Chaitanya Shivade
Applied Scientist at Amazon
At Ohio State
I was a PhD student at the Speech and Language Processing (SLaTe) lab. I was advised by Eric Fosler-Lussier and Albert Lai. My research was interdisciplanry involving collaboration with colleagues from deparments at Computer Science, Lingusitics, and Biomedical Informatics. I was a member of the Computational Lingusitics and Language Technology (CLLT) group.
Research
- Textual Inference for expediting clinical trial screening
- We developed techniques using natural language processing and machine learning to introduce automation into the process of clinical trial screening
- Shivade, Chaitanya, et al. Textual Inference for Eligibility Criteria Resolution in Clinical Trials. Journal of Biomedical Informatics (2015).
- Shivade, Chaitanya, Preethi Raghavan, and Siddharth Patwardhan. Addressing Limited Data for Textual Entailment Across Domains. Proceedings of ACL (2016).
- Shivade, Chaitanya, et al. Comparison of UMLS Terminologies to Identify Risk of Heart Disease in Clinical Notes. Journal of Biomedical Informatics (2015).
- Predicting 30-day readmissions in hospitals
- We developed a statistical model to identify patients with the risk of readmission across multiple diagnosis
- Hebert, Courtney, Chaitanya Shivade, Randi Foraker, Jared Wasserman, Caryn Roth, Hagop Mekhjian, Stanley Lemeshow, and Peter Embi. Diagnosis-specific readmission risk prediction using electronic health data: a retrospective cohort study. BMC Medical Informatics and Decision Making (2014)
- Understanding adjective and adverb scales
- We developed techniques to identify, understand and order the meaning gradable terms such as adjectives and adverbs in biomedical texts. (e.g. serious < life-threatening < fatal)
- Shivade, Chaitanya, Marie-Catherine de Marneffe, Eric Fosler-Lussier, and Albert M. Lai. Corpus-based discovery of semantic intensity scales. In Proceedings of NAACL 2015.
- Shivade, Chaitanya, Marie-Catherine de Marneffe, Eric Fosler-Lussier, and Albert M. Lai. Identification, characterization, and grounding of gradable terms in clinical text. In Proceedings of the Biomedical Natural Language Processing Workshop at ACL 2016.
Talks
- At University of San Diego, Division of Biomedical Informatics A review of approaches to identifying patient phenotype cohorts using electronic health records
Awards
- Best Dissertation Idea At the American Medical Informatics Association [AMIA] NLP Doctoral Symposium
Patents
- US20160203287A1 Chen, James L., Chaitanya Shivade, and David Liebner. Methods for predicting prognosis. U.S. Patent Application 14/912,961, filed July 14, 2016.