus baseline glomerular filtration rate >50 mL/min (Δcrea, 27.0±103.1 versus -0.2±12.1 µmol/L; P=0.010). CONCLUSIONS: Intravascular hemolysis is a frequent finding after PFA and increases with the number of PFA deliveries. Until the clinical impact of PFA-associated hemolysis is fully elucidated, a careful titration of PFA deliveries during the ablation procedure is warranted.

Safety and efficacy of pulsed-field ablation for atrial fibrillation in the elderly: A EU-PORIA sub-analysis.

BACKGROUND: The role of catheter ablation in elderly patients with atrial fibrillation (AF) is unclear. Pulsed field ablation (PFA) demonstrates a favorable clinical profile, however, data on elderly patients are lacking. AIMS: We aimed to assess the safety and efficacy of PFA in the elderly, using data from the EU-PORIA registry. METHODS: Periprocedural complications and long-term safety and efficacy outcomes of AF ablation using the pentaspline PFA catheter (Farapulse™) were compared between patients older than 80 years old and their younger counterparts, across seven European centers. RESULTS: Among the 1233 patients in the registry, 88 (7.1 %) were older than 80 years. Elderly patients were more often females (51.1 % vs 37.8 %, p = 0.01) with a lower median BMI (26.0, IQR:23.5-29.2 vs 26.9, IQR:24.4-30.4 kg/m2, p = 0.02), a higher median CHA2DS2-VASc score (4, IQR:3-5 vs 2, IQR:1-3, p < 0.001) and a higher incidence of hypertension (73.9 % vs 52.7 %, p < 0.001). In both groups, most patients had paroxysmal AF (58.0 % vs 60.3 %, p = 0.65). Ablation in the elderly was more frequently performed with minimally interrupted anticoagulation (87.5 % vs 59.7 %, p < 0.001). Despite comparable rates of overall complications (5.7 % vs 3.5 %, p = 0.29), elderly patients had a higher incidence of stroke (2.3 % vs 0.3 %, p = 0.04). At 12 months, major adverse clinical events (4.5 % vs. 2.1 %, p = 0.12) and arrhythmia-free survival (70 % vs 74 %, p = 0.69) were comparable in both groups. None of the recurrence-free elderly patients were on antiarrhythmic drugs at the end of follow-up. CONCLUSION: In this real-world cohort, the efficacy of PFA for AF was similar in elderly and younger patients. Despite comparable complication rates, a higher incidence of stroke was observed in the elderly.

Pulsed field ablation of the right superior pulmonary vein prevents vagal responses via anterior right ganglionated plexus modulation.

BACKGROUND: Pulsed field ablation (PFA) is selective for the myocardium. However, vagal responses and reversible effects on ganglionated plexi (GP) are observed during pulmonary vein isolation (PVI). Anterior-right GP ablation has been proven to effectively prevent vagal responses during radiofrequency-based PVI. OBJECTIVE: The purpose of this study was to test the hypothesis that PFA-induced transient anterior-right GP modulation when targeting the right superior pulmonary vein (RSPV) before any other pulmonary veins (PVs) may effectively prevent intraprocedural vagal responses. METHODS: Eighty consecutive paroxysmal atrial fibrillation patients undergoing PVI with PFA were prospectively included. In the first 40 patients, PVI was performed first targeting the left superior pulmonary vein (LSPV-first group). In the last 40 patients, RSPV was targeted first, followed by left PVs and right inferior PV (RSPV-first group). Heart rate (HR) and extracardiac vagal stimulation (ECVS) were evaluated at baseline, during PVI, and postablation to assess GP modulation. RESULTS: Vagal responses occurred in 31 patients (78%) in the LSPV-first group and 5 (13%) in the RSPV-first group (P <.001). Temporary pacing was needed in 14 patients (35%) in the LSPV-first group and 3 (8%) in the RSPV-first group (P = .003). RSPV isolation was associated with similar acute HR increase in the 2 groups (13 ± 11 bpm vs 15 ± 12 bpm; P = .3). No significant residual changes in HR or ECVS response were documented in both groups at the end of the procedure compared to baseline (all P >.05). CONCLUSION: PVI with PFA frequently induced vagal responses when initiated from the LSPV. Nevertheless, an RSPV-first approach promoted transient HR increase and reduced vagal response occurrence.

Catheter Ablation for Atrial Fibrillation in Structural Heart Disease: A Review.

Atrial fibrillation (AF) is the most common arrhythmia encountered in clinical practice. Patients with structural heart disease (SHD) are at an increased risk of developing this arrhythmia and are particularly susceptible to the deleterious hemodynamic effects it carries. In the last two decades, catheter ablation (CA) has emerged as a valuable strategy for rhythm control and is currently part of the standard care for symptomatic relief in patients with AF. Growing evidence suggests that CA of AF may have potential benefits that extend beyond symptoms. In this review, we summarize the current knowledge of this intervention on SHD patients.

