PhD defence by Rikke Linnemann Nielsen

On Thursday 13 August, Rikke Linnemann Nielsen will defend her PhD thesis: “Big Data approaches for prediction of clinically relevant outcomes”.

Time:  Thursday 13 August, at 14:00

Place: Bldg 303A, Auditorium 41.

Important: Active registration:

- Building 303A, auditorium 41.                                                                                                                                       

Due to Covid-19 there is a restriction on the number of participants who are physically present in auditorium 41.

Therefore, if you wish to be present in the auditorium, please register by sending an e-mail to Ramneek Gupta (

All participants present in auditorium 41 are responsible for complying with the applicable guidelines for distance etc.

Zoom link: 

Principal supervisor: Associate Professor Ramneek Gupta
Co-supervisor: Professor Kjeld Schmiegelow
Co-supervisor: Professor Anders Gorm

Professor Morten Nielsen, DTU Health Tech
Professor Marylyn Ritchie, University of Pennsylvania
Professor Siqi Liu, BGI
Professor Quiyuan Li, Xiamen University
William Haynes, Novo Nordisk A/S

Chairperson at defence:
Associate Professor Lars Rønn Olsen

In the past decades, technological advances have provided the ability to collect, store and analyze Big data in disease biology. The analysis of Big data requires us address challenges posed by the large volume, veracity and heterogeneity. This includes, for instance, data integration across high-throughput omics data, electronic medical records and environmental characteristics. Machine learning approaches have shown prospects for Big data analyses and provide individual-level predictions. Interpretation of this data promises to derive value towards understanding clinically relevant outcomes such as disease onset, disease progression, treatment efficacy or treatment toxicity at an individual patient level. The promise of precision medicine is to use the individual variability in genomics, environment and lifestyle to guide personalized prevention and treatment strategies.

The PhD thesis uses machine learning methodologies to predict health, disease and treatment outcomes. The prediction models include data integration of genetics and clinical characteristics including both single time point and longitudinal data. The presented models have potential applications in precision medicine. Three projects are presented: i) Weight loss predictions based on physiology (incl. postprandial responses), urine metabolomics, gut microbiome and genetics from two dietary cross-over trials (N=102 study participants). The developed model may function as an early screening tool when determining individual weight loss strategies. II) Prediction of type 2 diabetes progression by the time to insulin in N~6000 patients over a 10 year follow-up from electronic medical records and genetics. The model may eventually assist in reduction of clinical inertia in very high-risk patients or reduce health care interventions and costs by identifying very low-risk patients. III) Prediction of asparaginase-associated pancreatitis (AAP, N=1390 patients), a serious childhood acute lymphoblastic leukemia (ALL) treatment-associated toxicity. It is currently difficult to predict which patients are at risk of AAP from clinical and genetic features. Even more difficult is the prediction of patients that will develop a second AAP following re-exposure to asparaginase after the first AAP event. The models may eventually assist in further stratification of patients and adjustment of treatment protocols of childhood ALL.



Thu 13 Aug 20
14:00 - 17:00


DTU Sundhedsteknologi


Bldg. 303A, aud. 41 and zoom