Clinical Abbreviations Research Articles

Disambiguating Clinical Abbreviations by One-to-All Classification: Algorithm Development and Validation Study.

Electronic medical records store extensive patient data and serve as a comprehensive repository, including textual medical records like surgical and imaging reports. Their utility in clinical decision support systems is substantial, but the widespread use of ambiguous and unstandardized abbreviations in clinical documents poses challenges for natural language processing in clinical decision support systems. Efficient abbreviation disambiguation methods are needed for effective information extraction. This study aims to enhance the one-to-all (OTA) framework for clinical abbreviation expansion, which uses a single model to predict multiple abbreviation meanings. The objective is to improve OTA by developing context-candidate pairs and optimizing word embeddings in Bidirectional Encoder Representations From Transformers (BERT), evaluating the model's efficacy in expanding clinical abbreviations using real data. Three datasets were used: Medical Subject Headings Word Sense Disambiguation, University of Minnesota, and Chia-Yi Christian Hospital from Ditmanson Medical Foundation Chia-Yi Christian Hospital. Texts containing polysemous abbreviations were preprocessed and formatted for BERT. The study involved fine-tuning pretrained models, ClinicalBERT and BlueBERT, generating dataset pairs for training and testing based on Huang et al's method. BlueBERT achieved macro- and microaccuracies of 95.41% and 95.16%, respectively, on the Medical Subject Headings Word Sense Disambiguation dataset. It improved macroaccuracy by 0.54%-1.53% compared to two baselines, long short-term memory and deepBioWSD with random embedding. On the University of Minnesota dataset, BlueBERT recorded macro- and microaccuracies of 98.40% and 98.22%, respectively. Against the baselines of Word2Vec + support vector machine and BioWordVec + support vector machine, BlueBERT demonstrated a macroaccuracy improvement of 2.61%-4.13%. This research preliminarily validated the effectiveness of the OTA method for abbreviation disambiguation in medical texts, demonstrating the potential to enhance both clinical staff efficiency and research effectiveness.

Improving clinical abbreviation sense disambiguation using attention-based Bi-LSTM and hybrid balancing techniques in imbalanced datasets.

Clinical abbreviations pose a challenge for clinical decision support systems due to their ambiguity. Additionally, clinical datasets often suffer from class imbalance, hindering the classification of such data. This imbalance leads to classifiers with low accuracy and high error rates. Traditional feature-engineered models struggle with this task, and class imbalance is a known factor that reduces the performance of neural network techniques. This study proposes an attention-based bidirectional long short-term memory (Bi-LSTM) model to improve clinical abbreviation disambiguation in clinical documents. We aim to address the challenges of limited training data and class imbalance by employing data generation techniques like reverse substitution and data augmentation with synonym substitution. We utilise a Bi-LSTM classification model with an attention mechanism to disambiguate each abbreviation. The model's performance is evaluated based on accuracy for each abbreviation. To address the limitations of imbalanced data, we employ data generation techniques to create a more balanced dataset. The evaluation results demonstrate that our data balancing technique significantly improves the model's accuracy by 2.08%. Furthermore, the proposed attention-based Bi-LSTM model achieves an accuracy of 96.09% on the UMN dataset, outperforming state-of-the-art results. Deep neural network methods, particularly Bi-LSTM, offer promising alternatives to traditional feature-engineered models for clinical abbreviation disambiguation. By employing data generation techniques, we can address the challenges posed by limited-resource and imbalanced clinical datasets. This approach leads to a significant improvement in model accuracy for clinical abbreviation disambiguation tasks.

Generative large language models are all-purpose text analytics engines: text-to-text learning is all your need.

