graph.md

By graph.md

From the doctor's office to the patient's home! The patient will be the center of the entire process. The key issue is reverse the concept of telemedicine. Dear Patient P! Analyzing your data, we woul... more

Download on the App Store

Download on the App Store

Get it on Google Play

FAQs about graph.md:

How many episodes does graph.md have?

The podcast currently has 54 episodes available.

graph.md episodes:

July 10, 2024 diagnosis e2
diagnosis.health.eco.br
The health.eco.br proposes an intelligent model for the patient's virtual medical record. The main idea is that each person can have their own medical record in an application with their entire history of medical care (consultations, exams, hospitalizations, and diagnoses). The data would be owned by the patient and could be made available to doctors during consultations via a web interface. At the end of each service, the medical record would be updated through an automated integration process with the clinic or hospital system, permitted by the use of protocols used in the healthcare sector. The model stores information in concepts within a knowledge base through a system of intelligent agents.
new eco's main objective is to develop personal information systems with Graph DB integrated with next.js. Using the Neo4j Graph Data Science API with ML in Python together with Apache SparkX and Kafka tools, we are specifically interested in (D3.js and Neovis) to support the visualization of the knowledge synthesis platform's personal Graph DBs.
The basic idea is to provide a unified electronic space for people to store their personal information in intelligent Graph DBs. We created the new.eco.br digital ecosystem with smart filter tools and map visualization in our sintese.eco.br console that will bring added value to the project. The special interest is in using Timeline support integrated with the parameters generated by GDS (Graph Data Science). In this way, we intend to use the timeline visualization functionalities associated with the predictive parameters generated by AI. Thus providing possibilities of choice, and future simulation, to show users the positive impact of changes in their parameters (current situation).
For example, we want to show the positive impact on a person's health if they change their current (bad) habits to future (healthy) ones. This example can be used for different areas in people's lives: nutrition, supplementation, sleep, education, sports, relationships, etc. Thus, the focus would be to use the Timeline not only for the past but also for the future.
The electronic medical record is defined as a set of information about the patient, and their health status, including the medical care and treatments they received, maintained electronically and stored in such a way that they could be used by multiple health professionals integrated into a network. Commonly called medical records, it is considered an essential document for the care of individuals about their health and must provide all the necessary information to ensure that, through this, efficient care and continuity of treatments provided to the patient are possible.
It is also necessary to think about integrating the PEP (Electronic Patient Record) with other systems and equipment. The use of standards helps ensure that the data from the proposed system is structured to the point that it can be understood by other systems. With this, the need to use standards also emerged in the health area. They can be classified into patterns of identification; communication; content and structure; representation of clinical data; confidentiality, security and authentication; quality indicators, datasets, and guidelines.
In the project at health.eco.br, certification standards for electronic health record systems are used. An important standard applied in the system is HL7 (Health Level Seven), an international protocol for the exchange of electronic data in all healthcare environments, which defines not only the format of the message but also its contents, according to the circumstances in which they communicate. This protocol will be used to exchange the content (messages) of consultation information between health.eco.br and healthcare institutions through synchronous and asynchronous communication between healthcare systems.
...more
4min
July 10, 2024 diagnosis e1
diagnosis.health.eco.br
Many diagnoses in various areas of medicine and biomedicine are often complex, and expensive and include an extensive sequence of steps. These steps can be highly complex, compromising diagnostic performance, especially in remote regions of developing countries.
Quick access to patient information and exams, in an organized, decentralized, and distributed way, is very important for the medical community in general. The use of this data by Diagnostic Support Systems (CAD) can increase the accuracy of the result.
