Hello, this is the resume of graph data science specialist Daniel Cotrim. We are very pleased that you are interested in learning a little more about Daniel Cotrim's professional profile for potential opportunities to work together with graph md, on a project with your institution, integrating medicine with AI and GDS (Graph Data Science). He has been a programmer since he was fifteen, in Java for over 27 years, and in Python for over 10 years. His focus has been on Artificial Intelligence since 2007, when he began his master's degree in computing, and later, in 2010, his doctorate in computational neuroscience at the Faculty of Medicine of the Catholic University, located in Porto Alegre, in the state of Rio Grande do Sul, in the extreme south of Brazil. It is worth noting that the work he has done is deeply related to AI in the health area since that time. After all these years linked to Computing applied to Health, graph md was developed, creating an innovative telemedicine architecture in GDS, using the Neo4j graph database. With this new technology, it was possible to include all the world's classified diseases and the main symptoms from PubMed, which is an important scientific information bank in medicine. More than a thousand articles were studied and analyzed to identify concepts. The graph md has fifty thousand terms for symptoms and diseases; and four hundred thousand relationships between nodes. This model can be accessed at github.com/graphmd. According to the opinion of leading experts: It would not be possible to build the model with these characteristics in a relational database. In this Complex System, it is possible to demonstrate the viability of the idea of ​​proposing an intelligence storage solution based on the construction of personal patient databases. Soon, a person will have a system larger than that of a large company today.

Models are considered dependent data. Therefore, the proposal was to create an architecture that would significantly reduce the volume of relationships in the knowledge base. To solve the issue of isolation, the concept of knowledge synthesis bases was created, which gives the project its name. Zero-knowledge bases receive patient data and diagnostic maps so that intelligent agents can validate the most likely hypotheses. At this point, all AI is already pre-stored, in tags and relationships, within the synthesis database with their respective values ​​pre-calculated by AI algorithms that have been validated by medical specialists. This is where the concept of knowledge comes in to make the solution profitable. Patients do not pay for the graph.md app, but rather for the knowledge applied to their health needs.

Because, as we know, the analysis of a patient's diagnosis can have different levels of abstraction for different medical specialists! Through this innovative architecture, we built the super app graph.md in Java Spring Data Neo4J on the platform integrated with spring data high-performance environment, creating dynamic forms and navigating between fifty thousand terms with a maximum of fifty choices, or assignments, made through questions with guided or suggested answers. Thus allowing navigation in the graph md in an intelligent way. This interaction between patients and medical specialists to define a diagnostic hypothesis can be synchronous or asynchronous. The magic happens when we identify the initial context, that is, all patients who are approximately the same as the patient being analyzed who already have a diagnosis assigned in the knowledge base. By entering this group, we can check the complementary concepts of patients who have already been diagnosed, taking the shortest path within the graph to reach the goal. More sophisticated queries to find the classificatory terms can be used at this point. Concluding the process, we show the classificatory terms, or symptoms, for the target diagnoses. The fundamental issue is to reverse the concept of telemedicine: Dear Patient P! Analyzing your data in graph.md, we would like to schedule an in-person appointment with Specialist Physician E to avoid (seek to reduce) the risk of a certain disease D occurring. The main challenge is that health is not explicitly defined in the diagnostic process. It is considered only as a complement to the disease.

People mistakenly and invariably think: if I'm not sick, then I'm healthy! But the light at the end of the tunnel is that by studying symptoms throughout life, we can observe that the patient's physiology is related to both universes: Health and Disease. By observing a patient under the magnifying glass of a knowledge base in Graph DB, we can predict that if the physiology is in perfect working order, the disease is a certain safe degree away from the patient. If we eventually identify early warning signs in certain physiological dysfunctions, we can intervene to propose a return to this desired state even before the disease affects the patient.

The work carried out by graph.md in the health sector shows that in other areas, the modeling of a complex system applied to AI must observe the occurrence flows of the relationships between people's events and the companies' needs maps. Studying and developing highly complex systems, such as those that support medical diagnosis, for practically my entire life, the objective is to contribute, as a member of the graph md team, developing solutions in Graph Data Science, in this scenario of the beginning of the era of Artificial Intelligence directly linked to people's daily lives.