MARCO ANTONIO GUTIERREZ
Projetos de Pesquisa
Unidades Organizacionais
Instituto do Coração, Hospital das Clínicas, Faculdade de Medicina
LIM/65, Hospital das Clínicas, Faculdade de Medicina - Líder
LIM/65, Hospital das Clínicas, Faculdade de Medicina - Líder
15 resultados
Resultados de Busca
Agora exibindo 1 - 10 de 15
- Detecting Vascular Bifurcation in IVOCT Images Using Convolutional Neural Networks With Transfer Learning(2019) MIYAGAWA, Makoto; COSTA, Marly Guimaraes Fernandes; GUTIERREZ, Marco Antonio; COSTA, Joao Pedro Guimaraes Fernandes; COSTA FILHO, Cicero Ferreira FernandesOptical coherence tomography (OCT) technology enables experts to analyze coronary lesions from high-resolution intravascular images. Studies have shown a relationship between vascular bifurcation and a higher occurrence of wall thickening and lesions in these areas. Some level of automation could benefit experts since the visual analysis of pullback frames is a laborious and time-consuming task. Although convolutional neural networks (CNNs) have shown promising results in classifying medical images, in this paper, we found no studies using CNNs in IVOCT images to classify the vascular bifurcation. In this paper, we evaluated four different CNN architectures in the bifurcation classification task trained with the IVOCT images from nine pullbacks from nine different patients. We used data augmentation to balance the dataset, due to the small number of bifurcation-labeled frames, and also applied transfer learning methods to incorporate the knowledge from a lumen segmentation task into some of the evaluated networks. Our classification outperforms other works in this literature, presenting AUC = 99.72%, obtained by a CNN with transferred knowledge.
- Siamese Convolutional Neural Network for Heartbeat Classification Using Limited 12-Lead ECG Datasets(2023) VASCONCELLOS, Eduardo M. M.; FERREIRA, Bruno Georgevich; LEANDRO, Jorge S.; NETO, Baldoino F. S.; CORDEIRO, Filipe Rolim; CESTARI, Idagene A.; GUTIERREZ, Marco A.; SOBRINHO, Alvaro; CORDEIRO, Thiago D.The Electrocardiogram (ECG) is a low-cost exam commonly used to diagnose abnormalities in the cardiac cycle. Over the years, the scientific community has investigated the automatic classification of ECG signals driven by advanced Machine Learning (ML) techniques. Despite recent scientific advances, annotating large and diverse datasets to support the training of ML techniques is still very time-consuming and error-prone. Indeed, ML techniques whose training does not require extensive and well-annotated datasets are becoming even more prominent. Therefore, it is possible to correctly identify and classify abnormalities in the cardiac cycle (e.g., rare cardiologic disturbs) using limited data available in ECG datasets. However, the classification of heartbeats from digital tracings of ECG signals containing 12 leads from imbalanced datasets is challenging due to many existing heart diseases. This study investigates the few-shot learning paradigm based on Siamese Convolutional Neural Networks (SCNN), popular in imaging classification problems, to classify 12-Lead ECG heartbeats using a few training samples with supervised information. The proposed SCNN model presented an accuracy of up to 95% in a public dataset based on the hold-out validation method, implemented for different combinations of similarity and loss functions. Besides, using the 7-fold cross-validation method, the model presented a mean area under the curve of 89%. We also compared the class-by-class classification results with those of similar methods available in the literature, obtaining the same or better results based on performance metrics such as accuracy, precision, recall, and specificity.
