Dataset Information

A method for inferring medical diagnoses from patient similarities.

ABSTRACT:

Background

Clinical decision support systems assist physicians in interpreting complex patient data. However, they typically operate on a per-patient basis and do not exploit the extensive latent medical knowledge in electronic health records (EHRs). The emergence of large EHR systems offers the opportunity to integrate population information actively into these tools.

Methods

Here, we assess the ability of a large corpus of electronic records to predict individual discharge diagnoses. We present a method that exploits similarities between patients along multiple dimensions to predict the eventual discharge diagnoses.

Results

Using demographic, initial blood and electrocardiography measurements, as well as medical history of hospitalized patients from two independent hospitals, we obtained high performance in cross-validation (area under the curve >0.88) and correctly predicted at least one diagnosis among the top ten predictions for more than 84% of the patients tested. Importantly, our method provides accurate predictions (>0.86 precision in cross validation) for major disease categories, including infectious and parasitic diseases, endocrine and metabolic diseases and diseases of the circulatory systems. Our performance applies to both chronic and acute diagnoses.

Conclusions

Our results suggest that one can harness the wealth of population-based information embedded in electronic health records for patient-specific predictive tasks.

SUBMITTER: Gottlieb A

PROVIDER: S-EPMC3844462 | biostudies-literature | 2013 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A method for inferring medical diagnoses from patient similarities.

Gottlieb Assaf A Stein Gideon Y GY Ruppin Eytan E Altman Russ B RB Sharan Roded R

BMC medicine 20130902

<h4>Background</h4>Clinical decision support systems assist physicians in interpreting complex patient data. However, they typically operate on a per-patient basis and do not exploit the extensive latent medical knowledge in electronic health records (EHRs). The emergence of large EHR systems offers the opportunity to integrate population information actively into these tools.<h4>Methods</h4>Here, we assess the ability of a large corpus of electronic records to predict individual discharge diagn ...[more]

PMID: 24004670

Similar Datasets

Project description:BackgroundRacial/ethnic disparities in access to diagnostic services are pervasive for autistic children. However, a few studies have examined racial/ethnic health disparities among autistic adults, who commonly experience higher rates of health conditions than non-autistic adults. We aimed at examining the intersection of autism and race/ethnicity in association with psychiatric and medical diagnoses.MethodsThe study population included adult members of Kaiser Permanente Northern California enrolled from 2008 to 2012. We ascertained 1507 adults who had an autism diagnosis documented in their electronic medical records. We sampled a matched control group of adults without an autism diagnosis (N = 15,070) at a 10:1 ratio. Our sample was 46% White, 17% Hispanic, 16% Asian, 7% Black, and 14% other race/ethnicity. We compared health diagnoses (a) between autistic and non-autistic adults within strata of race/ethnicity and (b) across race/ethnicity within strata of autistic and non-autistic adults. Lastly, we examined the interaction between autism and race/ethnicity on both multiplicative and additive scales.ResultsAutistic adults were more likely to be diagnosed with most medical and psychiatric conditions compared with their non-autistic counterparts of the same race/ethnicity. Among autistic adults, Black, Hispanic, and Asian adults were less likely to be diagnosed with psychiatric conditions and Black and Hispanic autistic adults were more likely to be diagnosed with obesity than their White counterparts. In interaction models, we found that adults who were Black and autistic were disproportionately less likely to be diagnosed with psychiatric conditions and autoimmune disease and more likely to be diagnosed with hypertension than expected.ConclusionHealth vulnerabilities may be compounded at the intersection of autism and race/ethnicity. Future research should continue to apply an intersectional lens toward understanding and addressing these disparities. Our findings likely underestimate the health disparities that exist in uninsured autistic adults and those living in other parts of the United States.

Dataset Information

A method for inferring medical diagnoses from patient similarities.

Background

Methods

Results

Conclusions

Publications

A method for inferring medical diagnoses from patient similarities.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets