A5013963197- Machine Learning in Healthcare
- Autopsy Techniques and Outcomes
- Data-Driven Disease Surveillance
- Global Maternal and Child Health
- Protein Structure and Dynamics
- Parasites and Host Interactions
- Topic Modeling
- Migration, Health and Trauma
- Global Public Health Policies and Epidemiology
- Research on Leishmaniasis Studies
- Metabolism and Genetic Disorders
- Growth Hormone and Insulin-like Growth Factors
- Viral Infections and Immunology Research
- Anomaly Detection Techniques and Applications
- RNA and protein synthesis mechanisms
- Helminth infection and control
- Genomics and Rare Diseases
- Zoonotic diseases and public health
- Machine Learning in Bioinformatics
- Protein purification and stability
- Advanced Text Analysis Techniques
- Lipid metabolism and disorders
- Emergency and Acute Care Studies
- Computational Drug Discovery Methods
- Enzyme Production and Characterization
Hospital Santa Marcelina
2024
DASA (Brazil)
2024
Merck (Germany)
2023
St. Michael's Hospital
2015-2021
Centre for Global Health Research
2015-2021
University of Toronto
2017-2018
St Michaels Hospital
2017
University of Oxford
2012
Progress toward the United Nations 2030 Sustainable Development Goals requires improved information on mortality and causes of death. However, many fifty million annual deaths in low- middle-income countries remain unknown, as most occur at home without medical attention. In 2001 India began Million Death Study 1.3 nationally representative households. Nonmedical staff conduct verbal autopsies, which are structured interviews including a half-page narrative local language family's story...
Verbal autopsies (VA) are increasingly used in low- and middle-income countries where most causes of death (COD) occur at home without medical attention, deaths differ substantially from hospital deaths. Hence, there is no plausible "standard" against which VAs for may be validated. Previous studies have shown contradictory performance automated methods compared to physician-based classification CODs. We sought compare the classic naive Bayes classifier (NBC) versus existing classifiers,...
A verbal autopsy (VA) is a post-hoc written interview report of the symptoms preceding person's death in cases where no official cause (CoD) was determined by physician. Current leading automated VA coding methods primarily use structured data from VAs to assign CoD category. We present method automatically determine categories free-text narratives alone. After preprocessing and spelling correction, our extracts word frequency counts uses them as input four different machine learning...
Abstract In this pilot study, we aimed to evaluate the feasibility of whole genome sequencing (WGS) as a first‐tier diagnostic test for infants hospitalized in neonatal intensive care units Brazilian healthcare system. The cohort presented here results from joint collaboration between private and public hospitals Brazil considering initiative clinical laboratory provide timely diagnosis critically ill infants. We performed trio (proband parents) WGS 21 suspected genetic disease with an...
Abstract Background Predicting protein contacts solely based on sequence information remains a challenging problem, despite the huge amount of data at our disposal. Mutual Information (MI), an theory measure, has been extensively employed and modified to identify residues within (intra-protein) that are in contact. More recently MI its variants have also used prediction between proteins (inter-protein). Methods Here we assess predictive power for domain-domain contact prediction. We test...
We introduce a multi-task learning model for cause-of-death classification of verbal autopsy narratives that jointly learns to output interpretable key phrases. Adding these phrases outperforms the baseline and topic modeling features.
Background: Verbal autopsies (VA) are increasingly used in low- and middle-income countries where most causes of death (COD) occur at home without medical attention, deaths differ substantially from hospital deaths. Hence, there is no plausible “standard” against which VAs for may be validated. Previous studies have shown contradictory performance automated methods compared to physician-based classification CODs. We sought compare the classic naive Bayes classifier (NBC) versus existing...
Alpha-mannosidosis is an ultra-rare autosomal recessive genetic disorder that affects approximately 1:500,000 to 1:1000,000 live births globally, with no specific prevalence in any ethnic group or gender. It results from a MAN2B1 gene alteration, which codes for the lysosomal enzyme alpha-mannosidase. Its pathophysiology complex, there accumulation of mannose-rich oligosaccharides cellular lysosome, causing edema and compromising functions, but it does not explain completely clinical...
Hypertriglyceridemia (HTG) is a common dyslipidemia associated with an increased risk of cardiovascular disease and pancreatitis. It well stablished that the severe cases often present underlying genetic cause. In this study, we determined frequency variation spectrum genes involved in triglyceride metabolism series Brazilian patients HTG. A total 212 very high HTG, defined fasting triglycerides (TG) ≥ 880 mg/ dL, underwent multi-gene panel testing were included research. Germline...
The World Health Organization (WHO) recently published target product profiles (TPPs) for neglected tropical diseases (NTDs) to inform and accelerate the development of diagnostics tools necessary achieve targets in decade ahead. These TPPs describe minimal ideal requirements various diagnostic needs related NTD specific use-cases. An early step towards manufacture implementation new is critically review translate these into an initial design ultimately user requirement specifications (URS)....
Sensitivity and specificity of assignment by cause death, on Agincourt data (ages 15â 64 years).(DOC 71 kb)
Sensitivity and specificity of assignment by cause death, on Matlab data (ages 20â 64 years). (DOC 69 kb)
Background: Verbal autopsies (VA) are increasingly used in low- and middle-income countries where most causes of death (COD) occur at home without medical attention, deaths differ substantially from hospital deaths. Hence, there is no plausible “standard” against which VAs for may be validated. Previous studies have shown contradictory performance automated methods compared to physician-based classification CODs. We sought compare the classic naive Bayes classifier (NBC) versus existing...