Claudia Di Lorenzo Oliveira

BACKGROUND: Chagas disease (ChD), a neglected cardiovascular condition, affects 7.5 to 10.5 million people worldwide. Opportunistic screening during routine electrocardiography may allow earlier detection of unrecognized infections, enabling timely antiparasitic therapy or optimized cardiac care. This study prospectively evaluated the feasibility and diagnostic accuracy of an artificial intelligence (AI)–enabled ECG model enriched with 3 epidemiological questions (AI-ECG-EPI model) as an opportunistic screening for ChD in endemic regions of Brazil. METHODS: We conducted a prospective study embedded in the Telehealth Network of Minas Gerais, Brazil, across 107 municipalities in hyperendemic and endemic regions. From October 2023 to September 2024, adults undergoing routine tele-ECGs were eligible. The primary exposure was the output of the AI-ECG-EPI model: all positive individuals and a 3:1 sample of negative individuals were invited for serology (index test–dependent sampling), the reference standard. Weighted analyses (inverse probability) accounted for this sampling design. Diagnostic metrics included sensitivity, specificity, predictive values, likelihood ratios, area under the receiver operating characteristics curve, and area under the precision recall curve. The AI-ECG-EPI model was compared with both an epidemiological questions model and an AI ECG–only model. RESULTS: Among 75 779 eligible ECGs, the AI-ECG-EPI model was available for 59 154 individuals; 6812 (11.5%) screened positive. A total of 7804 selected were invited for serology, and 3509 completed testing (2517 positive; 992 negative). ChD was confirmed in 36.8% of positive and 9.3% of negative individuals. The weighted prevalence of ChD was 12.2%. Sensitivity was 34.3%, specificity 91.6%, harmonic mean of precision and recall score 0.52, and diagnostic odds ratio 5.69. The AI-ECG-EPI model showed higher discrimination than both the epidemiological questions model and AI ECG–only model (area under the receiver operating characteristics curve, 77.6% versus 70.0% and 64.5%, respectively; P <0.001). Precision recall curves showed higher precision across most recall levels. Exploratory analysis suggested improved risk reclassification (net reclassification improvement, 0.503). Performance was higher in women and in individuals with Chagas cardiomyopathy. CONCLUSIONS: In this community-based study, an AI-ECG-EPI model integrated into a public telecardiology network enabled feasible opportunistic screening for ChD in primary care. It showed acceptable diagnostic accuracy, with higher discrimination than epidemiological questions and AI ECG–only models, and identified many previously undiagnosed patients in endemic regions, supporting its potential role in scalable screening strategies. REGISTRATION: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT02646943.

Background: Chagas disease (CD) is a neglected endemic infectious disease. Primary health care (PHC) is responsible for delivering care to patients with CD in an integrated manner with other levels of the health system. We aimed to describe the use of implementation science (IS) as a tool for developing a care pathway for patients with CD from the perspective of the Brazilian Chronic Care Model (BCCM). Methods: This study was conducted in the large municipality of Minas Gerais, Brazil. A diagnostic phase was conducted to identify barriers and facilitators related to CD care. The findings from this stage were analyzed using IS frameworks. Subsequently, the BCCM was adapted as a logical model to guide the creation of a care pathway. Health managers and professionals from the municipalities were trained to implement the proposed actions. After implementation, process and outcome indicators were monitored over a 12-month period. Results: IS proved to be an effective strategy for applying BCCM to patients with CD in PHC settings. A total of 267 health professionals were trained. After the intervention, the following indicators increased: risk factor screening, serological testing, diagnosis, and antiparasitic treatment for CD. Conclusion: The development of a CD care pathway using IS tools integrated with the BCCM enabled the incorporation of processes into PHC and suggested that this model may be replicated in other contexts as well as for other chronic conditions requiring longitudinal care.

Objetivo Analisar a implementação do Programa de Promoção da Alimentação Adequada e Saudável em um município de pequeno porte de Minas Gerais, a partir da perspectiva dos profissionais de saúde.

