Prediction Models of Ischemic Stroke Using Deep Neural Networks
Name
Siim Kurvits
Abstract
The ischemic stroke is one of the leading causes of death worldwide. Although,there are many known risk factors for the disease the growing amount of electronicmedical data available gives opportunities for creating novel models for personal riskprediction. Usage of deep neural network (DNN) for developing such models can offersmany benefits such as potential to encode multiple types of data, less feature selection andengineering required, and sometimes even an increased prediction accuracy. This Thesisfocuses on developing a model for ischemic stroke prediction using electronic healthrecord (EHR) data. I show that TabNet, a state-of-the artDNNarchitecture for tabulardata analysis outperforms a simpler method, the FastAI tabular learner. Still, neitherof theDNNmethods achieved better results than the Random Forest. The ensemblemodels using Random Forest andDNNmodels were tested but only a small increase inthe performance was achieved compared to the singular model. These results indicatethat an ensemble-based methods such as Random Forest is sufficient for the data used.Nevertheless, with increased number of features and addition of more complex data typesmethods such as TabNet could still become valuable. All models developed resulted withhigh prediction power for ischemic stroke. This indicates that personal risk predictionsfor ischemic stroke can be given and the clinical utility of the models should be evaluatedfurther.
Graduation Thesis language
English
Graduation Thesis type
Master - Computer Science
Supervisor(s)
Toomas Haller, Ardi Tampuu
Defence year
2021