Browsing by Author "Gonzalez, Filipe"
Now showing 1 - 5 of 5
Results Per Page
Sort Options
- AI for the hemodynamic assessment of critically ill and surgical patients: focus on clinical applicationsPublication . Michard, Frederic; Mulder, Marijn P.; Gonzalez, Filipe; Sanfilippo, FilippoSeveral artificial intelligence (AI)-driven tools have emerged for the hemodynamic evaluation of critically ill and surgical patients. This article provides an overview of current developments and potential clinical applications of machine learning (ML) for blood pressure measurements, hypotension prediction, hemodynamic profiling, and echocardiography. ML algorithms have shown promise in enabling continuous, non-invasive blood pressure monitoring by analyzing pulse oximetry waveforms, though these methods require periodic calibration with traditional oscillometric brachial cuffs. Additionally, a variety of ML models have been trained to forecast impending hypotension. However, clinical research indicates that these algorithms often primarily rely on mean arterial pressure, leading to questions about their added predictive value. The issue of false-positive alerts is also significant and can result in unwarranted clinical interventions. In terms of hemodynamic profiling, ML algorithms have been proposed to automatically classify patients into specific hemodynamic endotypes. However, current evidence suggests these models tend to replicate conventional hemodynamic profiles found in medical textbooks or depicted on advanced hemodynamic monitors. This raises questions about their practical clinical utility, especially given occasional discrepancies that could impact treatment decisions. Point-of-care ultrasound (POCUS) has gained traction for evaluating cardiac function in patients experiencing circulatory shock. ML algorithms now embedded in some POCUS systems can assist by recognizing ultrasound images, guiding users for optimal imaging, automating and reducing the variability of key echocardiographic measurements. These capabilities are especially beneficial for novice operators, potentially enhancing accuracy and confidence in clinical decision-making. In conclusion, while several AI-based technologies show promise for refining hemodynamic assessment in both critically ill and surgical patients, their clinical value varies. Comprehensive validation studies and real-world testing are essential to identify which innovations will genuinely contribute to improving the quality of care.
- Automation of sub-aortic velocity time integral measurements by transthoracic echocardiography: clinical evaluation of an artificial intelligence-enabled tool in critically ill patientsPublication . Gonzalez, Filipe; Varudo, Rita; Leote, João; Martins, Cristina; Bacariza, Jacobo; Fernandes, Antero; Michard, FredericPoint-of-care ultrasound techniques are increasingly used for the bedside assessment of cardiac function and haemodynamics in critically ill patients. The sub-aortic or left ventricular outflow tract velocity time integral (VTI) can be measured using pulsed-Doppler ultrasonography from a transthoracic apical 5-chamber view. Quantifying VTI is useful to discriminate between vasoplegic states (hypotension with normal/high VTI) and low flow states (low VTI). Measuring VTI is also useful to predict fluid responsiveness, either by quantifying the respiratory swings in VTI when patients are mechanically ventilated, or by quantifying VTI changes during a passive leg raising manoeuvre or a fluid challenge.
- Diastology in the intensive care unit: challenges for the assessment and future directionsPublication . Gonzalez, Filipe; Santonocito, Cristina; Maybauer, Marc O.; Lopes, Luís Rocha; Almeida, Ana G.; Sanfilippo, FilippoMyocardial dysfunction is common in patients admitted to the intensive care unit (ICU). Septic disease frequently results in cardiac dysfunction, and sepsis represents the most common cause of admission and death in the ICU. The association between left ventricular (LV) systolic dysfunction and mortality is not clear for critically ill patients. Conversely, LV diastolic dysfunction (DD) seems increasingly recognized as a factor associated with poor outcomes, not only in sepsis but also more generally in critically ill patients. Despite recent attempts to simplify the diagnosis and grading of DD, this remains relatively complex, with the need to use several echocardiographic parameters. Furthermore, the current guidelines have several intrinsic limitations when applied to the ICU setting. In this manuscript, we discuss the challenges in DD classification when applied to critically ill patients, the importance of left atrial pressure estimates for the management of patients in ICU, and whether the study of cardiac dysfunction spectrum during critical illness may benefit from the integration of left ventricular and left atrial strain data to improve diagnostic accuracy and implications for the treatment and prognosis.
