Time series analysis of comprehensive maternal deaths in Brazil during the COVID-19 pandemic.

The effects of the COVID-19 pandemic on comprehensive maternal deaths in Brazil have not been fully explored. Using publicly available data from the Brazilian Mortality Information (SIM) and Information System on Live Births (SINASC) databases, we used two complementary forecasting models to predict estimates of maternal mortality ratios using maternal deaths (MMR) and comprehensive maternal deaths (MMRc) in the years 2020 and 2021 based on data from 2008 to 2019. We calculated national and regional standardized mortality ratio estimates for maternal deaths (SMR) and comprehensive maternal deaths (SMRc) for 2020 and 2021. The observed MMRc in 2021 was more than double the predicted MMRc based on the Holt-Winters and autoregressive integrated moving average models (127.12 versus 60.89 and 59.12 per 100,000 live births, respectively). We found persisting sub-national variation in comprehensive maternal mortality: SMRc ranged from 1.74 (95% confidence interval [CI] 1.64, 1.86) in the Northeast to 2.70 (95% CI 2.45, 2.96) in the South in 2021. The observed national estimates for comprehensive maternal deaths in 2021 were the highest in Brazil in the past three decades. Increased resources for prenatal care, maternal health, and postpartum care may be needed to reverse the national trend in comprehensive maternal deaths.

Development of HepatIA: A computed tomography annotation platform and database for artificial intelligence training in hepatocellular carcinoma detection at a Brazilian tertiary teaching hospital.

BACKGROUND: Hepatocellular carcinoma (HCC) is a prevalent tumor with high mortality rates. Computed tomography (CT) is crucial in the non-invasive diagnosis of HCC. Recent advancements in artificial intelligence (AI) have shown significant potential in medical imaging analysis. However, developing these AI algorithms is hindered by the scarcity of comprehensive, publicly available liver imaging datasets. OBJECTIVES: This study aims to detail the tools, data organization, and database structuring used in creating HepatIA, a medical imaging annotation platform and database at a Brazilian tertiary teaching hospital. HepatIA supports liver disease AI research at the institution. MATERIAL AND METHODS: The authors collected baseline characteristics and CT scans of 656 patients from 2008 to 2021. The database, designed using PostgreSQL and implemented with Django and Vue.js, includes 692 CT volumes from a four-phase abdominal CT protocol. Radiologists made segmentation annotations using the OHIF medical image viewer, incorporating MONAI Label for pre-annotation segmentation models. The annotation process included detailed descriptions of liver morphology and nodule characteristics. RESULTS: The HepatIA database currently includes healthy individuals and those with liver diseases such as HCC and cirrhosis. The database dashboard facilitates user interaction with intuitive plots and histograms. Key patient demographics include 64% males and an average age of 56.89 years. The database supports various filters for detailed searches, enhancing research capabilities. CONCLUSION: A comprehensive data structure was successfully created and integrated with the IT systems of a teaching hospital, enabling research on deep learning algorithms applied to abdominal CT scans for investigating hepatic lesions such as HCC.

Fusion of PCA and ICA in Statistical Subset Analysis for Speech Emotion Recognition.

Speech emotion recognition is key to many fields, including human-computer interaction, healthcare, and intelligent assistance. While acoustic features extracted from human speech are essential for this task, not all of them contribute to emotion recognition effectively. Thus, reduced numbers of features are required within successful emotion recognition models. This work aimed to investigate whether splitting the features into two subsets based on their distribution and then applying commonly used feature reduction methods would impact accuracy. Filter reduction was employed using the Kruskal-Wallis test, followed by principal component analysis (PCA) and independent component analysis (ICA). A set of features was investigated to determine whether the indiscriminate use of parametric feature reduction techniques affects the accuracy of emotion recognition. For this investigation, data from three databases-Berlin EmoDB, SAVEE, and RAVDES-were organized into subsets according to their distribution in applying both PCA and ICA. The results showed a reduction from 6373 features to 170 for the Berlin EmoDB database with an accuracy of 84.3%; a final size of 130 features for SAVEE, with a corresponding accuracy of 75.4%; and 150 features for RAVDESS, with an accuracy of 59.9%.

