With the development of artificial intelligence (AI), new applications are emerging based on this technology. Computer vision is one such application. AI can alleviate human tasks in various ways, allowing people to focus on other responsibilities, and in some cases, the technology is both faster and more reliable.
Ida Häggström, Associate Professor in Computer Vision and Medical Image Analysis at the Department of Electrical Engineering, Chalmers, works on medical image analysis using machine learning techniques.
“This means that I train computers to identify, for example, cancer. Computers perform tasks through algorithms, in this case, interpreting medical images.”
Using deep learning, Ida Häggström trains computers by presenting large amounts of data while providing the correct answers. This way, the computer learns to recognize patterns in images to read, for instance, X-rays or nuclear medicine images.
“For example, I have studied over 17,000 images from more than 5,000 lymphoma patients and created a learning system where a computer has been trained to identify signs of cancer in the lymphatic system.”
Such a system can ease the workload for radiologists, provide a second opinion, or prioritize patients who need attention most urgently.
Ida Häggström does not believe that computer vision will completely replace radiologists, at least not in the near future.
“I don't think human expertise will be entirely replaced anytime soon, but certain tasks may be.”
Another advantage Ida Häggström sees with computer vision is that it contributes to increased equality in healthcare.
“To make healthcare more equal, a digital radiologist in the form of a trained AI system that can operate anywhere fits well. So, regardless of the hospital, patients have access to the same expertise and can have their images reviewed within reasonable time.”
Even for rare diseases, an AI system may be better suited. A proper evaluation is based on a radiologist having seen thousands of images, but with rare diseases, one may have only seen a few images. By building an AI model, we can give it access to more information.
Ida Häggström collaborates with the Sahlgrenska Academy and Sahlgrenska University Hospital in various research projects, focusing on medical image analysis using machine learning techniques.
For more information, contact:
- Associate Professor, Signal Processing and Biomedical Engineering, Electrical Engineering