Predicting Age of Onset in TTR-FAP Patients with Genealogical Features
| Autor: | João Mendes-Moreira, Alípio Mário Jorge, Teresa Coelho, Maria Pedroto |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
0301 basic medicine
Elastic net regularization business.industry Computer science Decision tree 030105 genetics & heredity Machine learning computer.software_genre Regression Random forest Support vector machine 03 medical and health sciences 0302 clinical medicine Lasso (statistics) Linear regression Artificial intelligence Age of onset business computer 030217 neurology & neurosurgery |
| Zdroj: | CBMS |
| Popis: | This work describes a problem oriented approach to analyze and predict the Age of Onset of Patients diagnosed with Transthyretin Familial Amyloid Polyneuropathy (TTR-FAP). We constructed, from a set of clinical and familial records, three sets of features which represent different characteristics of a patient, before becoming symptomatic. Using those features, we tested a set of machine learning regression methods, namely Decision Tree (Regression Tree), Elastic Net, Lasso, Linear Regression, Random Forest Regressor, Ridge Regression and Support Vector Machine Regressor (SVM). Later, we defined a baseline model that represents the current medical practice to serve as a guideline for us to measure the accuracy of our approach. Our results show a significant improvement of machine learning methods when compared with the current baseline. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|