.JPG)
Named PICS, AI aims to identify possible diagnoses of diseases by suggesting protocols and prescriptions by Machine Learning
The market for virtual assistants and chatbots is constantly growing. According to a recent Spiceworks report, later this year, 40% of midsize companies expect to implement one or more intelligent assistants or AI-powered chatbots on their devices.
Attuned to trends, Pixeon, a leading healthcare technology company in Brazil, presented, at a recent event in the healthcare sector, its prototype and proof of concept of the Artificial Intelligence project, applied to clinical diagnoses. The AI called PICS (Pixeon Clinical Health Intelligence), aims to identify in natural language, text or voice, symptoms presented and identify possible diagnoses of diseases by probability, suggesting standardized protocols and prescriptions understood by the machine learning process (Machine Learning) .
PICS is prepared to understand narrated information considering a patient consultation context, that is, to hear and understand the speech during the consultation process, separating relevant information regarding symptoms and other relevant clinical characteristics.
“The PICS project aims to teach/build on the platform the ability to make diagnoses, give suggestions (insights) of protocols, exams, in short, the creation of a 'consultant-medical assistant', for health professionals”, highlights Armando Buchina, CEO of Pixeon.
Machine learning: learning process
Taking into account a scenario where the experience of the health professional and the conditions of both the doctor and the patient are highly relevant factors for the quality of the diagnosis, PICS aims to assist in the process of separating diseases and/or conditions. candidates, considering an event compared with historical data, international disease tables, own knowledge base, etc., enhanced by the use of machine learning and natural language processing resources.
Also, according to Buchina, speaking about the future of the project, “one of the goals is to teach PICS for approximately one year so that it has adequate accuracy and confidence in its information, since the Machine Learning will be carried out by medical specialty”.
Over the next five years, the company plans to inject R$ 83 million into its entire product portfolio, targeting innovations, process and functionality improvements, all designed to improve the patient's journey with its solutions, from scheduling an exam to delivering the report.
Attuned to trends, Pixeon, a leading healthcare technology company in Brazil, presented, at a recent event in the healthcare sector, its prototype and proof of concept of the Artificial Intelligence project, applied to clinical diagnoses. The AI called PICS (Pixeon Clinical Health Intelligence), aims to identify in natural language, text or voice, symptoms presented and identify possible diagnoses of diseases by probability, suggesting standardized protocols and prescriptions understood by the machine learning process (Machine Learning) .
PICS is prepared to understand narrated information considering a patient consultation context, that is, to hear and understand the speech during the consultation process, separating relevant information regarding symptoms and other relevant clinical characteristics.
“The PICS project aims to teach/build on the platform the ability to make diagnoses, give suggestions (insights) of protocols, exams, in short, the creation of a 'consultant-medical assistant', for health professionals”, highlights Armando Buchina, CEO of Pixeon.
Machine learning: learning process
Taking into account a scenario where the experience of the health professional and the conditions of both the doctor and the patient are highly relevant factors for the quality of the diagnosis, PICS aims to assist in the process of separating diseases and/or conditions. candidates, considering an event compared with historical data, international disease tables, own knowledge base, etc., enhanced by the use of machine learning and natural language processing resources.
Also, according to Buchina, speaking about the future of the project, “one of the goals is to teach PICS for approximately one year so that it has adequate accuracy and confidence in its information, since the Machine Learning will be carried out by medical specialty”.
Over the next five years, the company plans to inject R$ 83 million into its entire product portfolio, targeting innovations, process and functionality improvements, all designed to improve the patient's journey with its solutions, from scheduling an exam to delivering the report.
EN 
PT 