A Combined CNN Architecture for Speech Emotion Recognition.

Emotion recognition through speech is a technique employed in various scenarios of Human-Computer Interaction (HCI). Existing approaches have achieved significant results; however, limitations persist, with the quantity and diversity of data being more notable when deep learning techniques are used. The lack of a standard in feature selection leads to continuous development and experimentation. Choosing and designing the appropriate network architecture constitutes another challenge. This study addresses the challenge of recognizing emotions in the human voice using deep learning techniques, proposing a comprehensive approach, and developing preprocessing and feature selection stages while constructing a dataset called EmoDSc as a result of combining several available databases. The synergy between spectral features and spectrogram images is investigated. Independently, the weighted accuracy obtained using only spectral features was 89%, while using only spectrogram images, the weighted accuracy reached 90%. These results, although surpassing previous research, highlight the strengths and limitations when operating in isolation. Based on this exploration, a neural network architecture composed of a CNN1D, a CNN2D, and an MLP that fuses spectral features and spectogram images is proposed. The model, supported by the unified dataset EmoDSc, demonstrates a remarkable accuracy of 96%.

Update on the Epimed Monitor Adult ICU Database: 15 years of its use in national registries, quality improvement initiatives and clinical research.

In recent decades, several databases of critically ill patients have become available in both low-, middle-, and high-income countries from all continents. These databases are also rich sources of data for the surveillance of emerging diseases, intensive care unit performance evaluation and benchmarking, quality improvement projects and clinical research. The Epimed Monitor database is turning 15 years old in 2024 and has become one of the largest of these databases. In recent years, there has been rapid geographical expansion, an increase in the number of participating intensive care units and hospitals, and the addition of several new variables and scores, allowing a more complete characterization of patients to facilitate multicenter clinical studies. As of December 2023, the database was being used regularly for 23,852 beds in 1,723 intensive care units and 763 hospitals from ten countries, totaling more than 5.6 million admissions. In addition, critical care societies have adopted the system and its database to establish national registries and international collaborations. In the present review, we provide an updated description of the database; report experiences of its use in critical care for quality improvement initiatives, national registries and clinical research; and explore other potential future perspectives and developments.

Validation of secondary data sources of the retail food environment in the capital of Uruguay, an emerging Latin American country.

Secondary data sources are frequently used for characterizing physical access to food. Although several studies have reported that they tend to show a moderate agreement with field observation in WEIRD (Western Educated Industrialized Rich and Democratic) countries, little is known about their validity in non-WEIRD countries. The aim of the present research was to assess the validity of secondary data sources of the retail food environment in Montevideo, the capital of Uruguay, an emerging Latin American country. A random sample of 106 census tracts was obtained, covering 12% (62 km2) of the city's total area. Two secondary data sources were considered: administrative records and Google Maps. An aggregate database was created by manually removing duplicates. A total of 1051 unique outlets were listed in the database within the census tracts included in the sample. Field validation was performed by six teams of two observers. A total of 1200 food outlets were identified on the ground, including 463 (38.6%) outlets not listed on any database. On the contrary, 297 outlets listed in the databases (28.3%) were not found or were closed at the time of field validation. At the aggregate level, sensitivity and concordance were moderate (0.614 and 0.487, respectively), whereas positive predictive value was substantial (0.701). However, large heterogeneity in the validity of the database across census tracts was found. Sensitivity, positive predictive value, and concordance were positively associated with the socio-economic status index of the census tract. These results suggest that secondary data sources must be used with caution, particularly for the characterization of areas with low socio-economic status.

Standardizing fingerprint minutiae: A comprehensive inventory and statistical analysis based on Brazilian data.

