Artificiell intelligens och thoraxkirurgisk vetenskap (AICTS)
Artificial Intelligence in Cardiothoracic Science was funded 2006. Traditionally, medical decisions are based on the combined strength of clinical facts and the experience of the clinician. A growing understanding of the molecular, genetic and biochemical basis of diseases have greatly increased the degree of complexity in medical decision-making. To identify risk factors in medical registers, a non-linear method such as artificial neural network (ANN) may better describe the correlations between different health risk factors. A research area with this high degree of complexity; to identify, optimize and simulate outcomes is strongly dependent on scientific computing using large scalable high-speed computing systems.
The general aim for our research group, using large and unique global medical databases, to bring the use of artificial neural networks (ANNs) and simulation techniques in risk stratification research a step further, in achieve a higher quality of treatment and improve the outcome for patients with cardiothoracic diseases.
Machine learning compared with rule-in/rule-out algorithms and logistic regression to predict acute myocardial infarction based on troponin T concentrationsAnders Björkelund, Mattias Ohlsson, Jakob Lundager Forberg, Arash Mokhtari, Pontus Olsson de Capretz, Ulf Ekelund & Jonas Björk, 2021, I: Journal of the American college of emergency physicians open. 2, 2, e12363.
Forskningsoutput: Tidskriftsbidrag › Artikel i vetenskaplig tidskrift
Visa alla (81)