# Predicting Development of Chronic Obstructive Pulmonary Disease and its Risk Factor Analysis

**Authors:** Soojin Lee, Ingu Sean Lee, Samuel Kim  
**arXiv:** [2302.03137](https://arxiv.org/abs/2302.03137)  
**Date:** 2023-02-06  
**Summary page:** https://www.summarizepaper.com/en/paper/2302.03137v1/

## Key points

- Chronic Obstructive Pulmonary Disease (COPD) is characterized by irreversible airway obstruction and a significant societal burden.
- Smoking is the primary risk factor for COPD, but other components must also be considered.
- Previous studies have often focused on sociodemographic or genetic factors individually, rather than examining them together to assess COPD risk comprehensively.
- Researchers aim to identify and analyze various risk factors associated with COPD development using machine learning models that integrate sociodemographic, clinical, and genetic data.
- The study will involve longitudinal observation of comprehensive medical data to determine the most strongly correlated factors with COPD development.
- The researchers will utilize their proprietary research analysis platform, COMPASS, to extract and analyze important features from diverse sources of medical data.
- Data from the UK Biobank will be used, which contains detailed genetic, clinical, and sociodemographic information from over 500,000 participants.
- Sociodemographic and clinical factors associated with COPD risk have been selected based on previous research findings, along with genomic features linked to lung function and COPD.
- By identifying modifiable and non-modifiable risk factors for COPD development, early detection of the disease may be enabled leading to prompt management strategies and potentially preventing onset or progression of COPD.

## Summary

Chronic Obstructive Pulmonary Disease (COPD) is a debilitating condition characterized by irreversible airway obstruction and a significant societal burden. While smoking is widely recognized as the primary risk factor for COPD, there are additional components that must be taken into consideration. Previous studies have often focused on either sociodemographic or genetic factors individually, rather than examining them in conjunction to assess COPD risk comprehensively. In this study, researchers aim to identify and analyze various risk factors associated with the development of COPD by utilizing machine learning models that integrate sociodemographic, clinical, and genetic data. By taking a holistic approach and considering all relevant factors simultaneously, the goal is to gain a deeper understanding of the complex nature of COPD and its progression. This research will involve longitudinal observation of comprehensive medical data to determine which factors are most strongly correlated with the development of COPD. The researchers will leverage their proprietary research analysis platform, COMPASS, to extract and analyze important features from diverse sources of medical data. The study will draw upon data from the UK Biobank, a large-scale biomedical database containing detailed genetic, clinical, and sociodemographic information from over 500,000 participants. Sociodemographic and clinical factors associated with COPD risk have been selected based on previous research findings, while genomic features known to be linked with lung function and COPD have also been incorporated into the analysis. By identifying both modifiable and non-modifiable risk factors for COPD development, the researchers hope to enable early detection of the disease. This proactive approach could lead to prompt management strategies and ultimately aid in preventing the onset or progression of COPD. Through their comprehensive analysis of diverse datasets using advanced machine learning techniques, this study aims to contribute valuable insights towards improving diagnosis and treatment outcomes for individuals affected by COPD.