This study introduces a comprehensive inventory of 54 fingerprint minutiae and their variations aimed at standardizing characteristic point identification within forensic science. By analyzing a strategically sampled collection of fingerprints from the Brazilian Federal Police database, stratified by sex and geographic location, our research uncovers the complex interplay between various levels of fingerprint details (L1D, L2D, and L3D) and demographic factors such as sex and finger type. The sample encompassed the entire proposed list of minutiae, affirming the diversity and representativeness of the Brazilian populace, which had, for the first time, its minutiae frequency distributions studied. This investigation proposes a systematic approach for enhancing fingerprint identification accuracy by minimizing data categorization losses and lays the groundwork for more uniform comparative studies in the field. Our findings, derived from a review of contemporary studies and traditional identification manuals, suggest a step towards establishing a universally accepted standard for fingerprint minutiae classification.

Salvage Endovascular Thrombectomy for Splenic Vein Thrombosis After Pancreas Transplantation: A Single-Center Experience and Systematic Literature Review.

OBJECTIVES: Technical graft loss, usually thrombotic in nature, accounts for most of the pancreas grafts that are removed early after transplant. Although arterial and venous thrombosis can occur, the vein is predominantly affected, with estimated overall rate of thrombosis of 6% to 33%. In late diagnosis, the graft will need to be removed because thrombectomy will not restore its functionality. However, in early diagnosis, a salvage procedure should be attempted. MATERIALS AND METHODS: We conducted a retrospective, descriptive analysis of a prospective database of patients who underwent pancreas transplant from April 2008 to June 2020 at a single center. We evaluated post-transplant clinical glucose levels, imaging, treatment, and outcomes. We also performed a systematic review of publications for endovascular treatment of vascular graft thrombosis in pancreas transplant. RESULTS: In 67 pancreas transplants analyzed, 13 (19%) were diagnosed with venous thrombus. In 7 of 13 patients (54%), systemic anticoagulation was prescribed because of a non-occlusive thromboses, resulting in complete resolution for all 7 patients. Six patients (46%) required endovascular thrombectomy because of the presence of complete occlusive thrombosis; 4 of these patients (67%) needed a second procedure because of recurrence of the thrombosis. One of the 6 patients (17%) required a surgical approach, resulting in successful removal of the recurrent clot. Twelve of the 13 grafts (92%) were rescued. Graft survival at 1 year was 84%; graft survival at 3, 5, and 10 years remained at 70%. CONCLUSIONS: Pancreas vein thrombosis represents a frequent surgical complication and remains as a challenging problem. In our experience, early diagnoses and an endovascular approach combined with aggressive medical treatment and follow-up can be used for successful treatment and reduce graft loss.

Fulvestrant en la práctica clínica: análisis de efectividad en pacientes uruguayas con cáncer de mama HR+/HER2. / Fulvestrant in clinical practice: Effectiveness analysis in Uruguayan patients with HR+/HER2- breast cancer.

Introduction: Fulvestrant demonstrated benefits in overall survival and progression-free survival in patients with advanced breast cancer, who are hormone receptor-positive and human epidermal growth factor receptor 2 negative. The characteristics, evolution, and survival of patients with hormone receptor-positive, HER2-negative breast cancer treated with fulvestrant were evaluated according to the national treatment coverage protocols of the National Resources Fund, with the aim of understanding the efficacy of fulvestrant in patients treated in usual clinical practice and comparing our results with those from pivotal studies. Methods: A database from the National Resources Fund covering the period from 2009 to 2022 was used. Survival curves were assessed using the Kaplan-Meier method, and differences were analyzed using the Log-Rank test. Results: A total of 1085 patients with an average age of 63,66 years were included. Following a follow-up of 14 months, the median overall survival was 16 months, and the median progression-free survival was 6 months. The presence of liver and bone metastases was associated with a shorter overall survival. Patients from the public sector and those with a better performance status experienced longer overall survival. Conclusions: Our findings provide a valuable perspective for treatment management in a context of limited resources. Overall survival and progression-free survival were somewhat lower than those reported in pivotal clinical trials. The presence of liver and bone metastases was associated with worse prognosis and survival; additionally, patients with worse performance status had shorter overall survival. These findings underscore the need for personalized therapies, opening new lines of future research.