Objective To analyze the implementation of the Adequate and Healthy Eating Promotion Program in a small municipality in Minas Gerais, from the perspective of health professionals.

Chagas disease (ChD) remains a significant public health concern in the Americas, with challenges to accurately assessing its mortality burden due to under-reporting and misclassification. This study aimed to analyze mortality patterns of two cohorts of individuals with ChD—one comprising asymptomatic individuals with positive serology (REDS) and another with patients showing Chagas cardiomyopathy (SaMi-Trop)—to propose a method for estimating the potential under-registration of Chagas-related deaths and to find the factors influencing the identification of ChD as the underlying cause of death. We carried out a retrospective analysis of mortality data from these cohorts together with data on the Brazilian Mortality Information System. Causes of death were classified according to ICD-10 codes, and an expert review was used to find possible Chagas-related deaths. Logistic regression was used to explore predictors of ChD identification considering demographic and clinical variables. Of 2,488 patients, 381 died, 28.9% attributed to ChD, predominantly chronic ChD with cardiac involvement (B57.2). Using our method, we estimated a 53.8% potential under-registration rate for possible Chagas deaths. Males were negatively associated with Chagas disease identification, with an odds ratio of 0.52 (95%CI 0.24–1.1). No other significant associations were found, and the overall significance of the model was low. Our findings provide a potential measurement of under-registration, indicating that it may be substantial. These results underscore the need for improved identification and accurate reporting on death certificates. Strengthening the quality of mortality data is essential to understand Chagas-related mortality and guide public health strategies to reduce its impact.

Background: Worldwide, it is estimated that over 6 million people are infected with Chagas disease (ChD). It is a neglected disease that can lead to severe heart conditions in its chronic phase. While early treatment can avoid complications, the early-stage detection rate is low. We explore the use of deep neural networks to detect ChD from electrocardiograms (ECGs) to aid in the early detection of the disease. Methods: We employ a convolutional neural network model that uses 12-lead ECG data to compute the probability of a ChD diagnosis. Our model is developed using two datasets which jointly comprise over two million entries from Brazilian patients: The SaMi-Trop study focusing on ChD patients, enriched with data from the CODE study from the general population. The model’s performance is evaluated on two external datasets: the REDS-II, a study focused on ChD with 631 patients, and the ELSA-Brasil study, with 13,739 civil servant patients. Findings: Evaluating our model, we obtain an AUC-ROC of 0.80 (CI 95% 0.79-0.82) for the validation set (samples from CODE and SaMi-Trop), and in external validation datasets: 0.68 (CI 95% 0.63-0.71) for REDS-II and 0.59 (CI 95% 0.56-0.63) for ELSA-Brasil. In the latter, we report a sensitivity of 0.52 (CI 95% 0.47-0.57) and 0.36 (CI 95% 0.30-0.42) and a specificity of 0.77 (CI 95% 0.72-0.81) and 0.76 (CI 95% 0.75-0.77), respectively. Additionally, when considering only patients with Chagas cardiomyopathy as positive, the model achieved an AUC-ROC of 0.82 (CI 95% 0.77-0.86) for REDS-II and 0.77 (CI 95% 0.68-0.85) for ELSA-Brasil. Interpretation: The neural network detects chronic Chagas cardiomyopathy (CCC) from ECG—with weaker performance for early-stage cases. Future work should focus on curating large higher-quality datasets. The CODE dataset, our largest development dataset includes self-reported and therefore less reliable labels, limiting performance for non-CCC patients. Our findings can improve ChD detection and treatment, particularly in high-prevalence areas.

