Photo of Håkan Gustafsson

Håkan Gustafsson

Adjunct Associate Professor, Docent

Håkan Gustafsson's research is conducted in medical radiation physics and a central part of his work is the validation of AI systems for clinical use. This involves the development and application of structured, scientifically based methods to ensure that AI models function in a robust, reproducible and clinically meaningful way in real healthcare environments. He also participates in national initiatives to define how AI can be introduced into healthcare in a safe and transparent way.

In addition to his academic role, Håkan Gustafsson is employed as Head of Nuclear Specialist and Project Managers at Studsvik. There he strives to translate research results into industrial applications and commercial value, including by using AI to improve analysis and decision support in nuclear applications as well as research and development related to medical isotopes.

The research is conducted at Linköping University. Gustafsson holds a Master’s degree in physics, received his PhD in radiation physics in 2008 and Docent degree in radiation physics in 2013.

LinkedIn

ResearchWeb

Publications

2025

Pantelis Gialias, Maria Kristoffersen Wiberg, Anne-Kathrin Brehl, Tomas Bjerner, Håkan Gustafsson (2025) The use of artificial intelligence (AI) to safely reduce the workload of breast cancer screening: a retrospective simulation study Acta Radiologica (Article in journal) Continue to DOI
Johan Lyth, Pantelis Gialias, Magnus Husberg, Lars Bernfort, Tomas Bjerner, Maria Kristoffersen Wiberg, Lars-Åke Levin, Håkan Gustafsson (2025) Results from a Swedish model-based analysis of the cost-effectiveness of AI-assisted digital mammography European Radiology (Article in journal) Continue to DOI
Haiko Schurz, Klara Solander, Davida Astrom, Fernando Cossio, Taeyang Choi, Magnus Dustler, Claes Lundström, Håkan Gustafsson, Sophia Zackrisson, Fredrik Strand (2025) Simulating mismatch between calibration and target population in AI for mammography the retrospective VAIB study npj Digital Medicine, Vol. 8, Article 259 (Article in journal) Continue to DOI

2023

Fernando Cossío, Haiko Schurz, Mathias Engström, Carl Barck-Holst, Apostolia Tsirikoglou, Claes Lundström, Håkan Gustafsson, Kevin Smith, Sophia Zackrisson, Fredrik Strand (2023) VAI-B: a multicenter platform for the external validation of artificial intelligence algorithms in breast imaging Journal of Medical Imaging, Vol. 10 (Article in journal) Continue to DOI
Håkan Gustafsson (2023) Så kan AI valideras för klinisk implementering Läkartidningen, p. 27-31, Article 120:23065 (Article in journal)

Organisation