- Computational Analysis of Fluid Dynamics in the Transcatheter Aortic Valve Replacement(2020) GUTIERREZ, Marco A.
article Telecardiology guideline in Patient Care with Acute Coronary Syndrome and Other Respiratory Diseases(2015) OLIVEIRA JUNIOR, Mucio Tavares de; CANESIN, Manoel Fernandes; MARCOLINO, Milena Soriano; RIBEIRO, Antonio Luiz Pinho; CARVALHO, Antonio Carlos de Camargo; REDDY, Shankar; SANTOS, Adson Roberto Franca dos; FERNANDES, Alfredo Manoel da Silva; AMARAL, Amaury Zatorre; REZENDE, Ana Carolina de; NECHAR JUNIOR, Antonio; NASCIMENTO, Bruno Ramos do; PASTORE, Carlos Alberto; WEN, Chao Lung; GUALANDRO, Danielle Menosi; NAPOLI, Domingos Guilherme; FRANCA, Francisco Faustino A. C.; FEITOSA-FILHO, Gilson Soares; SAAD, Jamil Abdalla; PILLI, Jeanne; PAULA, Leonardo Jorge Cordeiro de; LODI-JUNQUEIRA, Lucas; CESAR, Luis Antonio Machado; BODANESE, Luiz Carlos; GUTIERREZ, Marco Antonio; ALKMIM, Maria Beatriz Moreira; NUNES, Mauricio Batista; MEDEIROS, Orlando Otavio de; MORENO, Ramon Alfredo; GUNDIM, Rosangela Simoes; MONTENEGRO, Sergio Tavares; NAZIMA, Willyan Issamu- Cerebral blood flow changes during intermittent acute hypoxia in patients with heart failure(2018) MANSUR, Antonio P.; ALVARENGA, Glaura Souza; KOPEL, Liliane; GUTIERREZ, Marco Antonio; CONSOLIM-COLOMBO, Fernanda Marciano; HAJJAR, Ludhmila Abrahao; LAGE, Silvia GelasObjective: Heart failure (HF) is associated with intermittent hypoxia, and the effects of this hypoxia on the cardiovascular system are not well understood. This study was performed to compare the effects of acute hypoxia (10% oxygen) between patients with and without HF. Methods: Fourteen patients with chronic HF and 17 matched control subjects were enrolled. Carotid artery changes were examined during the first period of hypoxia, and brachial artery changes were examined during the second period of hypoxia. Data were collected at baseline and after 2 and 4 minutes of hypoxia. Norepinephrine, epinephrine, dopamine, and renin were measured at baseline and after 4 minutes hypoxia. Results: The carotid blood flow, carotid systolic diameter, and carotid diastolic diameter increased and the carotid resistance decreased in patients with HF. Hypoxia did not change the carotid compliance, distensibility, brachial artery blood flow and diameter, or concentrations of sympathomimetic amines in patients with HF, but hypoxia increased the norepinephrine level in the control group. Hypoxia increased minute ventilation and decreased the oxygen saturation and end-tidal carbon dioxide concentration in both groups. Conclusion: Hypoxia-induced changes in the carotid artery suggest an intensification of compensatory mechanisms for preservation of cerebral blood flow in patients with HF.
