2023 World Conference on Lung Cancer (Posters)-EP17.01. Proposing Relevant Diagnosis Associated with Lung Cancer Based on Prior Diagnosis: Implementing Apriori Algorithm

EP17.01. Proposing Relevant Diagnosis Associated with Lung Cancer Based on Prior Diagnosis: Implementing Apriori Algorithm - PDF(Slides)

Back to course

Pdf Summary

A study conducted in Seoul, South Korea, aimed to use the Apriori algorithm to identify relevant diagnoses associated with lung cancer. The algorithm was applied to a dataset of 205,762 patients who visited a tertiary academic hospital and had 18,111,638 diagnoses recorded. The analysis generated a total of 82,873 association rules.<br /><br />The study found that lung cancer (C34) was associated with various diagnoses, including abnormal findings on lung imaging, chronic obstructive pulmonary disease (COPD), sleep disorders, other respiratory disorders, secondary malignant neoplasms of other or unspecified sites, pneumonia, respiratory conditions due to other external agents, respiratory failure, and emphysema.<br /><br />By using prior diagnoses, it is possible to predict additional future diagnoses and potentially detect pre-symptomatic diseases earlier. However, further research is needed to establish the causal and temporal relationship between these diagnoses.<br /><br />Through the visualization of association rules using arulesViz, the study aimed to provide a clear understanding of the associations between lung cancer and other diagnoses.<br /><br />The findings of this study have implications for the diagnosis and management of lung cancer patients. Identifying associated diagnoses can assist in early detection and treatment planning. This research highlights the potential of data mining algorithms, specifically the Apriori algorithm, in uncovering relevant associations between different medical conditions.<br /><br />In conclusion, this study employed the Apriori algorithm to analyze a large dataset of prior diagnoses and identified several relevant diagnoses associated with lung cancer. These findings have the potential to inform healthcare professionals in the diagnosis and management of lung cancer patients, ultimately contributing to improved patient outcomes.

Asset Subtitle

So Yeon Kim

Meta Tag

Speaker So Yeon Kim

Topic Global Health, Health Services & Health Economics: Digital Solutions

Keywords

Apriori algorithm

lung cancer

diagnoses

dataset

association rules

abnormal findings

chronic obstructive pulmonary disease

respiratory disorders

prior diagnoses

data mining algorithms