In Brazil, approximately 60 million people suffer from or acquire diseases daily, yet the time required for blood count diagnoses remains lengthy. This delay often worsens conditions, reduces patients' quality of life, and, in severe cases, results in irreversible harm or death. Artificial intelligence (AI) offers a potential solution to accelerate diagnostic reporting. This project focuses on developing AI-driven software to optimize blood count analysis and medical diagnoses. The methodology involved three stages. First, in "Medical Pattern," key variables linked to diseases detectable via blood counts were identified. These included conditions like diabetes, anemia, leukemia, dengue, and malaria, with significant indicators such as hemoglobin, leukocytes, platelets, glucose, and hormones. The second phase centered on theoretical and practical software development using artificial neural networks and Python-based regression models. The third phase involved testing with real datasets from 1,227 anonymized bl