-How do we leverage modern data science, radiomics, and mathematical methods to improve the radiological interpretation, prognostication, and image quality in cardiothoracic imaging?
-How do we discover novel imaging biomarkers in cardiothoracic imaging?
-How do we improve the practice of radiologist-clinician and radiologist-patient communication, with the assistance of modern natural language processing techniques?
These are some of the key questions that our laboratory has been exploring.
Big data-driven model development for lung cancer and interstitial lung disease imaging
Radiological natural language processing including large language models to analyze and generate optimal radiology reports.
Clinical translation, validation, and optimization of 0.55T Lung MRI (Siemens Freemax).
