English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Background: The increasing adoption of electronic health records (EHR) acrossthe US has created troves of computable data, to which machine learning methodshave been applied to extract useful insights. EHR data, represented as athree-dimensional analogue of a matrix (tensor), is decomposed intotwo-dimensional factors that can be interpreted as computational phenotypes.Methods: We apply constrained tensor factorization to derive computationalphenotypes and predict mortality in cohorts of patients with breast, prostate,colorectal, or lung cancer in the Northwestern Medicine Enterprise DataWarehouse from 2000 to 2015. In our experiments, we examined using a supervisedterm in the factorization algorithm, filtering tensor co-occurrences by medicalindication, and incorporating additional social determinants of health (SDOH)covariates in the factorization process. We evaluated the resultingcomputational phenotypes qualitatively and by assessing their ability topredict five-year mortality using the area under the curve (AUC) statistic.Results: Filtering by medical indication led to more concise and interpretablephenotypes. Mortality prediction performance (AUC) varied under the differentexperimental conditions and by cancer type (breast: 0.623 - 0.694, prostate:0.603 - 0.750, colorectal: 0.523 - 0.641, and lung: 0.517 - 0.623). Generally,prediction performance improved with the use of a supervised term and theincorporation of SDOH covariates. Conclusion: Constrained tensor factorization,applied to sparse EHR data of patients with cancer, can discover computationalphenotypes predictive of five-year mortality. The incorporation of SDOHvariables into the factorization algorithm is an easy-to-implement andeffective way to improve prediction performance.

Constrained tensor factorization for computational phenotyping and mortality prediction in patients with cancer