Snomed2Vec: Representation of SNOMED CT Terms with Word2Vec

Soriano, Ignacio Martinez; Peña, Juan Luis Castro; Fernández-Breis, Jesualdo Tomás; Roman, Ignacio San; Barriuso, Adrian Alonso; Baraza, David Guevara

doi:10.1109/cbms.2019.00138

Cited by 7 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the context of capturing climate change impacts on health, capturing standardized environmentally-specific clinical data permits the contextualization of vast amounts of data to predict service demands and population needs during natural disasters and climate-related events, health promotion and disease prevention and benchmarking of any subsequent interventions. Researchers have also demonstrated the use of these terminologies in developing or training models based on methods broadly defined as artificial intelligence (AI) [33,34]. SNOMED CT, for example, has been used in natural language processing tasks, utilizing machine learning methodologies to search and analyze free text clinical documentation entries [35].…”

Section: Concept Representation For Artificial Intelligence and Data ...mentioning

confidence: 99%

Representation of Environmental Concepts Associated with Health Impacts in Computer Standardized Clinical Terminologies

Block,

Lozada-Perezmitre,

Cho

et al. 2023

Yearb Med Inform

View full text Add to dashboard Cite

Objective: To evaluate the representation of environmental concepts associated with health impacts in standardized clinical terminologies. Methods: This study used a descriptive approach with methods informed by a procedural framework for standardized clinical terminology mapping. The United Nations Global Indicator Framework for the Sustainable Development Goals and Targets was used as the source document for concept extraction. The target terminologies were the Systematized Nomenclature of Medicine Clinical Terms (SNOMED CT) and the International Classification for Nursing Practice (ICNP). Manual and automated mapping methods were utilized. The lists of candidate matches were reviewed and iterated until a final mapping match list was achieved. Results: A total of 119 concepts with 133 mapping matches were added to the final SNOMED CT list. Fifty-three (39.8%) were direct matches, 37 (27.8%) were narrower than matches, 35 (26.3%) were broader than matches, and 8 (6%) had no matches. A total of 26 concepts with 27 matches were added to the final ICNP list. Eight (29.6%) were direct matches, 4 (14.8%) were narrower than, 7 (25.9%) were broader than, and 8 (29.6%) were no matches. Conclusion: Following this evaluation, both strengths and gaps were identified. Gaps in terminology representation included concepts related to cost expenditures, affordability, community engagement, water, air and sanitation. The inclusion of these concepts is necessary to advance the clinical reporting of these environmental and sustainability indicators. As environmental concepts encoded in standardized terminologies expand, additional insights into data and health conditions, research, education, and policy-level decision-making will be identified.

show abstract

Section: Concept Representation For Artificial Intelligence and Data ...mentioning

confidence: 99%

Representation of Environmental Concepts Associated with Health Impacts in Computer Standardized Clinical Terminologies

Block,

Lozada-Perezmitre,

Cho

et al. 2023

Yearb Med Inform

View full text Add to dashboard Cite

show abstract

A Survey of Deep Learning Based Natural Language Processing in Smart Healthcare

M’hamdi¹,

Lazaar²,

Mahboub³

2023

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

Optimizing word embeddings for small datasets: a case study on patient portal messages from breast cancer patients

Song,

Ni,

Warner

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Patient portal messages often relate to specific clinical phenomena (e.g., patients undergoing treatment for breast cancer) and, as a result, have received increasing attention in biomedical research. These messages require natural language processing and, while word embedding models, such as word2vec, have the potential to extract meaningful signals from text, they are not readily applicable to patient portal messages. This is because embedding models typically require millions of training samples to sufficiently represent semantics, while the volume of patient portal messages associated with a particular clinical phenomenon is often relatively small. We introduce a novel adaptation of the word2vec model, PK-word2vec (where PK stands for prior knowledge), for small-scale messages. PK-word2vec incorporates the most similar terms for medical words (including problems, treatments, and tests) and non-medical words from two pre-trained embedding models as prior knowledge to improve the training process. We applied PK-word2vec in a case study of patient portal messages in the Vanderbilt University Medical Center electric health record system sent by patients diagnosed with breast cancer from December 2004 to November 2017. We evaluated the model through a set of 1000 tasks, each of which compared the relevance of a given word to a group of the five most similar words generated by PK-word2vec and a group of the five most similar words generated by the standard word2vec model. We recruited 200 Amazon Mechanical Turk (AMT) workers and 7 medical students to perform the tasks. The dataset was composed of 1389 patient records and included 137,554 messages with 10,683 unique words. Prior knowledge was available for 7981 non-medical and 1116 medical words. In over 90% of the tasks, both reviewers indicated PK-word2vec generated more similar words than standard word2vec (p = 0.01).The difference in the evaluation by AMT workers versus medical students was negligible for all comparisons of tasks’ choices between the two groups of reviewers ($${\text{p}} = 0.774$$ p = 0.774 under a paired t-test). PK-word2vec can effectively learn word representations from a small message corpus, marking a significant advancement in processing patient portal messages.

show abstract

Snomed2Vec: Representation of SNOMED CT Terms with Word2Vec

Cited by 7 publications

References 14 publications

Representation of Environmental Concepts Associated with Health Impacts in Computer Standardized Clinical Terminologies

Representation of Environmental Concepts Associated with Health Impacts in Computer Standardized Clinical Terminologies

A Survey of Deep Learning Based Natural Language Processing in Smart Healthcare

Optimizing word embeddings for small datasets: a case study on patient portal messages from breast cancer patients

Contact Info

Product

Resources

About