Lots of changes. Added clinspacy_init() and changed how the package l…

…oads. Uses miniconda by default but the user can configure.

R/cui2vec_data.R

@@ -1,41 +1,50 @@
 #' Cui2vec concept embeddings
 #'
-#' This dataset contains sample medical transcriptions for various medical specialties.
+#' This dataset contains Unified Medical Langauge System (UMLS) concept embeddings from
+#' Andrew Beam's \href{https://github.com/beamandrew/cui2vec}{cui2vec R package}. There are
+#' 500 embeddings included for each concept.
 #'
-#' Acknowledgements
+#' Citation
 #'
-#' This data was scraped from mtsamples.com by Tara Boyle and is made available
-#' under a CC0: Public Domain license.
+#' Beam, A.L., Kompa, B., Schmaltz, A., Fried, I., Griffin, W, Palmer, N.P., Shi, X.,
+#' Cai, T., and Kohane, I.S.,, 2019. Clinical Concept Embeddings Learned from Massive
+#' Sources of Multimodal Medical Data. arXiv preprint arXiv:1804.01486.
 #'
-#' @format A data frame with 4999 rows and 6 variables:
+#' License
+#'
+#' This data is made available under a
+#' \href{https://creativecommons.org/licenses/by/4.0/}{CC BY 4.0 license}. The only change
+#' made to the original dataset is the renaming of columns.
+#'
+#' @format A data frame with 109053 rows and 501 variables:
 #' \describe{
-#'   \item{note_id}{A unique identifier for each note}
-#'   \item{description}{A description or chief concern}
-#'   \item{medical_specialty}{Medical specialty of the note}
-#'   \item{sample_name}{mtsamples.com note name}
-#'   \item{transcription}{Transcription of note text}
-#'   \item{keywords}{Keywords}
+#'   \item{cui}{A Unified Medical Language System (UMLS) Concept Unique Identifier (CUI)}
+#'   \item{emb_001}{Concept embedding vector #1}
+#'   \item{emb_002}{Concept embedding vector #2}
+#'   \item{...}{...}
+#'   \item{emb_500}{Concept embedding vector #500}
 #' }
-#' @source \url{https://www.kaggle.com/tboyle10/medicaltranscriptions/data}
+#' @source \url{https://figshare.com/s/00d69861786cd0156d81}
 'cui2vec_embeddings'
 
 #' Cui2vec concept definitions
 #'
-#' This dataset contains sample medical transcriptions for various medical specialties.
+#' This dataset contains definitions for the Unified Medical Language System (UMLS)
+#' Concept Unique Identifiers (CUIs). These come from Andrew Beam's
+#' \href{https://github.com/beamandrew/cui2vec}{cui2vec R package}.
 #'
-#' Acknowledgements
+#' License
 #'
-#' This data was scraped from mtsamples.com by Tara Boyle and is made available
-#' under a CC0: Public Domain license.
+#' This data is made available under a
+#' \href{https://github.com/beamandrew/cui2vec/blob/master/LICENSE.md}{MIT license}. The data
+#' is copyrighted in 2019 by Benjamin Kompa, Andrew Beam, and Allen Schmaltz. The only change
+#' made to the original dataset is the renaming of columns.
 #'
-#' @format A data frame with 4999 rows and 6 variables:
+#' @format A data frame with 3053795 rows and 3 variables:
 #' \describe{
-#'   \item{note_id}{A unique identifier for each note}
-#'   \item{description}{A description or chief concern}
-#'   \item{medical_specialty}{Medical specialty of the note}
-#'   \item{sample_name}{mtsamples.com note name}
-#'   \item{transcription}{Transcription of note text}
-#'   \item{keywords}{Keywords}
+#'   \item{cui}{A Unified Medical Language System (UMLS) Concept Unique Identifier (CUI)}
+#'   \item{semantic_type}{Semantic type of the CUI}
+#'   \item{definition}{Definition of the CUI}
 #' }
-#' @source \url{https://www.kaggle.com/tboyle10/medicaltranscriptions/data}
+#' @source \url{https://github.com/beamandrew/cui2vec}
 'cui2vec_definitions'

R/mtsamples.R

@@ -4,8 +4,11 @@
 #'
 #' Acknowledgements
 #'
-#' This data was scraped from mtsamples.com by Tara Boyle and is made available
-#' under a CC0: Public Domain license.
+#' This data was scraped from \href{https://mtsamples.com}{https://mtsamples.com} by Tara Boyle.
+#'
+#' License
+#' This data is made available under a
+#' \href{https://creativecommons.org/share-your-work/public-domain/cc0/}{CC0: Public Domain license}.
 #'
 #' @format A data frame with 4999 rows and 6 variables:
 #' \describe{

README.Rmd

@@ -29,20 +29,26 @@ You can install the GitHub version of clinspacy with:
 remotes::install_github('ML4LHS/clinspacy', INSTALL_opts = '--no-multiarch')
 ```
 
-## Example
+## Examples
 
 ```{r}
 library(clinspacy)
 
 clinspacy('This patient has diabetes and CKD stage 3 but no HTN.')
+
+clinspacy('This patient is taking omeprazole, Protonix, and lisinopril 10 mg. He has diabetes.',
+          semantic_types = 'Disease or Syndrome')
+
+clinspacy('This patient is taking omeprazole, Protonix, and lisinopril 10 mg. He has diabetes.',
+          semantic_types = 'Pharmacologic Substance')
 ```
 
 ## Using the mtsamples dataset
 
 ```{r}
 data(mtsamples)
 
-str(mtsamples[1:5,])
+mtsamples[1:5,]
 ```
 
 
@@ -51,8 +57,30 @@ str(mtsamples[1:5,])
 This function binds columns containing concept unique identifiers with which scispacy has 99% confidence of being present with values containing frequencies. Negated concepts, as identified by negspacy's NegEx implementation, are ignored and do not count towards the frequencies.
 
 ```{r}
-mtsamples_with_cuis = bind_clinspacy(mtsamples[1:5,], text = 'description')
+bind_clinspacy(mtsamples[1:5, 1:2],
+               text = 'description')
 
-str(mtsamples_with_cuis)
+bind_clinspacy(mtsamples[1:5, 1:2],
+               text = 'description',
+               semantic_types = 'Diagnostic Procedure')
 ```
 
+## Binding Concept Embeddings to a Data Frame
+
+```{r}
+bind_clinspacy_embeddings(mtsamples[1:5, 1:2],
+                          text = 'description',
+                          num_embeddings = 5)
+
+bind_clinspacy_embeddings(mtsamples[1:5, 1:2],
+                          text = 'description',
+                          num_embeddings = 5,
+                          semantic_types = 'Diagnostic Procedure')
+```
+
+# UMLS CUI definitions
+
+```{r}
+data(cui2vec_definitions)
+head(cui2vec_definitions)
+```