Background: The laboratory tests most used by blood banks to diagnose anemia are the hemoglobin (Hb) and microhematocrit (Hct) tests, measured from capillary samples. Objective: To analyze the two capillary screening methods for pre-donation anemia by comparing their agreement in diagnosing anemia. Method: A cross-sectional study in a population of 15,521 blood donation candidates for whom information was available on Hb and Hct, performed from capillary blood samples. Hb was determined using the HemoCue^® test and Hct by the centrifugation method. The Kappa coefficient was calculated to assess the agreement between the methods. Pearson’s correlation tests and gender-adjusted linear regression were used to assess the change in the response variable (Hb) as a function of the explanatory variable (Hct). Results: The majority of the study population were men (70.4%), aged between 18 and 44 years (72.1%), who declared themselves white or mixed skin color (85.6%), and had undergone at least 11 years of complete education (72.4%). The Kappa coefficient found was 92.7 and 99.2 for women and men, respectively. Pearson’s correlation showed a correlation coefficient of 0.98 and the linear regression graph showed an adequate relationship between the tests with R² = 0.97. Conclusions: Comparing the Hb and Hct capillary tests, it was found that Hct can be safely used to screen for anemia in pre-blood donation.

Background: Worldwide it is estimated that more than 6 million people are infected with Chagas disease (ChD). It is considered one of the most important neglected diseases and, when it reaches its chronic phase, the infected person often develops serious heart conditions. While early treatment can avoid complications, the condition is often not detected during its early stages. We investigate whether a deep neural network can detect ChD from electrocardiogram (ECG) tracings. The ECG is inexpensive and it is often performed during routine visits. Being able to evaluate ChD from this exam can help detect potentially hidden cases in an early stage. Methods: We use a convolutional neural network model, which takes the 12-lead ECG as input and outputs a scalar number associated with the probability of a Chagas diagnosis. To develop the model, we use two data sets, which jointly consist of over two million entries from Brazilian patients, compiled by the Telehealth Network of Minas Gerais within the SaMi-Trop (São Paulo-Minas Gerais Tropical Medicine Research Center) study focused on ChD patients and enriched with the CODE (Clinical Outcomes in Digital Electrocardiology) study focused on a general population. The performance is evaluated on two external data sets of 631 and 13,739 patients, collected in the scope of the REDS-II (Retrovirus Epidemiology Donor Study-II) study and of the ELSA-Brasil (Brazilian Longitudinal Study of Adult Health) study. The first study focuses on ChD patients and the second data set originates from civil servants from five universities and one research institute. Findings: Evaluating our model, we obtain an AUC-ROC value of 0.80 (CI 95% 0.79-0.82) for the validation data set (with samples from CODE and SaMi-Trop), and in external validation datasets: 0.68 (CI 95% 0.63-0.71) for REDS-II and 0.59 (CI 95% 0.56-0.63) for ELSA-Brasil. In these external validation datasets, we report a sensitivity of 0.52 (CI 95% 0.47-0.57) and 0.36 (CI 95% 0.30-0.42) and a specificity of 0.77 (CI 95% 0.72-0.81) and 0.76 (CI 95% 0.75-0.77), respectively, in REDS-II and ELSA-Brasil. We also evaluated the model for considering only patients with Chagas cardiomyopathy as positive. In this case, the model attains an AUC-ROC of 0.82 (CI 95% 0.77-0.86) for REDS-II and 0.77 (CI 95% 0.68-0.85) for ELSA-Brasil. Interpretation: The results indicate that the neural network can detect patients who developed chronic Chagas cardiomyopathy (CCC) from the ECG and – with weaker performance – detect patients before the CCC stage. Future work should focus on curating large and better datasets for developing such models. The CODE is the largest dataset available to us, and their labels are self-reported and less reliable than our other data sets, i.e. REDS-II and ELSA-Brasil. This, we believe, limits our model performance in the case of non-CCC patients. We are positive that our findings constitute the first step towards building tools for more efficient detection and treatment of ChD, especially in high-prevalent regions. Funding: This research is financially supported by the Swedish Foundation for Strategic Research (SSF) via the project ASSEMBLE (Contract number: RIT 15-0012), by the Wallenberg AI, Autonomous Systems and Software Program (WASP) funded by Knut and Alice Wallenberg Foundation, by Kjell och Märta Beijer Foundation , by the Brazilian Agencies CNPq, CAPES, and FAPEMIG, and by projects IATS, and CIIA-Saúde. The ELSA-Brasil study was supported by the Brazilian Ministries of Health and of Science and Technology (grants 01060010.00RS, 01060212.00BA, 01060300.00ES, 01060278.00MG, 01060115.00SP, and 01060071.00RJ). The SaMi-Trop and REDS-II cohort studies are supported by the National Institutes of Health (P50 AI098461-02, U19AI098461-06, 1U01AI168383-01). LG, SMB, ECS and ALPR receive unrestricted research scholarships from CNPq; ALPR received a Google Latin America Research Award scholarship. The funders had no role in the study design; collection, analysis, and interpretation of data; writing of the report; or decision to submit the paper for publication. Research in context: Evidence before this study: Chagas disease (ChD) is a neglected tropical disease, and the diagnosis relies on blood testing of patients from endemic areas. However, there is no clear recommendation on selecting patients for serological diagnosis in those living in endemic regions. Since most of the patients with Chronic ChD are asymptomatic or oligosymptomatic, the diagnostic rates are low, preventing patients from receiving adequate treatment. The Electro-cardiogram (ECG) is a widely available, low-cost exam, often available in primary care settings in endemic countries. Artificial intelligence (AI) algorithms on ECG tracings have allowed the detection of hidden conditions, such as cardiomyopathies and left ventricular systolic dysfunction. Added value of this study: To the best of our knowledge, this is the first study that presents an AI model for the automatic detection of ChD from the ECG. As part of the model development, we utilise established large cohorts of patients from the relevant population of all-comers in affected regions in the state of Minas Gerais, Brazil. We evaluate the model on data sets with high-quality ground truth labels obtained from the patients’ serological status. Our model has moderate diagnostic performance in recognition of ChD and better accuracy in detecting Chagas cardiomyopathy. Implications of all the available evidence: Our findings demonstrate a promising AI-ECG-based model capacity for discriminating patients with chronic Chagas cardiomyopathy (CCC). However, detecting ChD patients without CCC is still insufficient, and further developments that lead to higher performance are needed. We believe this can be achieved with the addition of epidemiological questions, and that our model can be a useful tool in helping pre-selecting patients for further testing in order to determine the infection with ChD. The use of AI-ECG-based strategies for recognizing CCC patients deserves to be tested in the clinical setting.