Case Report: A leadless and endovascular pacemaker teamwork.

Background: Cardiac Implantable Electronic Device infections increase short- and long-term mortality, along with healthcare costs. Leadless pacemakers (PM) were developed to overcome pocket- and minimize lead-related complications in selected high-risk patients. Recent advancements enable leadless devices to mechanically detect atrial activity, facilitating atrioventricular (AV) synchronous stimulation. Case summary: A 90-year-old woman, implanted with a dual-chamber pacemaker eight years ago due to sinus node dysfunction, presented with syncope. A diagnosis of complete AV block, in the setting of ventricular lead dysfunction was made. Due to a high risk of infection, the patient was implanted with a leadless PM capable of maintaining AV synchrony in VDD mode (MICRA™ model MC1AVR1). The transvenous PM was programmed to AAI-R mode to drive the atria, which, in turn, triggered the leadless PM to stimulate the ventricles. At six month follow-up, the AV synchrony rate was 85%. Conclusion: The combination of classic atrial pacing with leadless ventricular stimulation can be used in high-risk patients to reduce the risk of complications, in the setting of ventricular lead dysfunction. In this manner, AV synchrony can be maintained, improving hemodynamic parameters and quality of life. Low sinus rate variability at rest is essential to achieve a high AV synchrony rate in such cases.

A novel stepwise approach incorporating ethanol infusion in the vein of Marshall for the ablation of persistent atrial fibrillation.

Introduction: Apart from pulmonary vein isolation (PVI), several step-by-step procedures that aim to modify left atrial substrate have been proposed for the ablation of persistent atrial fibrillation (AF), yet the optimal strategy remains elusive. There are cumulative data suggesting an incremental benefit of adding vein of Marshall (VOM) ethanol infusion to PVI in patients with persistent AF. We sought to evaluate the feasibility and efficacy of a novel stepwise ablation approach, incorporating a VOM alcoholization step, for persistent AF. Methods: In this single-center study, we prospectively enrolled 66 consecutive patients with symptomatic persistent AF and failure of at least one antiarrhythmic drug (ADD). The ablation procedure consisted of (i) PVI, (ii) left atrial segmentation with VOM ethanol infusion and the deployment of linear radiofrequency lesions across the roof and the mitral isthmus and (iii) electrogram-based ablation of dispersion zones. The first two steps were performed in all patients, whereas the third step was carried out only in those still in AF at the end of the second step. Atrial tachycardias during the procedure were mapped and ablated. At the end of the procedure, cavotricuspid isthmus ablation was additionally performed in all patients. The primary endpoint was 12-month freedom from AF and atrial tachycardia after a single procedure and an initial three-month blanking period. Results: Total procedure time was 153 ± 38.5 min. Fluoroscopy time was 16 ± 6.5 min and the radiofrequency ablation time was 26.14 ± 0.26 min. The primary endpoint occurred in 54 patients (82%). At 12 months, 65% of patients were off any AAD. In the univariate Cox regression analysis, left ventricular ejection fraction < 40% was the only predictor of arrhythmia recurrence (HR 3.56; 95% CI, 1.04-12.19; p = 0.04). One patient developed a pericardial tamponade and another a minor groin hematoma. Conclusion: A novel stepwise approach, including a step of ethanol infusion in the VOM, is feasible, safe and provides a high rate of sinus rhythm maintenance at 12 months in patients with persistent AF.

Temporal Trends in Pacemaker Implantations Over the Past Decade in France: Impact of Transcatheter Aortic Valve Implantations.

In this study, we assessed the temporal trends of permanent pacemaker implantations in France from 2008 to 2018 using data from the Échantillon Généraliste de Bénéficiaires (EGB) administrative database, a representative sample of the French population. Additionally, we evaluated the impact of transcatheter aortic valve implantations on the overall pacemaker implantation rate. Our data suggest that the incidence of permanent pacemaker implantations in France increased significantly only in patients ≥80 years old, with post-transcatheter aortic valve implantations accounting for at least 2/3 of this increase.

A Hybrid Model for 30-Day Syncope Prognosis Prediction in the Emergency Department.

