We investigated whether proxies of athlete social media activity are useful features for a machine learning model to predict athletes’ performance in subsequent competitions. We extracted millions of tweets that NBA basketball players posted themselves or were tagged in and derived features reflecting players’ mood, social media behaviour, and sleep quality before games. Using these and other non-social media-related features, we performed statistical tests to examine whether the features significantly improve the accuracy of a random forest model for predicting players’ BPM scores in upcoming games.
We investigate the potential of morphological and hemodynamic features extracted from measured blood flow data in the aorta to classify hear-thealthy volunteers (HHV) and patients with bicuspid aortic valve (BAV). Furthermore, we determine features that distinguish male vs. female patients and elderly HHV vs. BAV patients. We propose a data analysis pipeline for cardiac status classification, encompassing feature selection, model training, and hyperparameter tuning.
My PhD thesis.
Knowledge of different disease phenotypes can help understand (a) which patient subpopulations seek treatment and (b) the response to treatment within each subgroup. In this paper, we presented a workflow to (i) determine distinct phenotypes of medical conditions in high-dimensional data, (ii) visualize these phenotypes to explore and compare essential subpopulation characteristics, and (iii) interactively inspect them and their change over time with an interactive web application. We evaluated our workflow by identifying four distinct phenotypes of tinnitus patients.
Study examining (1) how male and female tinnitus patients differ, (2) the extent to which gender is related to (tinnitus) questionnaire responses, and (3) which baseline questionnaire items are related to tinnitus distress, depression, and treatment response, only one gender.
- ← Newer
- 1 of 2
- Older →