I’m an orthopaedic surgeon at Danderyd’s Hospital, PhD and researcher at the Karolinska Institute. I am also the organiser for the Stockholm R-User Group, we are 400-500 a group of users that meet 4-6 times per year to discuss statistics, machine learning, and visualization under the R umbrella. I also have a research blog, G-forge, and a Github repository with my R-packages.
My main research interests are currently epidemiology and machine learning. I finished my thesis in May 2014 titled “Evaluation of patient related factors influencing outcomes after total hip replacement”. In collaboration with the Swedish Hip Arthroplasty Register we looked at basic factors, such as age, sex, and comorbidities, and how they influence total hip replacement outcomes. The work spurred me to look at advanced statistics and especially how to convey these graphically so that anyone can grasp the fundamentals of the results.
Apart from epidemiology I am interested in machine learning, especially neural networks and deep learning, and how to apply these methods in order to improve orthopaedic health-care. Machine learning has frequently been inaccessible to health care as it often requires massive amounts of data. With the digitalization of most of our tools and advent of new data collection technologies the data is now within our grasp. We are currently working with KTH data science department; a collaboration that we hope will lead to improved patient-safety, health-care efficiency, and patient outcomes.