Medical application for clinical prediction of mortality rate in patients with traumatic brain injury
DOI:
https://doi.org/10.5377/wani.v38i77.14659Keywords:
Neural Networks, Machine Learning, Predictive Model, Artificial Intelligence, MedicineAbstract
One of the fastest growing fields in recent years has been artificial intelligence, which is divided into subfields such as machine learning, which provides techniques and algorithms so that systems can learn and improve automatically. The objective of the project was to develop a medical application whose function is to optimize and facilitate decision-making regarding the care provided to patients with head trauma, through a predictive model that indicates the probability of death of these patients. The extreme programming (XP) methodology was applied, using machine learning techniques based on the CRASH-2 data set, which has 20,207 records of randomized traumatic brain injury patients. Artificial neural networks were used to build the predictive survival model. The neural network with the 6-(8-14-2)-1 architecture achieved an accuracy of 76%, sensitivity of 72.9%, and specificity of 94.1% on the test data set; demonstrating a promising discrimination capacity with good adaptation to internal validation.
Downloads
616
MP3 (Español (España)) 56
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2022 Bluefields Indian and Caribbean University
![Creative Commons License](http://i.creativecommons.org/l/by-nc-sa/4.0/88x31.png)
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.