Banca de QUALIFICAÇÃO: JEAN PAES LANDIM DE LUCENA

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : JEAN PAES LANDIM DE LUCENA
DATE: 31/07/2026
TIME: 08:30
LOCAL: Google Meet
TITLE:

Large Language Models and Risk Analysis in Healthcare: An Approach to Antimicrobial Resistance

KEY WORDS:

Large Language Models; Healthcare Stewardship; Antimicrobial Resistance; Risk Analysis

PAGES: 50
BIG AREA: Ciências Biológicas
AREA: Biologia Geral
SUMMARY:

The World Health Organization has drawn attention to the increase in mortality from health problems resulting from antimicrobial resistance. Estimates from the organization indicate that by around 2050, approximately 10 million people will die annually due to the ineffectiveness of antimicrobials. Studies suggest the inappropriate prescription of antibiotics as the core of this problem, something that stems mainly from the overload on health services and their workers who have to deal with a large amount of clinical data from their patients and constantly changing clinical protocols. In this sense, this work proposes a chatbot infrastructure based on the Large Language Models combined with the embeddings model. This framework is to be implemented as API for a Clinical Decision Support tool capable of suggesting therapeutic approaches in the context of infectious diseases. It uses the set of protocols from the Brazilian Ministry of Health as a knowledge base for Retrieval-Augmented Generation, in order to suggest adequate and up to date recommendations. The proposed approach aims to support evidence-based therapeutic decision-making and improve access to updated clinical guidance. Preliminary results indicate accurate retrieval of protocols, demonstrating the potential of the tool to assist healthcare professionals in clinical practice.

COMMITTEE MEMBERS:
Presidente - 3884005 - PATRICK CESAR ALVES TERREMATTE
Externo ao Programa - 3492344 - SILVAN FERREIRA DA SILVA JUNIOR - nullInterno - 3063244 - TETSU SAKAMOTO