Current Projects

Validating a natural language processing method for identifying complete pathologic response of breast cancer patients undergoing neoadjuvant chemotherapy from electronic pathology reports.   

Identifying response to cancer treatment, such as chemotherapy and surgery, usually requires manual review of the medical charts (i.e. pathology reports), given the reports are written in free-text. This poses significant difficulties for researchers to obtain this important information for a large number of patients due to its high cost and labor-intensity. This study aims to develop an automated Natural Language Processing method to effectively and efficiently extract treatment responses from the narrative data.