Machine Learning for Lung Disease
Lung diseases, are among the most common medical conditions in the world. According to the American Journal of Managed Care (AJMC), the mortality rate from chronic respiratory diseases has risen by almost 30% between 1980 and 2014 in the United States, and that data reflect significant differences in death rates across counties and regions. Diagnosis and prognosis of lung diseases early-on plays a key role in preventing long-term damages to the lung and health of patients. Thanks to the advances in machine learning and availability of X-rays and computerized tomography (CT) scans, we can use these methods for diagnosis and prognosis of lung disease. It is also of great importance to use machine learning for extracting the underlying patterns in the data, instead of just performing a classification task. This can help radiologists to understand the dynamics of the disease better and possibly study the effect of different treatments. In this project, I am working on interpretable machine learning models for diagnosis, prognosis, and understanding of lung diseases.