The development of computational models based on the specialist doctor's perspective aims to transform clinical information into knowledge, to propose the creation of more efficient diagnostic aid systems. This definition aims to facilitate the data analysis process and allow the formalization of information related to each diagnosis. The computational concepts used to create this technology are linked to Knowledge Engineering.
With the automated analysis of information banks (clinical data, laboratory tests, imaging tests, and diagnoses), the aim is to establish the degree of relationship between a specific diagnostic context and a given patient and their respective symptoms. In strict terms, knowledge is only created by individuals, and organizations need to support creative individuals or provide them with appropriate contexts. This process must be understood as a factor to organizationally expand the knowledge created by individuals, crystallizing it as part of the network within the institution. It is also observed that this process occurs within an expanding community of interaction, which crosses inter-organizational levels and boundaries. Ontology can be defined as an abstract and simplified explicit specification (conceptualization) of a domain to be represented. In the area of Intelligent Systems, if an entity can be represented, then it is an explicit specification of a conceptualization. Thus, conceptualization is defined as the structured interpretation of concepts and, relating them. In this case, those selected as relevant in a given domain are referred to. The term “explicit” refers to the set of these terms used and their restrictions are previously and explicitly defined. The importance of formalizing an ontology refers to the ability to process it by the computer, which excludes definitions in natural language. Its sharing is justified because it describes consensual knowledge, which is used by more than one individual, and accepted by a group. Thus, an ontological structure is the union of a non-empty set of attributes, concepts, and specifications in opposition to hierarchies, with the basic function of representing the domain through relationships and axioms.
The level of knowledge provides the means to "rationalize" the behavior of a system from the point of view of an external observer, formalization is established through an agent who has the intelligence to complete its objectives. The agent can execute a set of actions and choose them according to the principle of rationality. A KL (Knowledge Level) model, however, is not only composed of the domain that the agent appears to be using but, more importantly, it is the structure within which this knowledge is being used to achieve the objectives.
A CAD application has two aspects that influence its effectiveness. The first is associated with the information processing methodology, and the second is with the presentation of results through an intuitive interaction with the user. A single environment, which is primarily ergonomic, would enable the exploration of all related data within a large hybrid universe, with images of different types and textual information, or even structured, which would allow research in different sources of information, support diagnosis and/or health research.
...more
3min
July 10, 2024 patient e3
patient.health.eco.br
Where is your data? Shouldn't they be in our personal databases?
But do we have a personal knowledge base?
Unfortunately not yet today!
But now, why don't we have it?
Proposal for effective centralization of all your data in an Intelligent Graph DB in the cloud:
A characteristic that is surprising today is that people do not know at all where their data is. They fly around the world and are only accessed through login. Everything converges before the user interacts with the different applications and websites whose sole objective is purchasing. This process ends, the logout happens and the data disappears, no one knows where your data is. Except for the sad set of steps describing the app screens, but what if the doctor asks if you've ever had a specific symptom? Then the truth comes out! People do not have ownership over their data. They don't even know it exists! And the company's systems are made with specific mass production and consumption goals in mind. They are not modeled and treated in the correct way to be synthesized into knowledge bases. A simple change of focus, from the company to the person, makes all the difference here! Seeking to build knowledge bases of people and user groups with clear and concrete objectives. Enabling users to manage this synthetic knowledge base makes all the difference. And to think that graphs emerged in the early days of computing. Today in Graph Data Science we use some algorithms defined in the 70s, or even from earlier times. With the advancement of knowledge engineering combined with the visualization of complex data proposed by visualization APIs, we will be able to contribute to the construction of super apps. And if we think formally of a complex system, the objective would be compatibility, within the unification proposition proposed by Von Neumann who, before his early death, had the idea of integrating the current architecture that bears his name with the architecture of automatons, also defined by him. Now, leaving formalism aside, in the explanatory sense, the processing would be in the graph itself. It is worth thinking that even at that time the genius saw that the models were compatible.