Introducción: Fulvestrant demostró beneficio en sobrevida global y sobrevida libre de progresión en pacientes con cáncer de mama avanzado, con receptores hormonales positivos y receptor de factor de crecimiento epidérmico humano 2 negativo. Se evaluaron las características, la evolución y la sobrevida de pacientes con cáncer de mama receptor hormonal positivo, HER2 negativo, tratadas con fulvestrant, de acuerdo con los protocolos nacionales de cobertura de tratamiento del Fondo Nacional de Recursos. Su objetivo fue conocer la eficacia de fulvestrant en pacientes tratados en la práctica clínica habitual. Se compararon los resultados obtenidos en el presente trabajo con los resultados de los estudios pivotales. Métodos: Se utilizó la base de datos del Fondo Nacional de Recursos, que abarca el período de 2009 a 2022. La evaluación de las curvas de sobrevida se realizó mediante el método Kaplan-Meier y las diferencias se analizaron utilizando el test de Log-Rank. Resultados: Se incluyeron 1085 pacientes con una edad media de 63,66 años. Tras un seguimiento de 14 meses, la mediana de la sobrevida global fue de 16 meses y la de la sobrevida libre de progresión de 6 meses. La presencia de metástasis hepáticas y óseas se asoció con una menor sobrevida global. Los pacientes del sector público y aquellos con una mejor escala de estado funcional experimentaron una mayor sobrevida global. Conclusiones: Los resultados obtenidos ofrecen una perspectiva valiosa para la gestión de tratamientos en un contexto de recursos limitados. La sobrevida global y la sobrevida libre de progresión fueron algo inferiores a los reportados en los ensayos clínicos pivotales. La presencia de metástasis hepáticas y óseas se asoció a un peor pronóstico y una peor sobrevida. Además, los pacientes con peor escala de estado funcional tuvieron una menor sobrevida global. Estos hallazgos subrayan la necesidad de terapias personalizadas, abriendo nuevas líneas de investigación futura.

Predictors of 30-Day Hospital Readmission Following CABG in a Multicenter Database: A Cross-Sectional Study. / Preditores de Readmissão Hospitalar até 30 Dias de CRM em Banco de Dados Multicêntrico: Estudo de Coorte Transversal.

BACKGROUND: The analysis of indicators such as hospital readmission rates is crucial for improving the quality of services and management of hospital processes. OBJECTIVES: To identify the variables correlated with hospital readmission up to 30 days following coronary artery bypass grafting (CABG). METHODS: Cross-sectional cohort study by REPLICCAR II database (N=3,392) from June 2017 to June 2019. Retrospectively, 150 patients were analyzed to identify factors associated with hospital readmission within 30 days post-CABG using univariate and multivariate logistic regression. Analysis was conducted using software R, with a significance level of 0.05 and 95% confidence intervals. RESULTS: Out of 3,392 patients, 150 (4,42%0 were readmitted within 30 days post-discharge from CABG primarily due to infections (mediastinitis, surgical wounds, and sepsis) accounting for 52 cases (34.66%). Other causes included surgical complications (14/150, 9.33%) and pneumonia (13/150, 8.66%). The multivariate regression model identified an intercept (OR: 1.098, p<0.00001), sleep apnea (OR: 1.117, p=0.0165), cardiac arrhythmia (OR: 1.040, p=0.0712), and intra-aortic balloon pump use (OR: 1.068, p=0.0021) as predictors of the outcome, with an AUC of 0.70. CONCLUSION: 4.42% of patients were readmitted post-CABG, mainly due to infections. Factors such as sleep apnea (OR: 1.117, p=0.0165), cardiac arrhythmia (OR: 1.040, p=0.0712), and intra-aortic balloon pump use (OR: 1.068, p=0.0021) were predictors of readmission, with moderate risk discrimination (AUC: 0.70).