To solve major clinical natural language processing (NLP) tasks using a unified text-to-text learning architecture based on a generative large language model (LLM) via prompt tuning. We formulated 7 key clinical NLP tasks as text-to-text learning and solved them using one unified generative clinical LLM, GatorTronGPT, developed using GPT-3 architecture and trained with up to 20 billion parameters. We adopted soft prompts (ie, trainable vectors) with frozen LLM, where the LLM parameters were not updated (ie, frozen) and only the vectors of soft prompts were updated, known as prompt tuning. We added additional soft prompts as a prefix to the input layer, which were optimized during the prompt tuning. We evaluated the proposed method using 7 clinical NLP tasks and compared them with previous task-specific solutions based on Transformer models. The proposed approach achieved state-of-the-art performance for 5 out of 7 major clinical NLP tasks using one unified generative LLM. Our approach outperformed previous task-specific transformer models by ∼3% for concept extraction and 7% for relation extraction applied to social determinants of health, 3.4% for clinical concept normalization, 3.4%-10% for clinical abbreviation disambiguation, and 5.5%-9% for natural language inference. Our approach also outperformed a previously developed prompt-based machine reading comprehension (MRC) model, GatorTron-MRC, for clinical concept and relation extraction. The proposed approach can deliver the "one model for all" promise from training to deployment using a unified generative LLM.

Leveraging Large Language Models for Clinical Abbreviation Disambiguation.

Clinical abbreviation disambiguation is a crucial task in the biomedical domain, as the accurate identification of the intended meanings or expansions of abbreviations in clinical texts is vital for medical information retrieval and analysis. Existing approaches have shown promising results, but challenges such as limited instances and ambiguous interpretations persist. In this paper, we propose an approach to address these challenges and enhance the performance of clinical abbreviation disambiguation. Our objective is to leverage the power of Large Language Models (LLMs) and employ a Generative Model (GM) to augment the dataset with contextually relevant instances, enabling more accurate disambiguation across diverse clinical contexts. We integrate the contextual understanding of LLMs, represented by BlueBERT and Transformers, with data augmentation using a Generative Model, called Biomedical Generative Pre-trained Transformer (BIOGPT), that is pretrained on an extensive corpus of biomedical literature to capture the intricacies of medical terminology and context. By providing the BIOGPT with relevant medical terms and sense information, we generate diverse instances of clinical text that accurately represent the intended meanings of abbreviations. We evaluate our approach on the widely recognized CASI dataset, carefully partitioned into training, validation, and test sets. The incorporation of data augmentation with the GM improves the model's performance, particularly for senses with limited instances, effectively addressing dataset imbalance and challenges posed by similar concepts. The results demonstrate the efficacy of our proposed method, showcasing the significance of LLMs and generative techniques in clinical abbreviation disambiguation. Our model achieves a good accuracy on the test set, outperforming previous methods.

Deciphering clinical abbreviations with a privacy protecting machine learning system

Physicians write clinical notes with abbreviations and shorthand that are difficult to decipher. Abbreviations can be clinical jargon (writing “HIT” for “heparin induced thrombocytopenia”), ambiguous terms that require expertise to disambiguate (using “MS” for “multiple sclerosis” or “mental status”), or domain-specific vernacular (“cb” for “complicated by”). Here we train machine learning models on public web data to decode such text by replacing abbreviations with their meanings. We report a single translation model that simultaneously detects and expands thousands of abbreviations in real clinical notes with accuracies ranging from 92.1%-97.1% on multiple external test datasets. The model equals or exceeds the performance of board-certified physicians (97.6% vs 88.7% total accuracy). Our results demonstrate a general method to contextually decipher abbreviations and shorthand that is built without any privacy-compromising data.

Automatic Prediction of Severity Score of Incident Learning Reports in Radiation Oncology Using Natural Language Processing

