I believe that at least part of the answers to the great unknowns about diseases (how they originate or spread, how they evolve and why treatments work for some people and not others, for example) can be found in text contained in the myriad of publications, records and statements by researchers, clinicians and the patients themselves. If only we enable the systematic analysis of such data, taking care of accounting for biases and other barriers, significant progress can be made towards actionable health solutions. That is my goal and that of my team, to work with domain experts to develop the most advanced natural language processing, machine learning and artificial intelligence methods to help answer these questions."
The Gonzalez-Hernandez Laboratory has been a pioneer in integrating natural language processing and artificial intelligence for digital epidemiology. Our language processing methods are open to the research community and are designed for portability, with applications to mining information from real world data such as electronic health records, published literature and social media. In the last 5 years, we have made available advanced tools like DeepADEMiner, Kusuri, and SEED for extracting adverse events, medication names and symptoms mentioned in Twitter and electronic health records.
Meet Our Team
Learn more about our team of scientists with distinct expertise in medicine, public health, pharmacology, computational linguistics, machine learning, statistics or data science and who are dedicated to lead groundbreaking research.
Weissenbacher D, Ge S, Klein AZ, O’Connor K, Gross R, Hennessy S, Gonzalez-Hernandez G
JAMIA. 2021 Dec. 28(12): 2551-2561
Magge A, Tutubalina E, Miftahutdinov Z, Alimova I, Dirkson A, Verberne S, Weissenbacher D, Gonzalez-Hernandez G
JAMIA. 2021 Oct; 28(10): 2184-2192
Golder S, Smith K, O’Connor K, Gross R, Hennessy R, Hennessy S, Gonzalez-Hernandez G
Drug safety. 2021 Feb; 44(2): 167-179
Klein AZ, Sarker A, Weissenbacher D, Gonzalez-Hernandez G
NPJ digital medicine. 2019 Oct; 2(1): 1-9