Giovane Thomazini Soares

Universidade de São Paulo

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Domingos Alves

Universidade de São Paulo

Ana Clara de Andrade Mioto

Universidade de São Paulo

Filipe Andrade Bernardi

Universidade de São Paulo

Vinícius Lima

Universidade de São Paulo

Mariane Barros Neiva

Universidade de São Paulo

Diego Bettiol Yamada

Universidade de São Paulo

Victor Cassão

Universidade de São Paulo

Renan Barbieri Segamarchi

Universidade de São Paulo

Newton Shydeo Brandão Miyoshi

Centro Universitário Barão de Mauá

Têmis Maria Félix

Hospital de Clínicas de Porto Alegre

Cooperative Institutions

Universidade de São Paulo

Clinics Hospital of Ribeirão Preto

Universidade de Ribeirão Preto

Universidade Federal do Rio Grande do Sul

Hospital de Clínicas de Porto Alegre

Centro Universitário Barão de Mauá

Universidade Federal de São Carlos

Universidade Brasil

Brazilian Society of Computational and Applied Mathematics

Universidade Federal de Minas Gerais

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Preliminary Results to Predict Tuberculosis Outcomes Applying Traditional and Automated Machine Learning Models

Procedia Computer Science (2023)

Ana Clara de Andrade Mioto Mariana Tavares Mozini Renan Barbieri Segamarchi Giovane Thomazini Soares Pedro Emilio Andrade Martins

Tuberculosis (TB) remains one of the most lethal infectious diseases in the world and, despite being preventable and curable, kills 4.500 people daily, according to the World Health Organization (WHO). Brazil, being a country heavily affected by TB, works to improve social intervention programs, since the decrease in the patients vulnerability seems to have a positive effect for the cure of TB. The Brazilian public health system records data on TB treatment that can guide actions and interventions. In this context, machine learning (ML) algorithms have been used successfully to analyze health and medicine (H&M) datasets. An emerging area of ML called Automated Machine Learning (Auto-ML) was tested in this analysis to predict the following TB results: good and bad outcomes. Our results indicate that it is possible to build reasonable ML models with the available data.

Vulnerability

10.1016/j.procs.2023.01.424

Cite

Citations (3)

OUTB: application for decision-support in the outcomes of Tuberculosis

Procedia Computer Science (2024)

Mariana Tavares Mozini R. Rothschild Ana Clara de Andrade Mioto Filipe Andrade Bernardi Vinícius Lima

Brazil is among the 30 countries with the highest tuberculosis prevalence, and many studies are trying to understand and fight the harmful outcomes of this disease. In that sense, the following project proposes the development of a mobile application prototype that centralizes diagnosis aid tools for tuberculosis, making the process more transparent for the patient, using the Single Patient Application (SPA) concept. The application was built based on JavaScript and the frameworks chosen were Vue Framework along with Framework7 due to its focus on interface components. Communication with the back-end was done through AJAX calls. As a result, a prototype was built with the five main screens that the user will interact with: authentication/login screen, home screen, list of available algorithms, outcomes algorithm home screen and patient history. The interface design is expected to facilitate the planning of other application areas and contribute to defining business rules. Once implemented on the server side, these rules can support an organized, patient-centered tuberculosis database.

Interface (matter)

10.1016/j.procs.2024.06.241

Cite

Citations (0)

Avaliação paradigmática da saúde paulista sob as lentes da complexidade

Giovane Thomazini Soares

Notable similarities exist between complex systems and Health at the Interurban level of the state of São Paulo (SIUSP, interurban network, whose nodes are cities with more than 50 thousand inhabitants and Population Concentration Areas, ACPs).Thus, by analogy, the hypothesis that SIUSP presents emergents was built.This implies, given other properties of SIUSP, that the SIUSP is a complex system (from this perspective, the SIUSP is categorized in level of emergents and in level of interactive structure; in the first mentioned level, its edges refer to the health of their nodes, in the second level, its edges describe interactions between the nodes).The veracity of this implication, together with other assumptions already validated, incurs the need for São Paulo Public Health at Intermunicipal level (SPPIM, composed of formal connections between cities, such connections deal strictly with health and are instituted in a top-down manner; exemples of SPPIM are health regions) to be understood by a paradigm linked to the sciences of complexity (Holistic Worldview Paradigm, PVMH).The current SPPIM paradigm may be incompatible (given the perspective of properties of complex systems) with the PVMH.Thus, the consolidation that the SIUSP presents emergents provides a probable paradigmatic change of the way of understanding the SPPIM.The new paradigm that would support the understanding of SPPIM would be the PVMH (because the later is capable of handling properties of complex systems).For those important consequences, is tested the compatibility between the PVMH and the current paradigm of SPPIM.For the same reason, is investigated if SIUSP shows significant evidences that she has emergents.Better answers about the paradigmatic change are obtained if the last investigation mentioned is based on the possible new paradigm.In this context, the research aims to identify if there are meningful evidences that corroborate that the current paradigma of SPPIM must be replaced by the PVMH, aligned with the theory of complexity.To achieve this objective, the research is based on data collected from secondary source (DATASUS and IBGE), analyzed by multivariate statistics, whose results are interpreted by content analysis method.Thus, the research presents quantitative and qualitative aspects; it is also categorized as exploratory.It generates similarity networks of annual variations of cities and ACPs located in the state of São Paulo.These networks are representations of the SIUSP at emergents level.Based on these networks, it was identified that the instituted order, of regions and macro-regions of health, was not able to explain various behaviors of SIUSP.One of these networks made it possible to capture non-trivial interurban patterns that demonstrate strong temporal persistence.However, such networks have not shown great evidences of emergents properties.Thus, despite of the results presented and the identification of incompatibility between the PVMH and the current SPPIM paradigm, was concluded that it is not possible to state, with the information obtained, that exist meningful evidences that corroborate that the current paradigma of SPPIM must be replaced by the PVMH.