<h3>Purpose/Objective(s)</h3> Radiation oncology incident reports are an important tool for safety and quality improvement, but manual review is time-consuming and has low inter-rater reliability. In this study, we evaluated the ability of machine learning (ML) models to identify high-severity safety reports. <h3>Materials/Methods</h3> We employed two sets of manually labeled near-miss incident reports. The Institutional dataset (Inst.) has 7,094 reports from 2012-2021, graded with a severity score of 0 (no potential) - 4 (critical). The international SAFRON set has 571 reports graded in six levels from "minor" to "critical". We consider here two categories: high-severity reports (Institutional score 3-4 and SAFRON 2-6) and lower-severity reports. Each dataset was split into train, validation, and test sets. We compared two models: a Support Vector Machine (SVM) using TF*IDF text encodings and a transformer model pre-trained on PubMed text and clinical notes (BlueBERT) which is a state-of-the-art NLP method. We report performance metrics of the area under the receiver operating characteristic curve (AUC), which integrates the diagnostic ability of a binary classifier at all operating points, and F1, which is the harmonic mean of precision and recall. <h3>Results</h3> Table 1. Performance metrics for Institutional (15%) and SAFRON (20%) test sets, by model architecture and training set. ML methods were able to correctly predict severity with AUC scores of up to 0.720 on the institutional test set and 0.689 on the SAFRON test set, with BlueBERT outperforming SVM on both datasets. BlueBERT is able to generalize to the unseen dataset better than SVM. For example, BlueBERT (Inst.)—which refers to BlueBERT trained on Inst. data—outperforms SVM (Inst.) when it is applied to the SAFRON data on which it is not trained (Table 1: Rows 1 and 3). Similarly, BlueBERT (SAFRON) outperforms SVM (SAFRON) on the Inst. data (Table 1: Rows 2 and 4). The modest performance gain of BlueBERT may reflect clinical abbreviations and jargon that are not present in the pre-training corpus. Performance is limited by label noise due to multiple raters over 9 years. For comparison, Mullen et al. (2015) reported 37.6% inter-rater agreement for near-miss incident reports. <h3>Conclusion</h3> We developed two different ML models to score radiation oncology incident reports and achieved AUCs of up to 0.72 and F1 scores around 0.50. Our results suggest that BlueBERT is able to perform transfer learning across the institutional and SAFRON datasets. Our future work will focus on pre-training with other radiation oncology text using unsupervised learning to improve model performance to be used for triaging incident reports in clinical practice and in national and international databases such as SAFRON or RO-ILS.

Disambiguating Clinical Abbreviations Using a One-Fits-All Classifier Based on Deep Learning Techniques.

Background Abbreviations are considered an essential part of the clinical narrative; they are used not only to save time and space but also to hide serious or incurable illnesses. Misreckoning interpretation of the clinical abbreviations could affect different aspects concerning patients themselves or other services like clinical support systems. There is no consensus in the scientific community to create new abbreviations, making it difficult to understand them. Disambiguate clinical abbreviations aim to predict the exact meaning of the abbreviation based on context, a crucial step in understanding clinical notes. Objectives Disambiguating clinical abbreviations is an essential task in information extraction from medical texts. Deep contextualized representations models showed promising results in most word sense disambiguation tasks. In this work, we propose a one-fits-all classifier to disambiguate clinical abbreviations with deep contextualized representation from pretrained language models like Bidirectional Encoder Representation from Transformers (BERT). Methods A set of experiments with different pretrained clinical BERT models were performed to investigate fine-tuning methods on the disambiguation of clinical abbreviations. One-fits-all classifiers were used to improve disambiguating rare clinical abbreviations. Results One-fits-all classifiers with deep contextualized representations from Bioclinical, BlueBERT, and MS_BERT pretrained models improved the accuracy using the University of Minnesota data set. The model achieved 98.99, 98.75, and 99.13%, respectively. All the models outperform the state-of-the-art in the previous work of around 98.39%, with the best accuracy using the MS_BERT model. Conclusion Deep contextualized representations via fine-tuning of pretrained language modeling proved its sufficiency on disambiguating clinical abbreviations; it could be robust for rare and unseen abbreviations and has the advantage of avoiding building a separate classifier for each abbreviation. Transfer learning can improve the development of practical abbreviation disambiguation systems.