- Machine learning for the real-time assessment of left ventricular ejection fraction in critically ill patients: a bedside evaluation by novices and experts in echocardiographyPublication . Varudo, Rita; Gonzalez, Filipe; Leote, João; Martins, Cristina; Bacariza, Jacobo; Fernandes, Antero; Michard, FredericBackground: Machine learning algorithms have recently been developed to enable the automatic and real-time echocardiographic assessment of left ventricular ejection fraction (LVEF) and have not been evaluated in critically ill patients. Methods: Real-time LVEF was prospectively measured in 95 ICU patients with a machine learning algorithm installed on a cart-based ultrasound system. Real-time measurements taken by novices (LVEFNov) and by experts (LVEFExp) were compared with LVEF reference measurements (LVEFRef) taken manually by echo experts. Results: LVEFRef ranged from 26 to 80% (mean 54 ± 12%), and the reproducibility of measurements was 9 ± 6%. Thirty patients (32%) had a LVEFRef < 50% (left ventricular systolic dysfunction). Real-time LVEFExp and LVEFNov measurements ranged from 31 to 68% (mean 54 ± 10%) and from 28 to 70% (mean 54 ± 9%), respectively. The reproducibility of measurements was comparable for LVEFExp (5 ± 4%) and for LVEFNov (6 ± 5%) and significantly better than for reference measurements (p < 0.001). We observed a strong relationship between LVEFRef and both real-time LVEFExp (r = 0.86, p < 0.001) and LVEFNov (r = 0.81, p < 0.001). The average difference (bias) between real time and reference measurements was 0 ± 6% for LVEFExp and 0 ± 7% for LVEFNov. The sensitivity to detect systolic dysfunction was 70% for real-time LVEFExp and 73% for LVEFNov. The specificity to detect systolic dysfunction was 98% both for LVEFExp and LVEFNov. Conclusion: Machine learning-enabled real-time measurements of LVEF were strongly correlated with manual measurements obtained by experts. The accuracy of real-time LVEF measurements was excellent, and the precision was fair. The reproducibility of LVEF measurements was better with the machine learning system. The specificity to detect left ventricular dysfunction was excellent both for experts and for novices, whereas the sensitivity could be improved.
- Underuse of reperfusion therapy with systemic thrombolysis in high-risk acute pulmonary embolism in a Portuguese centerPublication . Martinho, Mariana; Calé, Rita; Grade Santos, João; Rita Pereira, Ana; Alegria, Sofia; Ferreira, Filipa; José Loureiro, Maria; Judas, Tiago; Ferreira, Melanie; Gomes, Ana; Morgado, Gonçalo; Martins, Cristina; Gonzalez, Filipe; Lohmann, Corinna; Delerue, Francisca; Pereira, HélderIntroduction: Reperfusion therapy is generally recommended in acute high-risk pulmonary embolism (HR-PE), but several population-based studies report that it is underused. Data on epidemiology, management and outcomes of HR-PE in Portugal are scarce. Objective: To determine the reperfusion rate in HR-PE patients, the reasons for non-reperfusion, and how it influences outcomes. Methods: In this retrospective cohort study of consecutive HR-PE patients admitted to a thromboembolic disease referral center between 2008 and 2018, independent predictors for non-reperfusion were assessed by multivariate logistic regression. PE-related mortality and long-term MACE (cardiovascular mortality, PE recurrence and chronic thromboembolic disease) were calculated according to the Kaplan-Meier method. Differences stratified by reperfusion were assessed using the log-rank test. Results: Of 1955 acute PE patients, 3.8% presented with hemodynamic instability. The overall reperfusion rate was 50%: 35 patients underwent systemic thrombolysis, one received first-line percutaneous embolectomy and one rescue endovascular treatment. Independent predictors of non-reperfusion were: age, with >75 years representing 12 times the risk of non-treatment (OR 11.9, 95% CI 2.7-52.3, p=0.001); absolute contraindication for thrombolysis (31.1%), with recent major surgery and central nervous system disease as the most common reasons (OR 16.7, 95% CI 3.2-87.0, p<0.001); and being hospitalized (OR 7.7, 95% CI 1.4-42.9, p=0.020). At a mean follow-up of 2.5±3.3 years, the survival rate was 33.8%. Although not reaching statistical significance for hospital mortality, mortality in the reperfusion group was significantly lower at 30 days, 12 months and during follow-up (relative risk reduction of death of 64% at 12 months, p=0.013). Similar results were found for MACE. Conclusions: In this population, the recommended reperfusion therapy was performed in only 50% of patients, with advanced age and absolute contraindications to fibrinolysis being the main predictors of non-reperfusion. In this study, thrombolysis underuse was associated with a significant increase in short- and long-term mortality and events.