...more
2min
July 10, 2024 paciente e3
paciente.health.eco.br
Onde estão os seus Dados? Eles não deveriam estar em nossos bancos de dados pessoais?
Mas temos uma base de conhecimento pessoal?
Infelizmente hoje ainda não!
Mas agora, por que não a temos?
Proposta de centralização efetiva de todos os seus dados em um Graph DB Inteligente na nuvem:
Uma característica que hoje causa estranheza é que as pessoas não sabem absolutamente onde estão seus dados. Eles voam pelo mundo e só são acessados através do login. Tudo converge diante do usuário que interage nos diversos aplicativos e sites cujo único objetivo é a compra. Esse processo termina, o logout acontece e os dados desaparecem, não se sabe ninguém que saiba onde estão seus dados. Exceto pelo triste conjunto de etapas que descrevem as telas do aplicativo, mas e se o médico perguntar se você já teve um determinado sintoma. Então a verdade aparece! As pessoas não têm propriedade sobre seus dados. Eles nem sabem da sua existência! E os sistemas da empresa são feitos com objetivos específicos de produção e consumo em massa. Eles não são modelados e tratados da forma correta para serem sintetizados em bases de conhecimento. Uma simples mudança de foco, da empresa para a pessoa, faz toda a diferença aqui! Buscando construir bases de conhecimento de pessoas e grupos de usuários com objetivos claros e concretos. Permitir que os usuários gerenciem essa base de conhecimento sintética faz toda a diferença. E pensar que os grafos surgiram nos primórdios da computação. E hoje em Graph Data Science usamos alguns algoritmos definidos na década de 70, ou mesmo de épocas anteriores. Com o avanço da engenharia do conhecimento aliado à visualização de dados complexos proposta pelas APIs de visualização, poderemos contribuir para a construção de super apps. E se pensarmos formalmente em um sistema complexo, o objetivo seria a compatibilidade, dentro da proposição de unificação proposta por Von Neumann que, antes de sua morte precoce, teve a ideia de integrar a arquitetura atual que leva seu nome com a da arquitetura de autômatos, também definida por ele. Agora, deixando um pouco de lado o formalismo, no sentido explicativo, o processamento ficaria no próprio grafo. Vale pensar que ainda naquela época o gênio viu que os modelos eram compatíveis.
...more
2min
July 09, 2024 paciente e2
paciente.health.eco.br
Estamos em um momento importante de valorização da propriedade da informação. A sensação do inconsciente coletivo é que as pessoas querem poder ser donas dos seus dados, mas a forma de utilização proposta pelo modelo atual não permite isso. Os sistemas são desenvolvidos pelas empresas apenas para realizar os mais diversos processos e procedimentos operacionais de registo. Os dados das pessoas nada mais são do que instâncias desses bancos relacionais, apenas uma minúscula célula, num fluxo contínuo de desenvolvimento infinito de novas versões para atender cada vez mais consumidores. Dentro deste paradigma, a propriedade da informação pelas pessoas nada mais será do que um sonho distante. Nunca deixará de ser uma utopia. E por que não, se pensarmos na sua efetiva não realização, uma distopia?
Impressionantemente, este problema insolúvel no atual contexto de alcançar um melhor nível de inteligência computacional é simples de resolver se concentrarmos os nossos esforços no desenvolvimento dos sistemas das pessoas. Para tanto, é necessário um protocolo de comunicação entre as partes. Com sistemas de gerenciamento e controle para movimentação de dados entre Graph DBs individuais e grupos específicos de pessoas, ou ainda salas coletivas de grupos temáticos segmentados por áreas do conhecimento. De qualquer forma, o fator decisivo neste caso é a centralização e o controle dos dados das pessoas que terão a propriedade real de suas bases de conhecimento.
Com filtros avançados de caminhamento nos grafos, trabalhamos com o conceito de interação completa dos eventos. Cada pessoa possui múltiplos eventos com registros de toda a extensão da interação nos Graph DBs. Traduzindo esse conceito, pode-se dizer que é possível percorrer um caminho de 50 conceitos e relacioná-los a uma pessoa a partir de 50 mil conceitos. Com o balanceamento de bases também é possível definir os conceitos de precisão e recall para definir a auto construtividade das bases e equipará-las ao alto desempenho no reconhecimento de padrões.
...more
2min
July 09, 2024 patient e2
patient.health.eco.br
