Search Articles

View query in Help articles search

Search Results (1 to 3 of 3 Results)

Download search results: CSV END BibTex RIS


Unsupervised Feature Selection to Identify Important ICD-10 and ATC Codes for Machine Learning on a Cohort of Patients With Coronary Heart Disease: Retrospective Study

Unsupervised Feature Selection to Identify Important ICD-10 and ATC Codes for Machine Learning on a Cohort of Patients With Coronary Heart Disease: Retrospective Study

We did a similar procedure for ATC codes in the PIN data set and aggregated the codes every 6 months, since most medications prescription refills did not extend beyond 6 months. We one-hot encoded the ICD-10-CA and ATC codes and their parent nodes for each record. For example, if the ICD-10-CA code “I251” was present, “I25,” “I20-I25,” and “Chapter IX” were also encoded in the one-hot table. Similarly, if the ATC code “C07 AB02” was present, “C07 AB,” “C07 A,” “C07,” and “C” were also encoded.

Peyman Ghasemi, Joon Lee

JMIR Med Inform 2024;12:e52896

A Multimorbidity Analysis of Hospitalized Patients With COVID-19 in Northwest Italy: Longitudinal Study Using Evolutionary Machine Learning and Health Administrative Data

A Multimorbidity Analysis of Hospitalized Patients With COVID-19 in Northwest Italy: Longitudinal Study Using Evolutionary Machine Learning and Health Administrative Data

From the drug prescriptions data set, the Anatomical Therapeutic Chemical (ATC) classification system codes were used. All distinct ATC codes up to the 4th level (the first 5 digits of the ATC codes) were considered in this study. One-hot encoding was applied to convert categorical codes into separate feature columns with binary values (0 or 1) indicating the absence or presence of drugs in each patient’s prescription history.

Dayana Benny, Mario Giacobini, Alberto Catalano, Giuseppe Costa, Roberto Gnavi, Fulvio Ricceri

JMIR Public Health Surveill 2024;10:e52353

Assessment and Improvement of Drug Data Structuredness From Electronic Health Records: Algorithm Development and Validation

Assessment and Improvement of Drug Data Structuredness From Electronic Health Records: Algorithm Development and Validation

During the validation step, information was added to each algorithm to determine whether the correct ATC code, wrong ATC code, or no ATC code was identified. If no algorithm identified the correct ATC code, it was determined by manual validation when possible. If an entry was found to generally have no drug prescription, it was marked as an additional entry without drug prescription with the keyword “nomed.”

Ines Reinecke, Joscha Siebel, Saskia Fuhrmann, Andreas Fischer, Martin Sedlmayr, Jens Weidner, Franziska Bathelt

JMIR Med Inform 2023;11:e40312

Cookie Consent

We use our own cookies and third-party cookies so that we can show you this website and better understand how you use it, with a view to improving the services we offer. If you continue browsing, we consider that you have accepted the cookies.