What is it about?
Brain tumors are tricky to diagnose and treat due to the brain’s complexity. Detecting them quickly improves patient chances. The study explores using Artificial Intelligence (AI), specifically deep learning, to save time and resources in finding tumors from imaging. We tested an AI model on MRI scans and achieved about 99% accuracy. We also emphasize the importance of explaining and being transparent about how the AI works to ensure human control and safety in the diagnostic process.
Featured Image
Photo by Milad Fakurian on Unsplash
Why is it important?
This paper presents a new AI model that uses MRI scans of the brains and tabular data to detect brain tumours. Such multi-modal models are new and seem promising. Furthermore, the paper also explores the need for explainability and human control in an AI system, especially when created to support medical diagnosis.
Perspectives
Working on this manuscript was really interesting. Not only we were able to reach good accuracy in brain tumor detection (potentially helping in providing a tool in the diagnostic process), but we also posed some interesting questions on the use of AI in medicine. We highlighted the need for explainability, and for leaving control to physicians, while also creating a strong basis for future work that will explore how to ensure these properties.
Andrea Esposito
Universita degli Studi di Bari Aldo Moro
Read the Original
This page is a summary of: Detecting Brain Tumors Through Multimodal Neural Networks, January 2024, Scitepress,
DOI: 10.5220/0012608600003654.
You can read the full text:
Contributors
The following have contributed to this page
