Den här sidan är utskriven från Högskolan i Halmstads webbplats (www.hh.se). Texten uppdaterades senast den 2018-08-17. Besök webbplatsen om du vill vara säker på att läsa den senaste versionen.

AI for predictive healthcare

Healthcare is facing major challenges; the current organization is not adapted to what will be required in the future. Changes in demographics are expected to create a greater demand for care and the discrepancy between the number of people in care-needing ages and the number of people in the labor force increases rapidly. In parallel to this, the disease panorama is changing, as well as the ability of the medical care to diagnose and treat diseases. New methods and drugs that reach the market tend to be advanced and costly. The healthcare system will simply put not fulfill its mission if it does not work differently from today.

In that perspective, health technology becomes central. There is considerable expectation that new technologies will fill the increasing gap between needs and resources. Large international investments are made in this area. Not least aimed at artificial intelligence (AI) since its methods are well suited for tasks that constitute the working days of healthcare workers: prediction and pattern recognition (read risk assessment and diagnostics). Neural networks/machine learning and other AI methods are particularly suitable for contributing to increased understanding of both complex organizational and biological systems. Healthcare needs to be part of the rest of society, where intelligent systems are developing and applying rapidly.

As a contribution to this development, Halmstad University, through CAISR and Region Halland, has established a research collaboration that enables real life studies where the conditions are optimized by having research staff employed with both parties. This allows researchers to work in the IT environment where the solutions will be applied. An important work during 2017 was to build the structure of cooperation by establishing contact routes, collaborative groups and mutual understanding of each other’s conditions. This has been done, among other things, through joint initiatives such as predictive studies of emergency care patients, and a new project aimed at understanding how we can help patients achieve prescribed treatment (iMedA – a project funded by the Swedish Innovation Agency), intelligent home development work and collaboration within the Halmstad Health Technology Center (HCH). We have also begun exploring the opportunities for dynamic, need-driven duty roster planning in healthcare units based on predicted needs. Halland has excellent conditions for developing individualized care since Region Halland has developed the necessary IT support and systems for analysis and research, covering the patient’s entire journey through the healthcare system.

During 2017 Halmstad University and Region Halland united and mobilized around going for a Vinnväxt initiative (a program funded by the Swedish Innovation Agency). The initiative as such aims at establishing a strong and sustainable innovation environment within Health Innovation in Halland, and further increasing integration between university, health and business.

Knowledge transfer between different applications has been central. Such disparate fields as intelligent vehicles and clinical decision support have much in common in their methodology, and CAISR strives to exploit these synergies. However, external knowledge is very valuable and Region Halland’s collaboration with researchers affiliated with Harvard Medical School has contributed a lot to the knowledge in ongoing projects.

A development area that needs to be addressed is the integration of data from different sources. The development on the sensor side is swift, but it is necessary to integrate the information from many sources to get the overall picture of a person’s or organization’s current state. In order to meet the need for a more comprehensive picture, we will continue the work on infrastructure initiatives. The continued development of HCH’s Test Environment Halland and the Halmstad Intelligent Home (an initiative within CAISR) is central in this regard. Investing in test beds where companies, developers and public actors can meet is important to support the development and introduction of new digital products and services.

We address the rapid digitization of society in a recently submitted application to the European Regional Development Fund (ERUF). The digitization leads to increased expectations for personalized solutions (services) with high quality and good accessibility. The rapid development we see in AI and the pressed resources means that a paradigm shift will take place in healthcare. When intelligent systems take over in areas such as prevention, detection, diagnosis and treatment, the flow in healthcare will change, as well as the professional roles. The way we prioritize care will change and home healthcare will develop and work methods will be streamlined. All signs indicate that more and more people will be cared for in their homes. Prevention services increase and people will gain (or will be expected to gain) greater influence over their health and care. The overall objective of our proposed project is to enable industry and the public sector to deliver the most competitive products and services for public and private care at home, with the individual at the center. A key part is to create a technical infrastructure that enables interoperability between different products and services, as well as enabling a transfer of information in the healthcare system. Among other things, the technical infrastructure will be evaluated and implemented in a private residential area – a test bed.

With these efforts we build a unique knowledge that gives us an opportunity to understand the complexity of healthcare and to prepare ourselves to understand the “new future”. Connecting real-world needs and new technological advances, not least with AI, is a prerequisite for meeting the challenges that face healthcare.

Updated 2018-08-17