We are at an important moment of valuing the ownership of information. The feeling of the collective unconscious is that people want to be able to own their data, but the way of using it proposed by the current model does not allow this. The systems are developed by companies only to carry out the most diverse operational registration processes and procedures. People's data are nothing more than instances of these relational banks, just tiny cells, in a continuous flow of infinite development of new versions to serve more and more consumers. Within this paradigm, people's ownership of information will be nothing more than a distant dream. It will never stop being a utopia. And why not, if we think about its effective non-realization, a dystopia?
Impressively, this insoluble problem in the current context of achieving a better level of computational intelligence is simple to solve if we focus our efforts on developing people's systems. To achieve this, a communication protocol between the parties is necessary. With management and control systems for moving data between individual Graph DBs and specific groups of people, or even collective rooms of thematic groups segmented by areas of knowledge. In any case, the decisive factor in this case is the centralization and control of the data of the people who will have real ownership of their knowledge bases.
With advanced graph path filters, we work with the concept of complete interaction of events. Each person has multiple events with records of the full extent of the interaction in Graph DBs. Translating this concept, it can be said that it is possible to follow a path of 50 concepts and relate them to a person based on 50 thousand concepts. With base balancing, it is also possible to define the concepts of precision and recall to define the self-constructivity of the bases and equate them with high performance in pattern recognition.
...more
2min
July 09, 2024 health e3 br
health.eco.br
A padronização dos nomes dos termos deve ter os mesmos significados para nomes de rótulos, nomes de relacionamentos, nomes de propriedades e também para o conteúdo preenchido dentro das propriedades.
Integridade: Consistência, confiabilidade e exclusividade das informações de nós e relacionamentos.
Um Sintoma define uma correlação entre efeito; qualificador; anatomia; e fisiologia.
O diagnóstico é um processo de associação a ser registrado no prontuário do paciente que correlaciona: sintomas, exames e doenças.
Existem elementos pós-diagnósticos: prognósticos e tratamento.
A razão para a Doença estar neste nível mais elevado é que ela se correlaciona com “sintomas”, portanto é um padrão reconhecido mais amplo.
A doença (específica) ancora a relação entre sintomas e doenças (classes), assim como a doença (específica) é a própria relação.
As doenças correlacionam sintomas; grupos de doenças (ligação); qualificador; causa (falta de vitamina e ou mineral, vírus e ou bactérias, drogas, estimulantes, substâncias, comportamento); anatomia (localização); e fisiologia (funcionamento).
A doença é, portanto, uma correlação: de uma classe de doença e uma especificação que tem uma localização.
Duas perguntas que devemos responder:
Quais sintomas são necessários para entrar no(s) grupo(s) de doenças?
Que sintomas são necessários para diferenciar uma doença específica do(s) grupo(s) de doença(s)?
O benefício é um sintoma de saúde que define uma correlação entre efeito; qualificador; anatomia; e fisiologia.
Benefícios (efeitos): associados a suplementos, comportamentos, alimentos, vitaminas e minerais.
O desafio é encontrar, através da análise dos termos, a relação entre benefícios, sintomas e doenças.
...more
2min
July 09, 2024 health e3 us
us.health.eco.br
Symptoms and diseases
The standardization of term names should have the same meanings for label names, relationship names, property names, and also for the content filled within the properties.
Integrity: Consistency, reliability, and uniqueness of information from nodes and relationships.
A Symptom defines a correlation between effect; qualifier; anatomy; and physiology.
Diagnosis is an association process to be recorded in the patient's medical record that correlates: symptoms, exams, and diseases.
There are post-diagnostic elements: prognosis and treatment.
The reason Disease is at this higher level is that it correlates with “symptoms”, so it is a broader recognized pattern.
The (specific) disease anchors the relationship between symptoms and diseases (classes), just as the (specific) disease is the relationship itself.
Diseases correlate symptoms; disease groups (link); qualifier; cause (lack of vitamin and/or minerals, viruses and/or bacteria, drugs, stimulants, substances, behavior); anatomy (location); and physiology (functioning).
Disease is therefore a correlation: of a class of disease and a specification that has a location.
Two questions we must answer:
What symptoms are necessary to enter the disease group(s)?
What symptoms are needed to differentiate a specific disease from the disease group(s)?
The benefit is a health symptom that defines a correlation between effect; qualifier; anatomy; and physiology.
Benefits (effects): associated with supplements, behaviors, foods, vitamins, and minerals.
The challenge is to find, through the analysis of terms, the relationship between benefits, symptoms, and diseases.
