Ph.d. defense by Jorne Biccler
Department of Clinical Medicine, Aalborg University and Department of Haematology, Aalborg University Hospital are pleased to invite to PhD defense by MSc, Jorne Biccler, who will defend the thesis entitled: Statistical and machine learning methods for the dynamic prediction of prognosis in haematological malignancies
15.03.2019 kl. 14.00 - 17.00
Time and place
14:00, Friday 15 March, 2019
In the auditorium, Forskningens Hus, Aalborg University Hospital
Tarec C. El-Galaly, Professor, Aalborg University
Martin Bøgsted, Professor, Aalborg University
Mats Jerkemand, Professor, Lund University
Peter de Nully Brown, MD, Copenhagen University Hospital
Søren Paaske Johnsen, Professor, Aalborg University (Chairmann)
Mette Nørgaard, Professor, Aarhus University
Paul C. Lambert, Professor, University of Leicester
About the PhD thesis
Lymphomas are cancers of the blood system. The outlook of newly diagnosed patients ranges from excellent to critical and is highly depended on the subtype and disease stage. Currently, the staging of these patients is often based on simple scores that were constructed using sub-optimal statistical techniques. A substantial part of this dissertation describes the loss of information that occurs due to the use of these simple scores and proposes new alternatives.
In addition to being depended on the stage and disease subtype, the outlook of these patients also evolves given that patients survive the critical period after their initial diagnosis. For example, in this dissertation we show that for young Hodgkin lymphoma patients the relapse risk becomes minimal given that they survive two years without dying or experiencing a relapse.
This PhD dissertation consists of 6 papers:
Three discuss the use of prognostic models for the survival of lymphoma patients and investigates the loss of information due to
Two describe how the prognosis evolves given that patients reach milestones such as remaining relapse free for a number of years after their initial diagnosis.
The final paper proposes a new estimation technique for survival models which is not as strongly influenced by outliers as the standard estimation method.
The department of Clinical Medicine, Aalborg University and Department of Haematology, Aalborg University Hospital
Auditoriet, Forskningens Hus, Sdr. Skovvej 15, 9000 Aalborg