Skip to content
printicon
Main menu hidden.
Published: 09 Mar, 2022

AI technology - on the rise in medicine

FEATURE In a collaboration that spans much of Europe, Jenny Persson and her research team are studying how cancer metastases can be treated in the most effective way. Here, as in the field of medicine in general, modern technology, AI, plays an important role.

Text: Lena Åminne

Jenny Persson, Professor at the Department of Molecular Biology and Chair of the Faculty of Medicine's AI Council, sees great opportunities for AI in the field of medicine. With AI, researchers can manage massive amounts of clinical and research data to conduct innovative basic and transnational research and innovation in an efficient way.

Predicting the spread of cancer

Jenny Persson and her research team in Umeå are studying the mechanisms and actions underlying the drug targets and therapeutic drugs used in clinics to effectively treat cancer metastases. One project, which has received major EU Horizon 2020 funding, focuses on using AI technology and machine learning to predict both the risks of cancer spread and whether cancer will respond to treatment.

The number of new cases of colorectal cancer metastases is increasing

Colorectal cancer is one of the most deadly types of cancer. 20% of patients have already metastasized by the time they are first diagnosed, while 25% will go on to develop cancer metastases. The number of new cases of colorectal cancer metastases is increasing, particularly among men and women younger than 50.

The importance of early detection

Metastases from colorectal cancer often spread to the lymph nodes, liver and lungs. Often metastases are initially asymptomatic, but by the time symptoms appear, the disease is so advanced that there is no effective treatment.

There are not enough reliable biomarkers to predict which patients are at high risk

However, today there are not enough reliable biomarkers to predict which patients are at high risk of metastasis. Therefore, early detection of the disease is essential to develop an effective treatment. Using AI technology, biomarker models can be developed using data from gene mutations and epigenetic changes - chemical modifications of DNA that occur in response to a disease - and other clinical data.

AI-based biomarker models can be used for earlier diagnosis, prognosis and treatment, and are therefore an important tool for the development of precision medicine, i.e. individually tailored treatment. 

Biomarker expert

Researchers look at factors such as genetics, diet and lifestyle.

- As an expert in the fields of cancer biomarkers and molecular pathology, my role in a large EU consortium is to look at gene mutations and genetic changes and various clinical and cancer-related factors, as well as to perform experimental studies with cancer cells and laboratory animal models, says Jenny Persson. We already know about some gene mutations linked to smoking, but we know less about, for example, how diet, bacterial infection and chronic inflammation can contribute to cancer metastasis and how tumours respond to treatment.

- We are now looking at many factors that could be models for predicting which patients are at risk of cancer progression and who do not respond to treatment. In this way, we can identify patients at high risk at the time of diagnosis and better predict which patients are more likely to develop malignant metastases.

This major EU project involves a computer science department in Italy and Germany, which has already built expertise in AI. It has built an infrastructure in the form of a database on different protein alterations, allowing it to classify and sort samples from thousands of individuals in order to identify risk groups.

AI in a wide range of areas

Jenny Persson sees great opportunities for AI in medical research, and in Umeå the interest in and need for AI is growing rapidly. Researchers in infection biology started using AI early on and have collaborated a lot with Uppsala University, especially when it comes to cancer.

But AI can be used in a wide range of medical fields. From helping the elderly to function in their daily lives, to creating flow models for the connection between nerves and the brain, or allowing pathologists to use automated programs to read samples.

Investing in AI - FACT

The Faculty of Medicine is making a long-term investment in AI in education programs to train the workforce of the future and in research by strengthening the application of AI technology broadly in medical research and innovation. Jenny Persson chairs and leads the work of the Faculty's AI Council.

The AI Council works to develop interdisciplinary collaboration on AI within the faculty and within universities, as well as with researchers around the world. This is done through various research activities, trainings and workshops such as seminar series, an AI day organized together with IceLab, a very productive Hackathon and an AI beginners' course open to everyone at Umeå University, but also to researchers and students from other universities.

The widespread adoption of AI is one of the main objectives of the Faculty of Medicine. Umeå University's newly formed transdisciplinary AI center, which includes all faculties, will be an important engine for AI in both research and education.