Interurban

Paradigm shift

10.11606/d.96.2020.tde-21072020-143540

Cite

Citations (0)

Scaling laws and spatial effects of Brazilian health regions: a research protocol

Procedia Computer Science (2023)

Giovane Thomazini Soares Diego Bettiol Yamada Filipe Andrade Bernardi Mariane Barros Neiva Luís Pedro Lombardi

The literature has already consolidated the importance of health regions for Brazilian public health. Complexity properties strongly mark such regions. In this context, there are abundant indications that health regions should be analyzed with approaches linked to the sciences of complexity. One of these approaches, the estimation of scaling laws, can describe important properties of socio-spatial elements. However, no studies estimate the scaling laws of Brazilian health regions. This research protocol can remedy this limitation, proposing the estimation of scaling laws of the previously mentioned regions, mainly considering variables relevant to Brazilian public health. Still, this paper can substantially mitigate other relevant limitations of usual research that estimate scaling laws of socio-spatial elements. These mitigations, which provide advances in the literature on estimating scaling laws, are given by the proposal of modeling (if necessary) spatial effects and estimating scaling laws for the entire population of the socio-spatial elements. According to the theory, the expected results are non-linear scaling laws, which will likely vary with space and time and coexist with relevant spatial effects. From such laws and effects, it will be possible to accurately characterize the performance of each health region through Spatial and Scale Adjusted Metropolitan Indicators and unravel spatio-temporal properties, stabilities, and instabilities of sets composed of health regions. The expected findings of this paper can help rearrange health regions and improve the quality of information used in Brazilian public health planning.

Scaling law

10.1016/j.procs.2023.01.417

Cite

Citations (2)

ICD-10 - ORPHA: An Interactive Complex Network Model for Brazilian Rare Diseases

Procedia Computer Science (2024)

Mariane Barros Neiva Bibiana Mello de Oliveira Amanda Maria Schmidt Victória Machado Scheibe Júlia Cordeiro Milke

A disease is considered rare if it has a low prevalence. It is estimated that around 400 million people worldwide have a rare disease, including 15 million in Brazil. Consequently, it became a public health priority for the World Health Organization and the Brazilian Health Ministry. In 2014, the Brazilian government launched a national policy regarding the care for rare patients', the Ordinance nº199. The national politic defines guidelines, procedures, and descriptions of rare disease codes to provide access and diagnosis in the public health system to reduce mortality and improve patient's quality of life. Diseases are identified according to the International Classification of Diseases 10th Revision, a widely used terminology in this context. However, there are also different terminologies to codify a rare disease, such as the ORPHAcode provided by Orphanet. This paper proposes a complex network model using the terminologies' relationship to show that the International Classification of Diseases 10th Revision may be generic for diagnosing rare Brazilian patients. Moreover, there is no perfect nomenclature to define rare diseases, but each context has a better application. So, mapping the relationship between each terminology is fundamental for creating consistent semantic relationships in biomedical ontologies, providing a functional environment for carrying out tasks involving more than one terminology.

Rare disease

10.1016/j.procs.2024.06.218

Cite

Citations (2)

A Web Portal for Real-Time Data Quality Analysis on the Brazilian Tuberculosis Research Network: A Case Study

Lecture notes in computer science (2023)

Victor Cassão Filipe Andrade Bernardi Vinícius Lima Giovane Thomazini Soares Newton Shydeo Brandão Miyoshi

10.1007/978-3-031-36024-4_24

Cite

Citations (2)

Batalha paradigmática: O confronto entre o paradigma cartesiano-newtoniano e o paradigma de visão de mundo holística

Giovane Thomazini Soares João Luiz Passador

10.22533/at.ed.112211708

Cite

Citations (0)

Informed Consent Form Automated Validation, The Brazilian Rare Disease Network Case Proposal

Procedia Computer Science (2023)

Mariane Barros Neiva André Luiz Teixeira Vinci Diego Bettiol Yamada Filipe Andrade Bernardi Giovane Thomazini Soares

The informed consent form (ICF) is required for all observational studies involving human subjects in Brazil. Besides the existence of e-Consent technologies, Brazilian guidelines issued by the Ministry of Health, the ICF must be obtained in the physical form in face-face services. Furthermore, the Brazilian Network of Rare Diseases (RARAS) project was proposed in a context marked by a scarcity of structured data on Rare Diseases (RD). One of the main objectives of RARAS is to understand and expose the RD scenario in Brazil. Since one of the stages in the RARAS project requires data collected from a patient's interview, the ICF is mandatory. Therefore, the importance of completeness of participants' process aroused the need of the technical team of RARAS to propose a protocol to automate the validation of scanned physical ICFs. The purpose is based on applying image preprocessing methods and a deep learning model. Regarding previous results in the literature, the expected outcome is to achieve around 90% accuracy in the classification of ICF for the RARAS project.

Christian ministry

10.1016/j.procs.2023.01.445

Cite

Citations (1)