...more
2min
July 09, 2024 health e2 us
us.health.eco.br
The focus of the Synthesis is to support medical diagnosis, transforming each interaction with the patient into knowledge, a fundamental basis for building a new culture related to AI applied to Digital Health.
Should telemedicine be considered video conferencing patient care? Or reversing the telemedicine process: face-to-face service, after processing the knowledge base algorithms! A medical diagnosis support system must be able to store part of the knowledge applied to medical diagnosis.
The possibility of telemedicine with the support of GDS (Graph Data Science) algorithms allows the development of a health.eco.br Super App to assist in the doctor's diagnosis and/or maintenance of your health.
Organization of the Patient's visual system: SYMPTOMS; ILLNESSES; HEALTH; EMERGENCY; FOOD; SUPPLEMENTS; CONTENT; DIAGNOSIS; RECORD; SERVICE; TREATMENT.
Universe of Concepts: PATIENT; DOCTOR; and PHYSIOLOGICAL.
Abstraction levels: GENERIC (GROUPS); SPECIFIC (SUBGROUPS); and CUSTOM (ELEMENTS).
Hubs: HEALTH and ILLNESS.Specialized content: in many cases, it is organized specifically for each content into topics that establish specific relationships at the base; in other cases, it presents a more specific structure (Etiology [origin and cause]; Pathophysiology; Signs and symptoms; Diagnosis; Treatment; Key points).
The most standardized content: has a pre-defined structural model: (example below) summary/description; Causes; Symptoms; Diagnosis; Treatment; and Additional Information.
Symptom groups are an important connection that ends up being the definition of the symptom itself, and the symptom itself is just a correlation of this defining group with a location and/or associated with other elements intensity, qualifier, location, viral type, bacterial type, etc).
Disease groups are another important link that ends up being the very definition of the disease, with the disease itself being just a correlation of this group with a location and/or associated with other elements (Viruses, etc).
...more
2min
July 09, 2024 health e2 br
health.eco.br
O foco da Síntese é apoiar o diagnóstico médico, transformando cada interação com o paciente em conhecimento, base fundamental para a construção de uma nova cultura relacionada à IA aplicada à Saúde Digital.
A telemedicina deve ser considerada atendimento ao paciente por videoconferência? Ou invertendo o processo da telemedicina: atendimento presencial, após processamento dos algoritmos da base de conhecimento! Um sistema de apoio ao diagnóstico médico deve ter capacidade de armazenar parte do conhecimento aplicado ao diagnóstico médico.
A possibilidade de telemedicina com apoio de algoritmos GDS (Graph Data Science) permite o desenvolvimento de um Super App health.eco.br para auxiliar no diagnóstico do médico e/ou manutenção da sua saúde.
O foco da Síntese é apoiar o diagnóstico médico, transformando cada interação com o paciente em conhecimento, base fundamental para a construção de uma nova cultura relacionada à IA aplicada à Saúde Digital.
A telemedicina deve ser considerada atendimento ao paciente por videoconferência? Ou invertendo o processo da telemedicina: atendimento presencial, após processamento dos algoritmos da base de conhecimento! Um sistema de apoio ao diagnóstico médico deve ter capacidade de armazenar parte do conhecimento aplicado ao diagnóstico médico.
A possibilidade de telemedicina com apoio de algoritmos GDS (Graph Data Science) permite o desenvolvimento de um Super App health.eco.br para auxiliar no diagnóstico do médico e/ou manutenção da sua saúde.
Organização do sistema visual do Paciente: SINTOMAS; DOENÇAS; SAÚDE; EMERGÊNCIA; COMIDA; SUPLEMENTOS; CONTEÚDO; DIAGNÓSTICO; REGISTRO; SERVIÇO; TRATAMENTO.
Universo de Conceitos: PACIENTE; DOUTOR; e FISIOLÓGICO.
Níveis de abstração: GENÉRICO (GRUPOS); ESPECÍFICOS (SUBGRUPOS); e PERSONALIZADO (ELEMENTOS).
Hubs: SAÚDE e DOENÇA.Conteúdo especializado: em muitos casos, é organizado especificamente para cada conteúdo em tópicos que estabelecem relações específicas na base; em outros casos, apresenta estrutura mais específica (Etiologia [origem e causa]; Fisiopatologia; Sinais e sintomas; Diagnóstico; Tratamento; Pontos-chave).
O conteúdo mais padronizado: possui um modelo estrutural pré-definido: (segue exemplo) resumo/descrição; Causas; Sintomas; Diagnóstico; Tratamento; Informações adicionais.
Os grupos de sintomas são uma ligação importante que acaba sendo a própria definição do sintoma, e o sintoma em si é apenas uma correlação desse grupo definidor com uma localização e/ou associado a outros elementos intensidade, qualificador, localização, tipo viral, tipo bacteriano , etc).
Os grupos de doenças são outra ligação importante que acaba sendo a própria definição da doença, sendo a doença em si apenas uma correlação deste grupo com uma localização e/ou associada a outros elementos (Vírus, etc).
...more
2min

FAQs about graph.md:

How many episodes does graph.md have?

The podcast currently has 54 episodes available.