- Chronic lung lesions in COVID-19 survivors: predictive clinical model(2022) CARVALHO, Carlos Roberto Ribeiro; CHATE, Rodrigo Caruso; SAWAMURA, Marcio Valente Yamada; GARCIA, Michelle Louvaes; LAMAS, Celina Almeida; CARDENAS, Diego Armando Cardona; LIMA, Daniel Mario; SCUDELLER, Paula Gobi; SALGE, Joao Marcos; NOMURA, Cesar Higa; GUTIERREZ, Marco AntonioObjective This study aimed to propose a simple, accessible and low-cost predictive clinical model to detect lung lesions due to COVID-19 infection. Design This prospective cohort study included COVID-19 survivors hospitalised between 30 March 2020 and 31 August 2020 followed-up 6 months after hospital discharge. The pulmonary function was assessed using the modified Medical Research Council (mMRC) dyspnoea scale, oximetry (SpO(2)), spirometry (forced vital capacity (FVC)) and chest X-ray (CXR) during an in-person consultation. Patients with abnormalities in at least one of these parameters underwent chest CT. mMRC scale, SpO(2), FVC and CXR findings were used to build a machine learning model for lung lesion detection on CT. Setting A tertiary hospital in Sao Paulo, Brazil. Participants 749 eligible RT-PCR-confirmed SARS-CoV-2-infected patients aged >= 18 years. Primary outcome measure A predictive clinical model for lung lesion detection on chest CT. Results There were 470 patients (63%) that had at least one sign of pulmonary involvement and were eligible for CT. Almost half of them (48%) had significant pulmonary abnormalities, including ground-glass opacities, parenchymal bands, reticulation, traction bronchiectasis and architectural distortion. The machine learning model, including the results of 257 patients with complete data on mMRC, SpO(2), FVC, CXR and CT, accurately detected pulmonary lesions by the joint data of CXR, mMRC scale, SpO(2) and FVC (sensitivity, 0.85 +/- 0.08; specificity, 0.70 +/- 0.06; F1-score, 0.79 +/- 0.06 and area under the curve, 0.80 +/- 0.07). Conclusion A predictive clinical model based on CXR, mMRC, oximetry and spirometry data can accurately screen patients with lung lesions after SARS-CoV-2 infection. Given that these examinations are highly accessible and low cost, this protocol can be automated and implemented in different countries for early detection of COVID-19 sequelae.
- Use of telemedicine to combat the COVID-19 pandemic in Brazil(2020) CARVALHO, Carlos Roberto Ribeiro; SCUDELLER, Paula Gobi; RABELLO, Guilherme; GUTIERREZ, Marco Antonio; JATENE, Fabio Biscegli
- Implementation of Tele-ICU during the COVID-19 pandemic(2021) MACEDO, Bruno Rocha de; GARCIA, Marcos Vinicius Fernandes; GARCIA, Michelle Louvaes; VOLPE, Marcia; SOUSA, Mayson Laercio de Araujo; AMARAL, Talita Freitas; GUTIERREZ, Marco Antonio; BARBOSA, Antonio Pires; SCUDELLER, Paula Gobi; CARUSO, Pedro; CARVALHO, Carlos Roberto RibeiroObjective: To describe the implementation of a Tele-ICU program during the COVID-19 pandemic, as well as to describe and analyze the results of the first four months of operation of the program. Methods: This was a descriptive observational study of the implementation of a Tele-ICU program, followed by a retrospective analysis of clinical data of patients with COVID-19 admitted to ICUs between April and July of 2020. Results: The Tele-ICU program was implemented over a four-week period and proved to be feasible during the pandemic. Participants were trained remotely, and the program had an evidence-based design, the objective being to standardize care for patients with COVID-19. More than 100,000 views were recorded on the free online platforms and the mobile application. During the study period, the cases of 326 patients with COVID-19 were evaluated through the program. The median age was 60 years (IQR, 49-68 years). There was a predominance of males (56%). There was also a high prevalence of hypertension (49.1%) and diabetes mellitus (38.4%). At ICU admission, 83.7% of patients were on invasive mechanical ventilation, with a median PaO2/FiO(2) ratio < 150. It was possible to use lung-protective ventilation in 75% of the patients. Overall, in-hospital mortality was 68%, and ICU mortality was 65%. Conclusions: Our Tele-ICU program provided multidisciplinary training to health care professionals and clinical follow-up for hundreds of critically ill patients. This public health care network initiative was unprecedented and proved to be feasible during the COVID-19 pandemic, encouraging the creation of similar projects that combine evidence-based practices, training, and Tele-ICU.