FUNDAMENTO: A análise de indicadores como taxa de readmissão hospitalar é crucial para aprimorar a qualidade dos serviços e gestão em processos hospitalares. OBJETIVO: Identificar as variáveis correlacionadas a readmissão hospitalar até 30 dias após cirurgia de revascularização miocárdica (CRM). MÉTODOS: Estudo de coorte transversal no banco de dados Registro Paulista de Cirurgia Cardiovascular II (REPLICCAR II)(N=3.392), de junho de 2017 a junho de 2019. Avaliaram-se retrospectivamente 150 pacientes para identificar os fatores correlacionados a readmissão hospitalar até 30 dias após-CRM via regressão logística univariada e multivariada. As análises foram realizadas no software R, com significância de 0,05 e intervalos de confiança de 95%. RESULTADOS: Cento e cinquenta pacientes foram readmitidos até 30 dias após a alta hospitalar de CRM (150/3.392, 4,42%) principalmente por infecções (mediastinite, ferida operatória e sepse) totalizando 52 casos (52/150, 34,66%), outras causas foram: complicações cirúrgicas (14/150, 9,33%) e pneumonia (13/150, 8,66%). Os preditores de readmissão identificados foram: O modelo de regressão multivariada apontou intercepto (OR: 1,098, p<0,00001), apneia do sono (OR: 1,117, p=0,0165), arritmia cardíaca (OR: 1,040, p=0,0712) e uso de balão intra-aórtico (OR: 1,068, p=0,0021) como preditores do desfecho, com uma AUC de 0,70. CONCLUSÃO: 4,42% dos pacientes foram readmitidos pós-CRM, principalmente por infecções. Fatores como apneia do sono (OR: 1,117, p=0,0165), arritmia cardíaca (OR: 1,040, p=0,0712) e uso de balão intra-aórtico (OR: 1,068, p=0,0021) foram preditores de readmissão, com uma discriminação de risco moderada (AUC: 0,70).

Conductive Hearing Loss Associates With Dementia, and Middle Ear Reconstruction Mitigates This Association: A Multinational Database Study.

OBJECTIVE: To test the hypothesis that conductive hearing loss (CHL) is associated with dementia, and that middle ear reconstruction (MER) associates with improved outcomes for these measures in a multinational electronic health records database. STUDY DESIGN: Retrospective cohort study with propensity-score matching (PSM). SETTING: TriNetX is a research database representing about 110 million patients from the United States, Taiwan, Brazil, and India. PATIENTS: Subjects older than 50 years with no HL and any CHL (ICD-10: H90.0-2). Subjects of any age with and without any MER (CPT: 1010174). MAIN OUTCOME MEASURES: Odds ratios (ORs) and hazard ratios with 95% confidence intervals (95% CIs) for incident dementia (ICD-10: F01, F03, G30). RESULTS: Of 103,609 patients older than 50 years experiencing any CHL, 2.74% developed dementia compared with 1.22% of 38,216,019 patients with no HL (OR, 95% CI: 2.29, 2.20-2.37). Of patients experiencing CHL, there were 39,850 who received MER. The average age was 31.3 years, with 51% female patients. A total of 343,876 control patients with CHL were identified; 39,900 patients remained in each cohort after 1:1 PSM for HL- and dementia-related risk factors. Matched risk for developing dementia among MER recipients was 0.33% compared with 0.58% in controls (OR: 0.58, 0.46-0.72). CONCLUSIONS: CHL increases the odds for dementia, and MER improves the odds for incident dementia. This study represents the first population study on the topic of CHL, MER, and dementia.

Potentially avoidable hospitalizations and associated factors among older people in French Guiana using the French National Health Data System.

