Olga Demler, Ph.D.
Senior Scientist
- olga.demler@inf.ethz.ch
- Address
-
Biomedical Informatics Group
Universitätstrasse 6
8092 Zürich - Room
- CAB F52.2
I hold a joint appointment as a Senior Scientist in the Biomedical Informatics Lab at ETH Zürich and as a Biostatistician and Assistant Professor in the Division of Preventive Medicine at Mass General Brigham / Harvard Medical School in Boston. This dual affiliation enables cross-disciplinary collaboration between clinical cardiovascular researchers in Boston and computer scientists at ETH Zürich.
My research focuses on developing AI- and machine learning–based methodologies for risk prediction, biomarker discovery, and diagnostic modeling, with applications in medical imaging, electronic health records, and large-scale omics data. I am particularly interested in:
- Advancing statistical methods for individualized risk prediction
- Addressing analytic challenges in high-throughput biomarker discovery, especially metabolomics
- Building diagnostic and prognostic models using multimodal clinical data
A key clinical theme of my work is understanding the mechanisms that drive the onset and progression of cardiovascular disease.
I earned my Ph.D. in Biostatistics from Boston University under the supervision of Profs. Ralph D’Agostino Sr. and Michael Pencina, focusing on the evaluation of risk prediction models. I joined the Division of Preventive Medicine at Mass General Brigham / Harvard Medical School in 2012, first as a postdoctoral fellow and now as an Assistant Professor. In 2022, I joined Prof. Gunnar Rätsch’s group at ETH Zürich, where I expanded my research into medical imaging analytics. I currently lead the design and analysis of high-throughput metabolomics studies, primarily in collaboration with Prof. Samia Mora’s group and colleagues at ETH Zürich. I also serve as Principal Investigator on several ongoing projects that translate innovations in analytics and AI into meaningful improvements in clinical research and patient care.
Latest Publications
Abstract Question Can established cardiovascular risk tools be adapted for local populations without sacrificing interpretability? Findings This cohort study including 95 326 individuals applied a machine learning recalibration method that uses minimal variables to the American Heart Association’s Predicting Risk of Cardiovascular Disease Events (AHA-PREVENT) equations for a New England population. This approach strengthened the AHA-PREVENT risk equations, improving calibration while maintaining similar risk discrimination. Meaning The results indicate that the interpretable machine learning-based recalibration method used in this study can be implemented to tailor risk stratification in local health systems.
Authors Aniket N Zinzuwadia, Olga Mineeva, Chunying Li, Zareen Farukhi, Franco Giulianini, Brian Cade, Lin Chen, Elizabeth Karlson, Nina Paynter, Samia Mora, Olga Demler
Submitted JAMA cardiology
Abstract Fracture prediction is essential in managing patients with osteoporosis and is an integral component of many fracture prevention guidelines. We aimed to identify the most relevant clinical fracture risk factors in contemporary populations by training and validating short- and long-term fracture risk prediction models in 2 cohorts. We used traditional and machine learning survival models to predict risks of vertebral, hip, and any fractures on the basis of clinical risk factors, T-scores, and treatment history among participants in a nationwide Swiss Osteoporosis Registry (N = 5944 postmenopausal women, median follow-up of 4.1 yr between January 2015 and October 2022; a total of 1190 fractures during follow-up). The independent validation cohort comprised 5474 postmenopausal women from the UK Biobank with 290 incident fractures during follow-up. Uno’s C-index and the time-dependent area under the receiver operating characteristics curve were calculated to evaluate the performance of different machine learning models (Random survival forest and eXtreme Gradient Boosting). In the independent validation set, the C-index was 0.74 [0.58, 0.86] for vertebral fractures, 0.83 [0.7, 0.94] for hip fractures, and 0.63 [0.58, 0.69] for any fractures at year 2, and these values further increased for longer estimations of up to 7 yr. In comparison, the 10-yr fracture probability calculated with FRAX Switzerland was 0.60 [0.55, 0.64] for major osteoporotic fractures and 0.62 [0.49, 0.74] for hip fractures. The most important variables identified with Shapley additive explanations values were age, T-scores, and prior fractures, while number of falls was an important predictor of hip fractures. Performances of both traditional and machine learning models showed similar C-indices. We conclude that fracture risk can be improved by including the lumbar spine T-score, trabecular bone score, numbers of falls and recent fractures, and treatment information has a significant impact on fracture prediction.
Authors Oliver Lehmann, Olga Mineeva, Dinara Veshchezerova, HansJörg Häuselmann, Laura Guyer, Stephan Reichenbach, Thomas Lehmann, Olga Demler, Judith Everts-Graber, The Swiss Osteoporosis Registry Study Group
Submitted Journal of Bone and Mineral Research