Disrupt, innovate, and transform in the era of COVID‐19

Today we live at an inflection point. For many, it's a world where things change at amazing speeds, from a patient's clinical status to a global position on risk mitigation. And it's this rate of change that has caught us either unprepared or unaware. While this notion of speed is far from new, in today's context, this speed is being defined by the emergence of COVID-19. Instant gratification is becoming less about gratification and more about a new normal driven by social and clinical necessity. Thus, I want what I when when i want it (IWWIWWIWI) may become a new clinical abbreviation that competes with the good old PRN. In this case, it's I want what I want when I want it. Perhaps we can even extend this from I want to I need. Some clinicians may remember the early days of arterial blood gas testing. The puncture of the radial artery with glass syringe, the bag of ice, and the rapid trip to the laboratory in the basement of the hospital defined an element of both technological sophistication and foot speed. Today, the glass syringe and quick footwork has, in many instances, been replaced by technology. But the story is much richer than a simple advancement of technology. This transformation has moved beyond just this blood test to the new reality of patient and consumer empowerment. Pulse oximetry has become the domain of both the physician and the consumer. Yes, I can check my oxygen partial pressure while I'm typing on my keyboard. Technology is becoming less an option and more of an imperative. From robotic microsurgery to artificial intelligence used to read chest x-rays, traditionally human tasks are being supported--if not replaced--by technological advances. Technology now offers us the opportunity to enhance many aspects of clinical practice from physical agility to cognitive function. As technology and artificial intelligence advance, the burden of many aspects of care can shift, freeing up the clinician for a higher and richer level of patient engagement. The heavy cognitive lifting of medicine may very well be transformed from medical school and on up revealing a new dynamic that facilitates adaptive thinking versus that conventional element of bookish memorization. Innovation and technology are now part and parcel of the practice of medicine. In many ways, has made us more human and these advantages extend to everyday tasks. Technology allows us to expand the richness of life to experience more sights, sounds, thoughts, and perhaps other senses that we have not even discovered. And it's these same advances that will allow medicine to evolve the personal relationship that exists between clinician and patient, and even the relationship between clinician and technology itself. And these profound changes, driven by technological advances, that create a new and richer reality, expanding our perceptions and introducing new sensory and computational skills to our human armamentarium. At the heart of our technological revolution are data. Some may argue that data are the new oil or even the new currency. But it might be better to consider data as a fundamental window into humanity. Just as the telescope and the microscope provided a new and eye-opening view of the universe, data will allow us to see vast connectivity and complexity that can define humanity. One simple comparison to today's COVID-19 pandemic and we can quickly see how the role of data allows us to tackle this disease. From bench science to social distancing, data are our best shot to understand and mitigate this disease. Yet the tsunami of information is simply beyond the comprehension of any human, even that uber-human in the white coat. The solution is not to work harder but to work smarter, and shift the burden to technology and AI. When the shift happens, another magical event will emerge to the surprise of clinicians and patients alike,the rise of empathy and engagement. Clinical senses—partnered with human sensory activity—will allow clinicians to touch with robotic exactness, see with electronic lens clarity, hear with digital fidelity, and listen with exacting sensitivity. What emerges is nothing even close to robotic but the advancement and extension of human capability. Today's technology dynamic forces the clinician to engage with a computer screen with their backs turned to the patient. Tomorrow's reality is that technology will drive aspects of care, like the physical exam, or peering into a microscope, and allow physicians to discover a richer and deeper relationship with medicine and mankind. And it's this realization that allows us to discover that innovation and technology aren't our masters, but tools to enhance our skills and improve care. In the final analysis, innovation is infectious and there is no cure.

Clinical Abbreviation Disambiguation Using Deep Contextualized Representation.

The objective of this study is to develop a method for clinical abbreviation disambiguation using deep contextualized representation and cluster analysis. We employed the pre-trained BioELMo language model to generate the contextualized word vector for abbreviations within each instance. Then principal component analysis was conducted on word vectors to reduce the dimension. K-Means cluster analysis was conducted for each abbreviation and the sense for a cluster was assigned based on the majority vote of annotations. Our method achieved an average accuracy of around 95% in 74 abbreviations. Simulation showed that each cluster required the annotation of 5 samples to determine its sense.

A method for harmonization of clinical abbreviation and acronym sense inventories.