Knowing the prevalence of potentially avoidable hospitalizations (PAHs) and the factors associated with them is essential if preventive action is to be taken. Studies on PAHs mainly concern adults, and very few have been carried out in South America. To the best of our knowledge, there has been no study on PAHs in French Guiana, particularly among older adults. This case-control study aimed to estimate the prevalence of PAHs in the Guianese population aged over 65 and to analyze their associated factors. We used the 2017-2019 data from the French National Health Service database (Système National des Données de Santé). The patients were age- and sex-matched 1 : 3 with controls without any PAH in 2019. Factors associated with PAHs were investigated through two conditional logistic regression models [one including the Charlson comorbidity index (CCI) and one including each comorbidity of the CCI], with calculation of the adjusted odds ratio (aOR) and 95% confidence interval (CI). The PAH incidence was 17.4 per 1000 inhabitants. PAHs represented 6.6% of all hospitalizations (45.6% related to congestive heart failure or hypertension). A higher CCI was associated with PAHs [aOR 2.2 (95% CI: 1.6, 3.0) and aOR 4.8 (95% CI: 2.4, 9.9) for 1-2 and ≥3 comorbidities, respectively, versus 0], as was immigrant health insurance status [aOR 2.3 (95% CI: 1.3, 4.2)]. Connective tissue disease, chronic pulmonary disease, congestive heart failure, diabetes, and peripheral vascular disease were comorbidities associated with an increased risk of PAHs. While the prevention of PAHs among immigrants is probably beyond the reach of the Guianese authorities, primary care and a public health policy geared toward prevention should be put in place for the French Guianese population suffering from cardiovascular disease in order to reduce PAHs.

A Data Ingestion Procedure towards a Medical Images Repository.

This article presents an ingestion procedure towards an interoperable repository called ALPACS (Anonymized Local Picture Archiving and Communication System). ALPACS provides services to clinical and hospital users, who can access the repository data through an Artificial Intelligence (AI) application called PROXIMITY. This article shows the automated procedure for data ingestion from the medical imaging provider to the ALPACS repository. The data ingestion procedure was successfully applied by the data provider (Hospital Clínico de la Universidad de Chile, HCUCH) using a pseudo-anonymization algorithm at the source, thereby ensuring that the privacy of patients' sensitive data is respected. Data transfer was carried out using international communication standards for health systems, which allows for replication of the procedure by other institutions that provide medical images. OBJECTIVES: This article aims to create a repository of 33,000 medical CT images and 33,000 diagnostic reports with international standards (HL7 HAPI FHIR, DICOM, SNOMED). This goal requires devising a data ingestion procedure that can be replicated by other provider institutions, guaranteeing data privacy by implementing a pseudo-anonymization algorithm at the source, and generating labels from annotations via NLP. METHODOLOGY: Our approach involves hybrid on-premise/cloud deployment of PACS and FHIR services, including transfer services for anonymized data to populate the repository through a structured ingestion procedure. We used NLP over the diagnostic reports to generate annotations, which were then used to train ML algorithms for content-based similar exam recovery. OUTCOMES: We successfully implemented ALPACS and PROXIMITY 2.0, ingesting almost 19,000 thorax CT exams to date along with their corresponding reports.

Epidemiology of high-energy trauma in Chile: An ecological analysis using public registries. / Epidemiología del trauma de alta energía en Chile: análisis ecológico con registros públicos.

High-energy trauma is defined as severe organic injuries resulting from events that generate a large amount of kinetic, electrical, or thermal energy. It represents a significant public health concern, accounting for 10% of global mortality. This article aims to describe the epidemiology of high-energy trauma in Chile. Specifically, it seeks to compare the mortality rate per 100 000 inhabitants among member countries of the World Health Organization (WHO), provide a descriptive analysis of notifications under the Explicit Health Guarantees (GES) for the health issue of polytraumatized patients, and analyze the trend in the mortality rate due to external causes in Chile. This study employs an ecological design using three open-access databases. First, the WHO database on deaths from traffic accidents in 2019 was used. Then, the GES database was consulted for the "Polytraumatized" issue between 2018 and 2022. Finally, the Chilean Department of Health Statistics database on causes of death between 1997 and 2020 was utilized. In 2019, Chile ranked in the middle regarding the mortality rate per 100 000 inhabitants due to traffic accidents. GES notifications for polytrauma predominantly involved men aged 20 to 40 years and those affiliated with the public health system, highlighting a primary focus for prevention efforts. Mortality from accidents showed a decreasing trend, with significant structural changes identified in 2000 and 2007.

