Machine Learning Models and Big Data Tools for Evaluating Kidney Acceptance
Machine Learning Models and Big Data Tools for Evaluating Kidney Acceptance
复制标题
DOI：
10.1016/j.procs.2021.05.019
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Lirim Ashiku;Md Al-Amin;S. Madria;C. Dagli
中科院分区：
文献类型：
--
作者：
Lirim Ashiku;Md Al-Amin;S. Madria;C. Dagli
按需医疗保健的兴起和电子健康记录的空前增长带来了大数据机会和使用机器学习的数据分析。使用传统数据库进行海量且分散的数据管理极具挑战性且管理成本高昂。它通常需要专门的分析工具来开发高级数据驱动功能和执行数据分析。本文探讨了开源框架“Apache Spark”的能力，该框架能够在节点集群上处理大量数据，以分析大数据并集成技术，从而在医疗保健环境中提供决策支持系统。接下来，我们提出基于 Apache Spark 的机器学习模型，以加快分配器官的决策，例如为合适的候选人选择肾脏，从而通过在分配的时间内找到受者来提高捐赠者的利用率。所提出的模型有助于识别愿意接受可能被丢弃的肾脏的候补候选人。
The rise of on-demand healthcare and the unprecedented growth of electronic health records has given rise to big data opportunities and data analysis using machine learning. The massive and disparate data management using conventional databases is incredibly challenging and expensive to manage. It often requires specialized analytical tools for developing advanced data-driven capabilities and performing data analytics. This paper explores the capability of an open-source framework ‘Apache Spark’ capable of processing large amounts of data on clusters of nodes to analyze Big data and integrate technologies to provide decision support systems in healthcare settings. Next, we propose machine learning models on top of Apache Spark to expedite the decision-making in allocating organs such as kidney selection for the right candidate, thus increasing donor utilization by locating a recipient within the allotted time. The proposed models help in identifying waitlisted candidates willing to accept kidneys that may otherwise be discarded.