Format

Send to

Choose Destination
AMIA Annu Symp Proc. 2018 Apr 16;2017:1554-1563. eCollection 2017.

Leveraging Collaborative Filtering to Accelerate Rare Disease Diagnosis.

Author information

1
Department of Health Sciences Research, Mayo Clinic, Rochester, MN, USA.

Abstract

In the USA, rare diseases are defined as those affecting fewer than 200,000 patients at any given time. Patients with rare diseases are frequently misdiagnosed or undiagnosed which may due to the lack of knowledge and experience of care providers. We hypothesize that patients' phenotypic information available in electronic medical records (EMR) can be leveraged to accelerate disease diagnosis based on the intuition that providers need to document associated phenotypic information to support the diagnosis decision, especially for rare diseases. In this study, we proposed a collaborative filtering system enriched with natural language processing and semantic techniques to assist rare disease diagnosis based on phenotypic characterization. Specifically, we leveraged four similarity measurements with two neighborhood algorithms on 2010-2015 Mayo Clinic unstructured large patient cohort and evaluated different approaches. Preliminary results demonstrated that the use of collaborative filtering with phenotypic information is able to stratify patients with relatively similar rare diseases.

PMID:
29854225
PMCID:
PMC5977716
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for PubMed Central
Loading ...
Support Center