Natural language processing for precision clinical diagnostics and treatment

In this thesis, I focus upon application of natural language processing to clinical diagnostics and treatment within the palliative care and serious illness field. I explore a variety of natural language processing methods, including deep learning, rule-based, and classic machine learning, and applied to the identication of documentation reflecting advanced care planning measures, serious illnesses, and serious illness symptoms. I introduce two tools that can be used to analyze clinical notes from electronic health records: ClinicalRegex, a regular expression interface, and PyCCI, an a clinical text annotation tool. Additionally, I discuss a palliative care-focused research project in which I apply machine learning natural language processing methods to identifying clinical documentation in the palliative care and serious illness field. Advance care planning, which includes clarifying and documenting goals of care and preferences for future care, is essential for achieving end-of-life care that is consistent with the preferences of dying patients and their families. Physicians document their communication about these preferences as unstructured free text in clinical notes; as a result, routine assessment of this quality indicator is time consuming and costly. Integrating goals of care conversations and advance care planning into decision-making about palliative surgery have been shown to result in less invasive care near the time of death and improve clinical outcomes for both the patient and surviving family members. Natural language processing methods offer an efficient and scalable way to improve the visibility of documented serious illness conversations within electronic health record data, helping to better quality of care.

Department dc:contributor.department

Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science.

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2018

Author dc:creator

• Chien, Isabel

Advisor dc:contributor.advisor

• Manolis Kellis.

• Electrical Engineering and Computer Science.

Statement dc:rights

• MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission.

Licence dc:rights.uri

• http://dspace.mit.edu/handle/1721.1/7582

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/1721.1/119754

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/119754