Every year, approximately 35,000 Danes are diagnosed with cancer, and most of these patients must undergo surgery to have the cancer removed. Unfortunately, in up to half of all cases involving certain types of cancer, it is not possible to remove all cancerous tissue with certainty during the first surgery. As a result, many patients must undergo repeat surgeries and intensive follow-up treatments such as radiation therapy and chemotherapy due to the risk of recurrence. This has costs for the individual and for society.
Precision through advanced imaging and AI
With a new Danish project titled AIMICS – An AI-driven Multidimensional Imaging Platform to Improve Cancer Surgery – researchers aim to develop a new technology with a simple yet ambitious goal: to ensure that all cancerous tissue is removed during the patient’s first surgery. The project combines technologies such as 3D ultrasound, mid-infrared imaging, micro-CT, and artificial intelligence to provide surgeons with a more precise basis for decision-making at the operating table.
The technology can be used during cancer surgery in the operating room to determine whether all the cancer has been removed. Today, it takes a week to 14 days to analyze the tissue that has been removed. But with the project’s technology, an analysis can be performed right at the operating table, allowing surgeons to immediately determine whether more needs to be removed or if the surgery can be completed.
“For patients, it is absolutely crucial that we can cure them with the first operation without the need for double treatment with many long-term consequences. With this technology, we hope that the surgeon can have a much more precise basis for decision-making during the operation and thus increase the likelihood that all the cancer is removed immediately. This can both improve survival rates and enhance the quality of life for cancer patients,” says Tobias Todsen, clinical professor at DTU and senior physician at the Department of Otolaryngology and Audiology at Rigshospitalet.
A key prerequisite for the project is access to high-quality imaging data combined with biological and medical knowledge about cancer.
“AI is a central foundation for the project. We can only carry out the project using AI-based methods, which require high-quality data - especially imaging data. A major part of the project will be collecting, annotating, and combining the various datasets to train precise AI models that can be used directly in clinical practice. Our ambition is for patients to be scanned using ultrasound and mid-infrared spectroscopy, and the images from these scans will provide an accurate assessment of whether all cancer has been removed within minutes,” says Anders Dahl, professor and section head of Visual Computing at DTU Compute.
The technology is initially being tested on patients with oral or skin cancer, but its potential extends far beyond to other types of cancer where surgery is performed. If the project succeeds, it will mean higher survival rates, better quality of life for patients, and significant savings for the healthcare system. And things look promising. Preliminary clinical trials show that the new technology can improve the surgical success rate by up to 93 percent.
Cross-disciplinary collaboration is crucial
Behind the project is a partnership consisting of Rigshospitalet, DTU, and the company 3Sonic, of which Tobias Todsen is a co-founder.
“We are living in a time that demands major changes in the healthcare system of the future. This means it is crucial that we break down silos and think innovatively. If we are to have a sustainable healthcare system in 20 years, we must become far better at collaborating across disciplines—including between doctors and engineersand this project is a prime example of that,” says Tobias Todsen.
The partners contribute different areas of expertise:
- Rigshospitalet contributes clinical expertise.
- DTU provides technical expertise.
- 3Sonic ensures that the technology can be scaled up and disseminated to benefit as many people as possible.
The project is being developed as part of TUH, Technical University Hospital of Greater Copenhagen - a collaboration between the Capital Region, the Region of Zealand, and DTU, established to bring clinical practice and engineering science closer together for the benefit of patients.