Syncope is a challenging problem in the emergency department (ED) as the available risk prediction tools have suboptimal predictive performances. Predictive models based on machine learning (ML) are promising tools whose application in the context of syncope remains underexplored. The aim of the present study was to develop and compare the performance of ML-based models in predicting the risk of clinically significant outcomes in patients presenting to the ED for syncope. We enrolled 266 consecutive patients (age 73, IQR 58-83; 52% males) admitted for syncope at three tertiary centers. We collected demographic and clinical information as well as the occurrence of clinically significant outcomes at a 30-day telephone follow-up. We implemented an XGBoost model based on the best-performing candidate predictors. Subsequently, we integrated the XGboost predictors with knowledge-based rules. The obtained hybrid model outperformed the XGboost model (AUC = 0.81 vs. 0.73, p < 0.001) with acceptable calibration. In conclusion, we developed an ML-based model characterized by a commendable capability to predict adverse events within 30 days post-syncope evaluation in the ED. This model relies solely on clinical data routinely collected during a patient's initial syncope evaluation, thus obviating the need for laboratory tests or syncope experienced clinical judgment.

Multimodal deep learning for COVID-19 prognosis prediction in the emergency department: a bi-centric study.

Predicting clinical deterioration in COVID-19 patients remains a challenging task in the Emergency Department (ED). To address this aim, we developed an artificial neural network using textual (e.g. patient history) and tabular (e.g. laboratory values) data from ED electronic medical reports. The predicted outcomes were 30-day mortality and ICU admission. We included consecutive patients from Humanitas Research Hospital and San Raffaele Hospital in the Milan area between February 20 and May 5, 2020. We included 1296 COVID-19 patients. Textual predictors consisted of patient history, physical exam, and radiological reports. Tabular predictors included age, creatinine, C-reactive protein, hemoglobin, and platelet count. TensorFlow tabular-textual model performance indices were compared to those of models implementing only tabular data. For 30-day mortality, the combined model yielded slightly better performances than the tabular fastai and XGBoost models, with AUC 0.87 ± 0.02, F1 score 0.62 ± 0.10 and an MCC 0.52 ± 0.04 (p < 0.32). As for ICU admission, the combined model MCC was superior (p < 0.024) to the tabular models. Our results suggest that a combined textual and tabular model can effectively predict COVID-19 prognosis which may assist ED physicians in their decision-making process.

An accurate and time-efficient deep learning-based system for automated segmentation and reporting of cardiac magnetic resonance-detected ischemic scar.

BACKGROUND AND OBJECTIVES: Myocardial infarction scar (MIS) assessment by cardiac magnetic resonance provides prognostic information and guides patients' clinical management. However, MIS segmentation is time-consuming and not performed routinely. This study presents a deep-learning-based computational workflow for the segmentation of left ventricular (LV) MIS, for the first time performed on state-of-the-art dark-blood late gadolinium enhancement (DB-LGE) images, and the computation of MIS transmurality and extent. METHODS: DB-LGE short-axis images of consecutive patients with myocardial infarction were acquired at 1.5T in two centres between Jan 1, 2019, and June 1, 2021. Two convolutional neural network (CNN) models based on the U-Net architecture were trained to sequentially segment the LV and MIS, by processing an incoming series of DB-LGE images. A 5-fold cross-validation was performed to assess the performance of the models. Model outputs were compared respectively with manual (LV endo- and epicardial border) and semi-automated (MIS, 4-Standard Deviation technique) ground truth to assess the accuracy of the segmentation. An automated post-processing and reporting tool was developed, computing MIS extent (expressed as relative infarcted mass) and transmurality. RESULTS: The dataset included 1355 DB-LGE short-axis images from 144 patients (MIS in 942 images). High performance (> 0.85) as measured by the Intersection over Union metric was obtained for both the LV and MIS segmentations on the training sets. The performance for both LV and MIS segmentations was 0.83 on the test sets. Compared to the 4-Standard Deviation segmentation technique, our system was five times quicker (<1 min versus 7 ± 3 min), and required minimal user interaction. CONCLUSIONS: Our solution successfully addresses different issues related to automatic MIS segmentation, including accuracy, time-effectiveness, and the automatic generation of a clinical report.

Artificial intelligence-assisted quantification of COVID-19 pneumonia burden from computed tomography improves prediction of adverse outcomes over visual scoring systems.