El trauma de alta energía se define como lesiones orgánicas graves resultantes de eventos que generan una gran cantidad de energía cinética, eléctrica o térmica. Constituye una importante preocupación de salud pública, representando el 10% de la mortalidad mundial. El objetivo de este artículo es describir la epidemiología del trauma de alta energía en Chile. Específicamente, se busca comparar la tasa de mortalidad por 100 000 habitantes entre los países miembros de la Organización Mundial de la Salud (OMS), realizar un análisis descriptivo de las notificaciones por Garantías Explícitas en Salud (GES) del problema de salud "politraumatizado", y analizar la tendencia de la tasa de fallecidos por causa externa en Chile. El presente estudio tiene un diseño ecológico, utilizando tres bases de datos de acceso abierto. Primero, se utilizó la base de datos de la OMS sobre fallecidos por accidentes automovilísticos en 2019. Luego, se consultó la base de datos del programa Garantías Explícitas en Salud para el problema "politraumatizado" entre los años 2018 y 2022. Finalmente, se utilizó la base de datos del Departamento de Estadísticas de Salud de Chile sobre causas de muerte entre 1997 y 2020. En 2019, Chile ocupó una posición intermedia en cuanto a la tasa de mortalidad por 100 000 habitantes debido a accidentes de tráfico. Las notificaciones el programa Garantías Explícitas en Salud por politraumatismo fueron predominantemente en hombres de entre 20 y 40 años, afiliados al sistema de salud pública. Por este motivo, el foco principal de prevención debe centrarse en este grupo. La mortalidad por accidentes mostró una tendencia decreciente, identificándose cambios estructurales significativos en los años 2000 y 2007.

Pilot deployment of a cloud-based universal medical image repository in a large public health system: A protocol study.

This paper outlines the protocol for the deployment of a cloud-based universal medical image repository system. The proposal aims not only at the deployment but also at the automatic expansion of the platform, incorporating Artificial Intelligence (AI) for the analysis of medical image examinations. The methodology encompasses efficient data management through a universal database, along with the deployment of various AI models designed to assist in diagnostic decision-making. By presenting this protocol, the goal is to overcome technical challenges and issues that impact all phases of the workflow, from data management to the deployment of AI models in the healthcare sector. These challenges include ethical considerations, compliance with legal regulations, establishing user trust, and ensuring data security. The system has been deployed, with a tested and validated proof of concept, possessing the capability to receive thousands of images daily and to sustain the ongoing deployment of new AI models to expedite the analysis process in medical image exams.

Prostaglandin analogues signal detection by data mining in the FDA Adverse Event Reporting System database.

OBJECTIVE: This study aims to identify safety signals of ophthalmic prostaglandin analogues through data mining the Food and Drug Administration Adverse Event Reporting System (FAERS) database. METHODS: A data mining search by proportional reporting ratio, reporting OR, Bayesian confidence propagation neural network, information component 0.25 and χ2 for safety signals detection was conducted to the FAERS database for the following ophthalmic medications: latanoprost, travoprost, tafluprost and bimatoprost. RESULTS: 12 preferred terms were statistically associated: diabetes mellitus, n=2; hypoacusis, n=2; malignant mediastinal neoplasm, n=1; blood immunoglobulin E increased, n=1; cataract, n=1; blepharospasm, n=1; full blood count abnormal, n=1; skin exfoliation, n=1; chest discomfort, n=1; and dry mouth, n=1. LIMITATION OF THE STUDY: The FAERS database's limitations, such as the undetermined causality of cases, under-reporting and the lack of restriction to only health professionals reporting this type of event, could modify the statistical outcomes. These limitations are particularly relevant in the context of ophthalmic drug analysis, as they can affect the accuracy and reliability of the data, potentially leading to biased or incomplete results. CONCLUSIONS: Our findings have revealed a potential relationship due to the biological plausibility among malignant mediastinal neoplasm, full blood count abnormal, blood immunoglobulin E increased, diabetes mellitus, blepharospasm, cataracts, chest discomfort and dry mouth; therefore, it is relevant to continue investigating the possible drug-event association, whether to refute the safety signal or identify a new risk.

