Academic Background

Bachelor’s Degree in Information Systems for Management

Institution: ESTG - Polytechnic of Porto

Duration: Sep 2019 – Feb 2023

Final Grade: 13/20

Course Overview: The Bachelor’s degree in Information Systems for Management offers a comprehensive foundation combining business management and information technologies. The program equips students with both technical and scientific skills to work in roles at the intersection of IT and business, enabling the design and management of information systems to support organizational processes.

Projects & Activities:

  • “From Linear to Circular Ideas” – Participated in an innovation competition focused on sustainable solutions.

  • “Find It” – Academic project creating a collaborative repository for knowledge sharing between students and professors. Applied concepts included software development, UI design, and UML modeling.

  • Bachelor’s Degree Document: Intership report Churn Prediction (PDF)

Key Skills: UI Design, Software Development, Business Process Management, ERP Systems, Data Science

Master’s Degree in Artificial Intelligence Engineering

Institution: ISEP - Porto School of Engineering

Duration: Sep 2023 – Jul 2025

Final Grade: 16/20

Course Overview: The Master’s in Artificial Intelligence Engineering provides advanced training in AI, machine learning, and data engineering, with a strong focus on practical applications in real-world scenarios. Students develop expertise in expert systems, natural language processing, and decision support systems, preparing them to create innovative AI solutions across industries, particularly in healthcare.

Projects & Challenges

For a detailed explanation of all my challenges, you can check my MSc Challenges Blog.

  • Challenge 1 – AnxietyDetect AI: Developed a system using Drools and Prolog to identify anxiety disorders, implementing logic rules for accuracy and reliability.

  • Challenge 2 – trIAge: AI system to classify patients by wristband color and assign hospitals based on distance and waiting times, optimizing patient distribution.

  • Challenge 3 – DermaChat: Platform diagnosing skin issues through images or text, using machine learning and NLP. Includes a chatbot for user guidance.

  • Challenge 4 – Multi-Agent Hospital Management System: Autonomous communication between ambulances, vehicles, pedestrians, and hospitals to optimize emergency resource allocation.

Key Skills: Expert Systems, Data Engineering, Machine Learning, Natural Language Processing (NLP), Decision Support Systems (DSS)

Master’s Thesis

Title: Master’s Thesis Project – Application of Advanced Artificial Intelligence Techniques for the Diagnosis of Gastrointestinal Diseases

Duration: Nov 2024 – Jul 2025

Final Grade: 19/20

Overview: As part of my Master’s degree, I conducted in-depth research focused on the application of advanced artificial intelligence techniques for the diagnosis of gastrointestinal diseases. This work was carried out in partnership with GECAD (Knowledge Engineering and Decision Support Group) within the scope of the international PROFIT project, which aims to optimize hospital processes through digitalization and AI implementation.

My thesis explores the integration of computer vision and deep learning methods for colonoscopy analysis, a fundamental procedure for early detection and treatment of colorectal lesions. Colonoscopy, although the gold standard, heavily depends on the physician’s expertise, which can lead to variability in diagnoses and delayed detection. Additionally, invasive biopsies required for confirmation involve risks and patient discomfort.

To overcome these challenges, I developed a system based on Convolutional Neural Networks (CNNs) and segmentation models, including ResNet, DenseNet, EfficientNet, and Inception, enabling more accurate and consistent identification and classification of gastrointestinal lesions. Results show that hybrid models, which combine multiple architectures, outperform models based solely on transfer learning, with the best performance achieved by the ResNet + EfficientNet + DenseNet hybrid model, reaching an accuracy of 86.67%. This hybrid approach shows promise in supporting clinical decision-making, reducing diagnostic variability, and minimizing the need for invasive biopsies.

Key Skills: Python, Deep Learning, CSS, Explainable AI (XAI), Transfer Learning