You are here


2019/20 I. félév
Árpád tér 2. II. em. 220. sz.
15:15 16:00
Kicsi András
Machine understanding of textual radiological reports

My talk will present an artificial intelligence based method for quick comprehension of textual radiological reports. Thousands of reports are created annually which represent textual information that is often expressed in the native language of the radiologist. Comparison and later examination of their content are cumbersome. To the best of our knowledge, we are the first to introduce an intelligent comprehension and detailed visualization technique that supports Hungarian language reports.  Our method relies on deep learning and various additional artificial intelligence procedures. The machine learning classification builds upon data from almost 500 anonymized reports manually annotated by radiologists according to specific annotation guidelines. The classification process detects various anatomic locations and pathologies with quantification, which are further specified with natural language processing techniques. Several linguistic characteristics are also addressed, negative statements and non-pathological disorders are separated from the general text. Currently, we limited our focus on lumbar spine MRI reports.