This Database contains 3397 patients for the consideration of Testosterone Deficiency. The Research Ethics Committee of the State University of Feira de Santana, Bahia, Brazil, approved the study with the ethical approval code: 3,057,301.
All six features were obtained by low-cost exams/informations, which enable other researchers to add new data and try different algorithms.
Features:
- Age: in years (we only considered patients aged from 45-85 years);
- Triglycerides (TG): Integer Numbers (mg/dl);
- Waist Circumference (WC): in centimeters (medical literature usually thresholds at 102cm);
- HDL: Integer numbers (mg/dl);
- Hypertension: Binary, in which 1 is presence, and 0 is absence;
- Diabetes: Binary, in which 1 is presence, and 0 is absence.
Target:
- T: Testosterone (Medical Literature suggests a deficiency when T<300 ng/dl).
Our thanks to UroS (Research group in Urology and Populational Subgroups), from Medical School, Universidade Estadual de Feira de Santana.
@article{novaes2021prediction, title={Prediction Of Secondary Testosterone Deficiency Using Machine Learning: Comparative Analysis Of Ensemble And Base Classifiers, Probability Calibration, And Sampling Strategies In A Slightly Imbalanced Dataset}, author={Novaes, Monique Tonani and de Carvalho, Osmar Luiz Ferreira and Ferreira, Pedro Henrique Guimar{~a}es and Tiraboschi, Taciana Leonel Nunes and Silva, Carolina Santos and Zambrano, Jean Carlos and Gomes, Cristiano Mendes and de Paula Miranda, Eduardo and de Carvalho J{'u}nior, Osmar Ab{'\i}lio and de Bessa J{'u}nior, Jos{'e}}, journal={Informatics in Medicine Unlocked}, pages={100538}, year={2021}, publisher={Elsevier} }