Previous research has developed methods to construct acronym sense inventories from a single institutional corpus. Although beneficial, a sense inventory constructed from a single institutional corpus is not generalizable, because acronyms from different geographic regions and medical specialties vary greatly. Develop an automated method to harmonize sense inventories from different regions and specialties towards the development of a comprehensive inventory. The method involves integrating multiple source sense inventories into one centralized inventory and cross-mapping redundant entries to establish synonymy. To evaluate our method, we integrated 8 well-known source inventories into one comprehensive inventory (or metathesaurus). For both the metathesaurus and its sources, we evaluated the coverage of acronyms and their senses on a corpus of 1 million clinical notes. The corpus came from a different institution, region, and specialty than the source inventories. In the evaluation using clinical notes, the metathesaurus demonstrated an acronym (short form) micro-coverage of 94.3%, representing a substantial increase over the two next largest source inventories, the UMLS LRABR (74.8%) and ADAM (68.0%). The metathesaurus demonstrated a sense (long form) micro-coverage of 99.6%, again a substantial increase compared to the UMLS LRABR (82.5%) and ADAM (55.4%). Given the high coverage, harmonizing acronym sense inventories is a promising methodology to improve their comprehensiveness. Our method is automated, leverages the extensive resources already devoted to developing institution-specific inventories in the United States, and may help generalize sense inventories to institutions who lack the resources to develop them. Future work should address quality issues in source inventories and explore additional approaches to establishing synonymy.

A long journey to short abbreviations: developing an open-source framework for clinical abbreviation recognition and disambiguation (CARD).

The goal of this study was to develop a practical framework for recognizing and disambiguating clinical abbreviations, thereby improving current clinical natural language processing (NLP) systems' capability to handle abbreviations in clinical narratives. We developed an open-source framework for clinical abbreviation recognition and disambiguation (CARD) that leverages our previously developed methods, including: (1) machine learning based approaches to recognize abbreviations from a clinical corpus, (2) clustering-based semiautomated methods to generate possible senses of abbreviations, and (3) profile-based word sense disambiguation methods for clinical abbreviations. We applied CARD to clinical corpora from Vanderbilt University Medical Center (VUMC) and generated 2 comprehensive sense inventories for abbreviations in discharge summaries and clinic visit notes. Furthermore, we developed a wrapper that integrates CARD with MetaMap, a widely used general clinical NLP system. CARD detected 27 317 and 107 303 distinct abbreviations from discharge summaries and clinic visit notes, respectively. Two sense inventories were constructed for the 1000 most frequent abbreviations in these 2 corpora. Using the sense inventories created from discharge summaries, CARD achieved an F1 score of 0.755 for identifying and disambiguating all abbreviations in a corpus from the VUMC discharge summaries, which is superior to MetaMap and Apache's clinical Text Analysis Knowledge Extraction System (cTAKES). Using additional external corpora, we also demonstrated that the MetaMap-CARD wrapper improved MetaMap's performance in recognizing disorder entities in clinical notes. The CARD framework, 2 sense inventories, and the wrapper for MetaMap are publicly available at https://sbmi.uth.edu/ccb/resources/abbreviation.htm . We believe the CARD framework can be a valuable resource for improving abbreviation identification in clinical NLP systems.

A Preliminary Study of Clinical Abbreviation Disambiguation in Real Time.

To save time, healthcare providers frequently use abbreviations while authoring clinical documents. Nevertheless, abbreviations that authors deem unambiguous often confuse other readers, including clinicians, patients, and natural language processing (NLP) systems. Most current clinical NLP systems "post-process" notes long after clinicians enter them into electronic health record systems (EHRs). Such post-processing cannot guarantee 100% accuracy in abbreviation identification and disambiguation, since multiple alternative interpretations exist. Authors describe a prototype system for real-time Clinical Abbreviation Recognition and Disambiguation (rCARD) - i.e., a system that interacts with authors during note generation to verify correct abbreviation senses. The rCARD system design anticipates future integration with web-based clinical documentation systems to improve quality of healthcare records. When clinicians enter documents, rCARD will automatically recognize each abbreviation. For abbreviations with multiple possible senses, rCARD will show a ranked list of possible meanings with the best predicted sense at the top. The prototype application embodies three word sense disambiguation (WSD) methods to predict the correct senses of abbreviations. We then conducted three experments to evaluate rCARD, including 1) a performance evaluation of different WSD methods; 2) a time evaluation of real-time WSD methods; and 3) a user study of typing clinical sentences with abbreviations using rCARD. Using 4,721 sentences containing 25 commonly observed, highly ambiguous clinical abbreviations, our evaluation showed that the best profile-based method implemented in rCARD achieved a reasonable WSD accuracy of 88.8% (comparable to SVM - 89.5%) and the cost of time for the different WSD methods are also acceptable (ranging from 0.630 to 1.649 milliseconds within the same network). The preliminary user study also showed that the extra time costs by rCARD were about 5% of total document entry time and users did not feel a significant delay when using rCARD for clinical document entry. The study indicates that it is feasible to integrate a real-time, NLP-enabled abbreviation recognition and disambiguation module with clinical documentation systems.