A transfer learning approach to identify Plasmodium in microscopic images.

Plasmodium parasites cause Malaria disease, which remains a significant threat to global health, affecting 200 million people and causing 400,000 deaths yearly. Plasmodium falciparum and Plasmodium vivax remain the two main malaria species affecting humans. Identifying the malaria disease in blood smears requires years of expertise, even for highly trained specialists. Literature studies have been coping with the automatic identification and classification of malaria. However, several points must be addressed and investigated so these automatic methods can be used clinically in a Computer-aided Diagnosis (CAD) scenario. In this work, we assess the transfer learning approach by using well-known pre-trained deep learning architectures. We considered a database with 6222 Region of Interest (ROI), of which 6002 are from the Broad Bioimage Benchmark Collection (BBBC), and 220 were acquired locally by us at Fundação Oswaldo Cruz (FIOCRUZ) in Porto Velho Velho, Rondônia-Brazil, which is part of the legal Amazon. We exhaustively cross-validated the dataset using 100 distinct partitions with 80% train and 20% test for each considering circular ROIs (rough segmentation). Our experimental results show that DenseNet201 has a potential to identify Plasmodium parasites in ROIs (infected or uninfected) of microscopic images, achieving 99.41% AUC with a fast processing time. We further validated our results, showing that DenseNet201 was significantly better (99% confidence interval) than the other networks considered in the experiment. Our results support claiming that transfer learning with texture features potentially differentiates subjects with malaria, spotting those with Plasmodium even in Leukocytes images, which is a challenge. In Future work, we intend scale our approach by adding more data and developing a friendly user interface for CAD use. We aim at aiding the worldwide population and our local natives living nearby the legal Amazon's rivers.

A standardized clinical database for research in Chagas disease: The NHEPACHA network.

The NHEPACHA Iberoamerican Network, founded on the initiative of a group of researchers from Latin American countries and Spain, aims to establish a research framework for Chagas disease that encompasses diagnosis and treatment. For this purpose, the network has created a questionnaire to gather relevant data on epidemiological, clinical, diagnostic, and therapeutic aspects of the disease. This questionnaire was developed based on a consensus of expert members of the network, with the intention of collecting high-quality standardized data, which can be used interchangeably by the different research centers that make up the NHEPACHA network. Furthermore, the network intends to offer a clinical protocol that can be embraced by other researchers, facilitating comparability among published studies, as well as the development of therapeutic response and progression markers.

Racial disparities and maternal mortality in Brazil: findings from a national database.

OBJECTIVE: To assess maternal mortality (MM) in Brazilian Black, Pardo, and White women. METHODS: We evaluated the maternal mortality rate (MMR) using data from the Brazilian Ministry of Health public databases from 2017 to 2022. We compared MMR among Black, Pardo, and White women according to the region of the country, age, and cause. For statistical analysis, the Q2 test prevalence ratio (PR) and confidence interval (CI) were calculated. RESULTS: From 2017 to 2022, the general MMR was 68.0/100,000 live births (LB). The MMR was almost twice as high among Black women compared to White (125.81 vs 64.15, PR = 1.96, 95%CI:1.84-2.08) and Pardo women (125.8 vs 64.0, PR = 1.96, 95%CI: 1.85-2.09). MMR was higher among Black women in all geographical regions, and the Southeast region reached the highest difference among Black and White women (115.5 versus 60.8, PR = 2.48, 95%CI: 2.03-3.03). During the covid-19 pandemic, MMR increased in all groups of women (Black 144.1, Pardo 74.8 and White 80.5/100.000 LB), and the differences between Black and White (PR = 1.79, 95%CI: 1.64-1.95) and Black and Pardo (PR = 1.92, 95%CI: 1.77-2.09) remained. MMR was significantly higher among Black women than among White or Pardo women in all age ranges and for all causes. CONCLUSION: Black women presented higher MMR in all years, in all geographic regions, age groups, and causes. In Brazil, Black skin color is a key MM determinant. Reducing MM requires reducing racial disparities.