Can AI mend your broken heart?
Cardiovascular disease is the most common cause of death in Sweden. To prevent and discover cardiovascular disease, and to anticipate the continuous trajectory and care needs for those affected, researchers and doctors in the project AIR Lund are working with artificial intelligence – AI – and self-learning computer programmes.
"I see great possibilities with this project. We are hoping to contribute to a better base for decisions for healthcare personnel, and a safer and more thought-out journey through the healthcare system for those affected by cardiovascular disease.”
Mattias Ohlsson, Professor of Information Technology
Through artificial intelligence and machine learning, we can train computer programmes to perform a certain task. Self-learning computer programmes have been proven useful in many different medical areas, for example to recognise complicated disease patterns, interpret x-rays and make different kinds of risk assessments. Now, researchers are investigating how we can use the technology to save lives, and to improve the quality of life for those affected by cardiovascular disease. Mattias Ohlsson, Professor of Information Technology, has long seen the possibilities in using AI in healthcare:
“I have always been interested in machine learning and how to make predictions, for example how to predict a healthcare trajectory. I was quick to specialise in healthcare data. I see great possibilities with this project. We are hoping to contribute to a better base for decisions for healthcare personnel, and a safer and more thought-out journey through the healthcare system for those affected by cardiovascular disease.”
Prevention, diagnosis, and prognosis
The project has three different focus areas: prevention, diagnosis, and prognosis. The preventive work is about finding patterns and combinations of risk factors to predict cardiovascular disease, and to use preventive actions to avoid illness.
“You can use AI to predict processes and situations that we would like to be ready for, or avoid”; says Mattias Ohlsson and continues:
“We are also working with diagnostics. Here we use a self-learning computer programme to improve the diagnostics and suggest how to handle patients with suspected disease or other serious disease.”
The project's third focus is to establish prognosis to find patterns in life events, either medical or socioeconomical. The events, which need to be stored in registers, can together predict the risks of disease, the course of disease and the care needs for those affected by cardiovascular disease.
“In some cases, genetic factors, lifestyle and surrounding factors are used as input to the AI systems. The conclusions that the systems make can contribute to measures being taken earlier, and to a higher precision in diagnostics and treatment for the individual”, says Mattias Ohlsson.
Preventative and individualised care
The research project also studies risks in integrity and ethics connected to using artificial intelligence and self-learning computer programmes as a base for decision making in healthcare. For example, in judicial matters concerning responsibility and transparency, or in cases of ethical dilemmas and risk of misleading or discriminating results of analysis and suggested decisions.
“We want the self-learning computer programmes to improve their ability to explain their findings and risk estimates. In that way, we hope to solve some of the dilemmas that could possibly occur when using self-learning computer programmes in healthcare”, says Mattias Ohlsson.
One of the global goals for sustainability in the UN Agenda 2030 is about good health and well-being and has the target to reduce mortality from non-communicable diseases and promote mental health. The target can be applied to the research project, which hopes to improve the quality and effectivity of care for those who are already affected by cardiovascular disease and prevent others from falling ill.
“Our hope is to contribute to multidisciplinary research within the area and at the same time pave the way for more AI research within healthcare. The goal is more preventative and individualised care, which will benefit both the individual and society”, Mattias Ohlsson finishes.
Text: Christa Amnell and Linnéa Andersson
Photo: Ida Fridvall and Dan Bergmark
About the project
AIR Lund is active between 2019 and 2023 and is coordinated by Lund University. The Project Manager is Jonas Björk, Professor at Lund University. At Halmstad University, the project is managed by Mattias Ohlsson, Professor of Information Technology, and Ali Amirahmadi, Doctoral student in Machine Learning. The project includes Halmstad University, Lund University and Region Halland. The project is financed by the Swedish Research Council.
AIR Lund is part of the effort on information driven care that is conducted at Halmstad University together with the public and private sectors. The research profile CAISR Health, the industrial research centre Heath Data Centre and the innovation centre Leap for Life are part of this effort. The project is also part of Halmstad University’s focus area Health Innovation.
No valid user selected.