Northwestern University Feinberg School of Medicine

Biomedical Natural Language Processing Lab


Biomedical Language Processing Laboratory seeks to improve healthcare outcomes by enabling secondary uses of data and create advanced algorithms to enhance clinical knowledge systems. Following is a description of our current work. Our lab welcomes opportunities for collaboration.

Improving the Efficiency and Efficacy in Authoring Essential Clinical FAQs

Medical errors are one of the leading causes of death in the United States. It has been observed that point-of-care access to relevant clinical knowledge supports decision-making and decreases medical errors, thereby improving patient safety and healthcare costs. Our proposed research aims to empower physicians specialized in the area (specialists) in quickly gathering evidence from literature or finding citations supporting or qualifying their expert opinion. It will also generate the answers and suggest updates to the existing answers for their perusal.

Evidence-based Clinical Knowledge Summarization

Online clinical resources such as UpToDate provide recommendations to support clinicians at the point of care. A recommendation summarizes the knowledge from high-quality publications with a good study design. Currently, clinicians and physicians have to manually generate these recommendations. So to mitigate the time taken, manual labor as well as bias errors, our lab is exploring Natural Language Processing techniques to automatically generate the knowledge summaries or recommendations. Our research aims to utilize existing online resources, such as UpToDate, to understand the knowledge gap between online resources and MEDLINE. This helps us to develop algorithms to provide new knowledge that is not yet available in online resources.

Clinical Natural Language Processing

Electronic Health Records (EHR) are the most important source of a patient’s medical knowledge. However, information processing from EHR is very challenging due to presence of large complex unstructured data in form of clinical notes. We apply natural language processing and information extraction techniques to extract patient information such as medications prescribed and dosagefrom EHR. This information is useful to clinicians at the point-of-care to decide the best suitable treatment for their patients.

Information Extraction-based Approach to Automate Clinical Trial Prescreening

To reduce the large amount of time spent screening, identifying, and recruiting patients into clinical trials, researchers need prescreening systems that are able to automate the data extraction and decision-making tasks that are typically relegated to clinical research study coordinators. Our lab is working on an information extraction-based approach to obtain important data elements and values from eligibility criteria and clinical notes for easier matching.

Recent Publications

For a list of publication from this lab, see the faculty profile of our lab’s principal investigator, Siddhartha Jonnalagadda, PhD.