- Long-term respiratory follow-up of ICU hospitalized COVID-19 patients: Prospective cohort study(2023) CARVALHO, Carlos Roberto Ribeiro; LAMAS, Celina Almeida; CHATE, Rodrigo Caruso; SALGE, Joao Marcos; SAWAMURA, Marcio Valente Yamada; ALBUQUERQUE, Andre L. P. de; JR, Carlos Toufen; LIMA, Daniel Mario; GARCIA, Michelle Louvaes; SCUDELLER, Paula Gobi; NOMURA, Cesar Higa; GUTIERREZ, Marco Antonio; BALDI, Bruno GuedesBackgroundCoronavirus disease (COVID-19) survivors exhibit multisystemic alterations after hospitalization. Little is known about long-term imaging and pulmonary function of hospitalized patients intensive care unit (ICU) who survive COVID-19. We aimed to investigate long-term consequences of COVID-19 on the respiratory system of patients discharged from hospital ICU and identify risk factors associated with chest computed tomography (CT) lesion severity. MethodsA prospective cohort study of COVID-19 patients admitted to a tertiary hospital ICU in Brazil (March-August/2020), and followed-up six-twelve months after hospital admission. Initial assessment included: modified Medical Research Council dyspnea scale, SpO(2) evaluation, forced vital capacity, and chest X-Ray. Patients with alterations in at least one of these examinations were eligible for CT and pulmonary function tests (PFTs) approximately 16 months after hospital admission. Primary outcome: CT lesion severity (fibrotic-like or non-fibrotic-like). Baseline clinical variables were used to build a machine learning model (ML) to predict the severity of CT lesion. ResultsIn total, 326 patients (72%) were eligible for CT and PFTs. COVID-19 CT lesions were identified in 81.8% of patients, and half of them showed mild restrictive lung impairment and impaired lung diffusion capacity. Patients with COVID-19 CT findings were stratified into two categories of lesion severity: non-fibrotic-like (50.8%-ground-glass opacities/reticulations) and fibrotic-like (49.2%-traction bronchiectasis/architectural distortion). No association between CT feature severity and altered lung diffusion or functional restrictive/obstructive patterns was found. The ML detected that male sex, ICU and invasive mechanic ventilation (IMV) period, tracheostomy and vasoactive drug need during hospitalization were predictors of CT lesion severity(sensitivity,0.78 +/- 0.02;specificity,0.79 +/- 0.01;F1-score,0.78 +/- 0.02;positive predictive rate,0.78 +/- 0.02; accuracy,0.78 +/- 0.02; and area under the curve,0.83 +/- 0.01). ConclusionICU hospitalization due to COVID-19 led to respiratory system alterations six-twelve months after hospital admission. Male sex and critical disease acute phase, characterized by a longer ICU and IMV period, and need for tracheostomy and vasoactive drugs, were risk factors for severe CT lesions six-twelve months after hospital admission.
- Blood Pressure Estimation From Photoplethysmography by Considering Intra- and Inter-Subject Variabilities: Guidelines for a Fair Assessment(2023) COSTA, Thiago Bulhoes Da Silva; DIAS, Felipe Meneguitti; CARDENAS, Diego Armando Cardona; TOLEDO, Marcelo Arruda Fiuza De; LIMA, Daniel Mario De; KRIEGER, Jose Eduardo; GUTIERREZ, Marco AntonioCardiovascular diseases are the leading causes of death, and blood pressure (BP) monitoring is essential for prevention, diagnosis, assessment, and treatment. Photoplethysmography (PPG) is a low-cost opto-electronic technique for BP measurement that allows the acquisition of a modulated light signal highly correlated with BP. There are several reports of methods to estimate BP from PPG with impressive results; in this study, we demonstrate that the previous results are excessively optimistic because of their train/test split configuration. To manage this limitation, we considered intra- and inter-subject data arrangements and demonstrated how they affect the results of feature-based BP estimation algorithms (i.e., XGBoost, LightGBM, and CatBoost) and signal-based algorithms (i.e., Residual U-Net, ResNet-18, and ResNet-LSTM). Inter-subject configuration performance is inferior to intra-subject configuration performance, regardless of the model. We also showed that, using only demographic attributes (i.e., age, sex, weight, and subject index number), a regression model achieved results comparable to those obtained in an intra-subject scenario.Although limited to a public clinical database, our findings suggest that algorithms that use an intra-subject setting without a calibration strategy may be learning to identify patients and not predict BP.