Bioengineering

About the Domain

Bioengineering, a rapidly evolving scientific domain combining engineering and life sciences, integrates core engineering principles, practices and technologies applied to healthcare, biology, environmental and health sciences, enabling engineering solutions to challenges in these domains. At INESC TEC, we foster the advancement of scientific knowledge in bioengineering through fundamental and applied research, advanced training and innovation – developing, for instance, innovative biosensors, wearable medical devices for digital health and well-being monitoring, biorobotics, human-computer interfaces, and new approaches to biomedical image analysis for cancer diagnosis.

In this field, we bring together competences that allow us to develop innovative technologies for diagnosis, monitoring, and treatment, promoting more precise, personalised, and accessible healthcare. Some of these key areas of expertise include:

Neuroengineering

Medical Image Computing

Biomedical Sensors

Biosignal Processing

Research Challenges

Research challenges in Bioengineering reflect the interdisciplinary and innovative nature of this field, focusing on the creation of technological solutions that address emerging needs in healthcare, biology, and the environment. From the development of advanced biosensors for several contexts to the analysis of complex medical images and biologically inspired robotics, our research aims to bring technology ever closer to the human being, promoting health, well-being, and sustainability.

Main Achievements

Our research in Bioengineering led to breakthroughs at the intersection of AI, bionics, biometrics, and medical image analysis. We strive to make technology more accurate, explainable, and useful for healthcare professionals and researchers, and our developments are internationally acknowledged. Some of our main achievements include:

Explainability in medical image analysis

We developed a method that improves how AI identifies patterns in medical images, leading to more consistent results, which are easier to interpret. Applied to skin lesion detection, this method has proven more effective than traditional AI models. Learn more here.

Greater accuracy in facial recognition confidence

We created a new approach to more accurately determine whether two facial images belong to the same person. Our method makes comparisons more reliable and interpretable, helping to enhance both security and accuracy in facial recognition systems. More information in this publication.

Explainable AI to support image analysis

We developed tools that support healthcare professionals in diagnosis and examination interpretation by reducing bias and increasing confidence in AI decisions. They provide explanations that safeguard patient privacy, evaluate and compare different explanations to ensure transparent and useful AI, and present information in a way that facilitates understanding of the results. Learn more: here, here, and here.