1. How to submit my research paper? What’s the process of publication of my paper?
The journal receives submitted manuscripts via email only. Please submit your research paper in .doc or.pdf format to the submission email: ijpmbs@ejournal.net.
You’ll be given a paper number if your submission is successful. Your paper then will undergo peer review process, which may take approximately one and a half months under normal circumstances, three tops.
After blind peer review, you will receive the notification letter with the review result of your paper...
2. Can I submit an abstract?
The journal publishes full research papers.[Read More]
 

Predicting the Effect of Parathyroidectomy on Patient Survival in Secondary Hyperparathyroidism with Machine Learning

Oktoria 1, Cheng-Hong Yang 1, and Jin-Bor Chen 2
1. Department of Electronic Engineering, National Kaohsiung University of Applied Sciences, Kaohsiung, Taiwan
2. Division of Nephrology, Department of Internal Medicine, Kaohsiung Chang Gung Memorial Hospital and Chang

Abstract—The main goal of parathyroidectomy (PTX) is to remove the offending gland(s) while protecting the remaining normal parathyroid glands as well as the recurrent laryngeal nerves and the thyroid gland. In this study, the writer hypothesized that Machine Learning (ML) could predict the effect of PTX based on readily available clinical and laboratory indicators. There were 158 consecutive HD patients who underwent PTX before 2009 and 275 consecutive hemodialysis (HD) patients without PTX as controls from those visiting the Kaohsiung Chang Gung Memorial Hospital, Taiwan between 2009 and 2013. The study first held by testing several categories of supervised ML classifiers: 1) Bayesian network classifier, 2) k-Nearest Neighbors, 3) rule-based classifiers, and 4) tree-based classifiers. All ML classifiers were tested using 10-fold cross validation. The performance of each classifier was evaluated based on sensitivity (recall), specificity, positive predictive value (precision), area under the receiver operating characteristic (ROC) curve, and overall accuracy. After testing >20 different algorithms, we selected tree-based classifier (Random Forest) that has the highest value of correctly classified instances, namely 76.91% (area under receiver operating characteristic = 0.78).

Index Terms—parathyroidectomy, machine learning, classifier, random forest

Cite:Oktoria, Cheng-Hong Yang, and Jin-Bor Chen, "Predicting the Effect of Parathyroidectomy on Patient Survival in Secondary Hyperparathyroidism with Machine Learning," International Journal of Pharma Medicine and Biological Sciences, Vol. 6, No. 2, pp. 58-62, April 2017. doi: 10.18178/ijpmbs.6.2.58-62
Copyright © 2012-2024 International Journal of Pharma Medicine and Biological Sciences, All Rights Reserved
E-mail: ijpmbs@ejournal.net