A compressed large language model embedding dataset of ICD 10 CM descriptions
Abstract This paper presents novel datasets providing numerical representations of ICD-10-CM codes by generating description embeddings using a large language model followed by a dimension reduction via autoencoder. The embeddings serve as informative input features for machine learning models by ca...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2023-12-01
|
| Series: | BMC Bioinformatics |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12859-023-05597-2 |
| _version_ | 1797376754155782144 |
|---|---|
| author | Michael J. Kane Casey King Denise Esserman Nancy K. Latham Erich J. Greene David A. Ganz |
| author_facet | Michael J. Kane Casey King Denise Esserman Nancy K. Latham Erich J. Greene David A. Ganz |
| author_sort | Michael J. Kane |
| collection | DOAJ |
| description | Abstract This paper presents novel datasets providing numerical representations of ICD-10-CM codes by generating description embeddings using a large language model followed by a dimension reduction via autoencoder. The embeddings serve as informative input features for machine learning models by capturing relationships among categories and preserving inherent context information. The model generating the data was validated in two ways. First, the dimension reduction was validated using an autoencoder, and secondly, a supervised model was created to estimate the ICD-10-CM hierarchical categories. Results show that the dimension of the data can be reduced to as few as 10 dimensions while maintaining the ability to reproduce the original embeddings, with the fidelity decreasing as the reduced-dimension representation decreases. Multiple compression levels are provided, allowing users to choose as per their requirements, download and use without any other setup. The readily available datasets of ICD-10-CM codes are anticipated to be highly valuable for researchers in biomedical informatics, enabling more advanced analyses in the field. This approach has the potential to significantly improve the utility of ICD-10-CM codes in the biomedical domain. |
| first_indexed | 2024-03-08T19:43:17Z |
| format | Article |
| id | doaj.art-3ee01e62b0304e67a4c9f372574dfb9f |
| institution | Directory Open Access Journal |
| issn | 1471-2105 |
| language | English |
| last_indexed | 2024-03-08T19:43:17Z |
| publishDate | 2023-12-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Bioinformatics |
| spelling | doaj.art-3ee01e62b0304e67a4c9f372574dfb9f2023-12-24T12:30:57ZengBMCBMC Bioinformatics1471-21052023-12-0124111310.1186/s12859-023-05597-2A compressed large language model embedding dataset of ICD 10 CM descriptionsMichael J. Kane0Casey King1Denise Esserman2Nancy K. Latham3Erich J. Greene4David A. Ganz5Department of Biostatistics, School of Public Health, Yale UniversityThe Jackson School of Global Affairs, Yale UniversityDepartment of Biostatistics, School of Public Health, Yale UniversityResearch Program in Men’s Health: Aging and Metabolism, Boston Claude D. Pepper Older Americans Independence Center for Function Promoting Therapies, Brigham and Women’s HospitalDepartment of Biostatistics, School of Public Health, Yale UniversityDepartment of Medicine, VA Greater Los Angeles/UCLAAbstract This paper presents novel datasets providing numerical representations of ICD-10-CM codes by generating description embeddings using a large language model followed by a dimension reduction via autoencoder. The embeddings serve as informative input features for machine learning models by capturing relationships among categories and preserving inherent context information. The model generating the data was validated in two ways. First, the dimension reduction was validated using an autoencoder, and secondly, a supervised model was created to estimate the ICD-10-CM hierarchical categories. Results show that the dimension of the data can be reduced to as few as 10 dimensions while maintaining the ability to reproduce the original embeddings, with the fidelity decreasing as the reduced-dimension representation decreases. Multiple compression levels are provided, allowing users to choose as per their requirements, download and use without any other setup. The readily available datasets of ICD-10-CM codes are anticipated to be highly valuable for researchers in biomedical informatics, enabling more advanced analyses in the field. This approach has the potential to significantly improve the utility of ICD-10-CM codes in the biomedical domain.https://doi.org/10.1186/s12859-023-05597-2Large language modelAutoencoderICD-10-CMElectronic health recordsEHRNLP |
| spellingShingle | Michael J. Kane Casey King Denise Esserman Nancy K. Latham Erich J. Greene David A. Ganz A compressed large language model embedding dataset of ICD 10 CM descriptions BMC Bioinformatics Large language model Autoencoder ICD-10-CM Electronic health records EHR NLP |
| title | A compressed large language model embedding dataset of ICD 10 CM descriptions |
| title_full | A compressed large language model embedding dataset of ICD 10 CM descriptions |
| title_fullStr | A compressed large language model embedding dataset of ICD 10 CM descriptions |
| title_full_unstemmed | A compressed large language model embedding dataset of ICD 10 CM descriptions |
| title_short | A compressed large language model embedding dataset of ICD 10 CM descriptions |
| title_sort | compressed large language model embedding dataset of icd 10 cm descriptions |
| topic | Large language model Autoencoder ICD-10-CM Electronic health records EHR NLP |
| url | https://doi.org/10.1186/s12859-023-05597-2 |
| work_keys_str_mv | AT michaeljkane acompressedlargelanguagemodelembeddingdatasetoficd10cmdescriptions AT caseyking acompressedlargelanguagemodelembeddingdatasetoficd10cmdescriptions AT deniseesserman acompressedlargelanguagemodelembeddingdatasetoficd10cmdescriptions AT nancyklatham acompressedlargelanguagemodelembeddingdatasetoficd10cmdescriptions AT erichjgreene acompressedlargelanguagemodelembeddingdatasetoficd10cmdescriptions AT davidaganz acompressedlargelanguagemodelembeddingdatasetoficd10cmdescriptions AT michaeljkane compressedlargelanguagemodelembeddingdatasetoficd10cmdescriptions AT caseyking compressedlargelanguagemodelembeddingdatasetoficd10cmdescriptions AT deniseesserman compressedlargelanguagemodelembeddingdatasetoficd10cmdescriptions AT nancyklatham compressedlargelanguagemodelembeddingdatasetoficd10cmdescriptions AT erichjgreene compressedlargelanguagemodelembeddingdatasetoficd10cmdescriptions AT davidaganz compressedlargelanguagemodelembeddingdatasetoficd10cmdescriptions |