OBJECTIVE: We aimed to evaluate the effectiveness of utilizing artificial intelligence (AI) to quantify the extent of pneumonia from chest CT scans, and to determine its ability to predict clinical deterioration or mortality in patients admitted to the hospital with COVID-19 in comparison to semi-quantitative visual scoring systems. METHODS: A deep-learning algorithm was utilized to quantify the pneumonia burden, while semi-quantitative pneumonia severity scores were estimated through visual means. The primary outcome was clinical deterioration, the composite end point including admission to the intensive care unit, need for invasive mechanical ventilation, or vasopressor therapy, as well as in-hospital death. RESULTS: The final population comprised 743 patients (mean age 65  ±â€¯ 17 years, 55% men), of whom 175 (23.5%) experienced clinical deterioration or death. The area under the receiver operating characteristic curve (AUC) for predicting the primary outcome was significantly higher for AI-assisted quantitative pneumonia burden (0.739, p = 0.021) compared with the visual lobar severity score (0.711, p < 0.001) and visual segmental severity score (0.722, p = 0.042). AI-assisted pneumonia assessment exhibited lower performance when applied for calculation of the lobar severity score (AUC of 0.723, p = 0.021). Time taken for AI-assisted quantification of pneumonia burden was lower (38 ± 10 s) compared to that of visual lobar (328 ± 54 s, p < 0.001) and segmental (698 ± 147 s, p < 0.001) severity scores. CONCLUSION: Utilizing AI-assisted quantification of pneumonia burden from chest CT scans offers a more accurate prediction of clinical deterioration in patients with COVID-19 compared to semi-quantitative severity scores, while requiring only a fraction of the analysis time. ADVANCES IN KNOWLEDGE: Quantitative pneumonia burden assessed using AI demonstrated higher performance for predicting clinical deterioration compared to current semi-quantitative scoring systems. Such an AI system has the potential to be applied for image-based triage of COVID-19 patients in clinical practice.

Changes in smell and taste perception related to COVID-19 infection: a case-control study.

The main aim of the present study was to psychophysically evaluate smell and taste functions in hospitalized COVID-19 patients and to compare those results with a group of healthy subjects. Another aim of the study was to assess the relationship of changes in patients' smell and taste functions with a number of clinical parameters, symptoms, and other physiological signs as well as with severity of disease. Olfactory and gustatory functions were tested in 61 hospitalized patients positive for SARS-CoV-2 infection and in a control group of 54 healthy individuals. Overall, we found a significant impairment of olfactory and gustatory functions in COVID-19 patients compared with the control group. Indeed, about 45% of patients self-reported complaints about or loss of either olfactory or gustatory functions. These results were confirmed by psychophysical testing, which showed a significantly reduced performance in terms of intensity perception and identification ability for both taste and smell functions in COVID-19 patients. Furthermore, gustatory and olfactory impairments tended to be more evident in male patients suffering from more severe respiratory failure (i.e., pneumonia with need of respiratory support need during hospitalization).

Learning Analytics Applied to Clinical Diagnostic Reasoning Using a Natural Language Processing-Based Virtual Patient Simulator: Case Study.

BACKGROUND: Virtual patient simulators (VPSs) log all users' actions, thereby enabling the creation of a multidimensional representation of students' medical knowledge. This representation can be used to create metrics providing teachers with valuable learning information. OBJECTIVE: The aim of this study is to describe the metrics we developed to analyze the clinical diagnostic reasoning of medical students, provide examples of their application, and preliminarily validate these metrics on a class of undergraduate medical students. The metrics are computed from the data obtained through a novel VPS embedding natural language processing techniques. METHODS: A total of 2 clinical case simulations (tests) were created to test our metrics. During each simulation, the students' step-by-step actions were logged into the program database for offline analysis. The students' performance was divided into seven dimensions: the identification of relevant information in the given clinical scenario, history taking, physical examination, medical test ordering, diagnostic hypothesis setting, binary analysis fulfillment, and final diagnosis setting. Sensitivity (percentage of relevant information found) and precision (percentage of correct actions performed) metrics were computed for each issue and combined into a harmonic mean (F1), thereby obtaining a single score evaluating the students' performance. The 7 metrics were further grouped to reflect the students' capability to collect and to analyze information to obtain an overall performance score. A methodological score was computed based on the discordance between the diagnostic pathway followed by students and the reference one previously defined by the teacher. In total, 25 students attending the fifth year of the School of Medicine at Humanitas University underwent test 1, which simulated a patient with dyspnea. Test 2 dealt with abdominal pain and was attended by 36 students on a different day. For validation, we assessed the Spearman rank correlation between the performance on these scores and the score obtained by each student in the hematology curricular examination. RESULTS: The mean overall scores were consistent between test 1 (mean 0.59, SD 0.05) and test 2 (mean 0.54, SD 0.12). For each student, the overall performance was achieved through a different contribution in collecting and analyzing information. Methodological scores highlighted discordances between the reference diagnostic pattern previously set by the teacher and the one pursued by the student. No significant correlation was found between the VPS scores and hematology examination scores. CONCLUSIONS: Different components of the students' diagnostic process may be disentangled and quantified by appropriate metrics applied to students' actions recorded while addressing a virtual case. Such an approach may help teachers provide students with individualized feedback aimed at filling competence drawbacks and methodological inconsistencies. There was no correlation between the hematology curricular examination score and any of the proposed scores as these scores address different aspects of students' medical knowledge.