Resumo O objetivo deste artigo é analisar a implantação do Programa de Promoção da Alimentação Adequada e Saudável na Atenção Primária à Saúde em municípios de Minas Gerais. Pesquisa de Métodos Mistos com Estratégia Incorporada Concomitante de dados. Na abordagem quantitativa utilizou-se pesquisa avaliativa na perspectiva da análise de implantação. Na abordagem qualitativa realizou-se estudo exploratório-descritivo utilizando a análise de conteúdo. Os dados qualitativos e quantitativos foram incorporados para análise a partir de sua imbricação. O Grau de Implantação do programa foi 92,5% representando uma implantação adequada. Na imbricação dos dados, a metodologia das ações propostas no programa e a realização da atividade de Educação Permanente foram pontos fortes convergentes que emergiram. Recursos Humanos insuficientes para planejamento e execução do programa e falta de cozinha para oficina culinária foram apontados como desafios. A implantação do programa foi adequada e poderá ser estendida para outros municípios brasileiros, de forma a apoiar os profissionais a estruturar intervenções de promoção da alimentação adequada e saudável na rotina de trabalho na Atenção Primária.

Chagas disease (ChD) is a neglected tropical disease that is caused by the protozoan parasite Trypanosoma cruzi and can negatively impact quality of life (QoL). This study aimed to assess and compare QoL between individuals with and without ChD.