Open-source Data Analysis and Machine Learning for Asthma Hospitalisation Rates

Rooney, Laura; Chute, Chaloner; Buchanan, William J; Smales, Adrian; Hepburn, Leigh-Anne

Abstract

Long-term conditions in Scotland account for 80% of all GP consultations; they also account for 60% of all deaths in Scotland. Asthma and Chronic Obstructive Pulmonary Disease (COPD) are common long-term respiratory diseases [1]. Asthma is a heterogeneous disease, usually characterized by chronic airway inflammation. It is defined by the history of respiratory symptoms such as wheeze, shortness of breath, chest tightness and cough that vary over time and in intensity, together with variable expiratory airflow limitation [2]. So far, we know that there are many different things – such as viruses, allergens, and pollution – that cause asthma or trigger attacks but not why or how they do it. This paper outlines how an open source dataset can be used to estimate asthma hospitalisation rates and uses machine learning to predict these rates, within ±7.5%, and for an 86.67% success rate.

Citation

Rooney, L., Chute, C., Buchanan, W. J., Smales, A., & Hepburn, L. (2018). Open-source Data Analysis and Machine Learning for Asthma Hospitalisation Rates. In Proceedings of ThinkMind - GLOBAL HEALTH 2018, The Seventh International Conference on Global Health Challenges

• Conference Name: Global Health 2018
• Conference Location: Athens, Greece
• Start Date: Nov 18, 2018
• End Date: Nov 22, 2018
• Acceptance Date: Nov 1, 2018
• Publication Date: Nov 18, 2018
• Deposit Date: Nov 20, 2018
• Publicly Available Date: Nov 21, 2018
• Publisher: International Academy, Research, and Industry Association
• Series ISSN: 2308-4553
• Book Title: Proceedings of ThinkMind - GLOBAL HEALTH 2018, The Seventh International Conference on Global Health Challenges
• ISBN: 9781612086828
• Keywords: asthma, copd, machine learning, open source
• Public URL: http://researchrepository.napier.ac.uk/Output/1370166
• Publisher URL: http://www.thinkmind.org/index.php?view=article&articleid=global_health_2018_1_10_70037