A sense inventory for clinical abbreviations and acronyms created using clinical notes and medical dictionary resources.

To create a sense inventory of abbreviations and acronyms from clinical texts. The most frequently occurring abbreviations and acronyms from 352,267 dictated clinical notes were used to create a clinical sense inventory. Senses of each abbreviation and acronym were manually annotated from 500 random instances and lexically matched with long forms within the Unified Medical Language System (UMLS V.2011AB), Another Database of Abbreviations in Medline (ADAM), and Stedman's Dictionary, Medical Abbreviations, Acronyms & Symbols, 4th edition (Stedman's). Redundant long forms were merged after they were lexically normalized using Lexical Variant Generation (LVG). The clinical sense inventory was found to have skewed sense distributions, practice-specific senses, and incorrect uses. Of 440 abbreviations and acronyms analyzed in this study, 949 long forms were identified in clinical notes. This set was mapped to 17,359, 5233, and 4879 long forms in UMLS, ADAM, and Stedman's, respectively. After merging long forms, only 2.3% matched across all medical resources. The UMLS, ADAM, and Stedman's covered 5.7%, 8.4%, and 11% of the merged clinical long forms, respectively. The sense inventory of clinical abbreviations and acronyms and anonymized datasets generated from this study are available for public use at http://www.bmhi.umn.edu/ihi/research/nlpie/resources/index.htm ('Sense Inventories', website). Clinical sense inventories of abbreviations and acronyms created using clinical notes and medical dictionary resources demonstrate challenges with term coverage and resource integration. Further work is needed to help with standardizing abbreviations and acronyms in clinical care and biomedicine to facilitate automated processes such as text-mining and information extraction.

A new clustering method for detecting rare senses of abbreviations in clinical notes

Abbreviations are widely used in clinical documents and they are often ambiguous. Building a list of possible senses (also called sense inventory) for each ambiguous abbreviation is the first step to automatically identify correct meanings of abbreviations in given contexts. Clustering based methods have been used to detect senses of abbreviations from a clinical corpus [1]. However, rare senses remain challenging and existing algorithms are not good enough to detect them. In this study, we developed a new two-phase clustering algorithm called Tight Clustering for Rare Senses (TCRS) and applied it to sense generation of abbreviations in clinical text. Using manually annotated sense inventories from a set of 13 ambiguous clinical abbreviations, we evaluated and compared TCRS with the existing Expectation Maximization (EM) clustering algorithm for sense generation, at two different levels of annotation cost (10 vs. 20 instances for each abbreviation). Our results showed that the TCRS-based method could detect 85% senses on average; while the EM-based method found only 75% senses, when similar annotation effort (about 20 instances) was used. Further analysis demonstrated that the improvement by the TCRS method was mainly from additionally detected rare senses, thus indicating its usefulness for building more complete sense inventories of clinical abbreviations.

Methods for Building Sense Inventories of Abbreviations in Clinical Notes

ObjectiveTo develop methods for building corpus-specific sense inventories of abbreviations occurring in clinical documents. DesignA corpus of internal medicine admission notes was collected and instances of each clinical abbreviation in the corpus were clustered to different sense clusters. One instance from each cluster was manually annotated to generate a final list of senses. Two clustering-based methods (Expectation Maximization—EM and Farthest First—FF) and one random sampling method for sense detection were evaluated using a set of 12 clinical abbreviations. MeasurementsThe clustering-based sense detection methods were evaluated using a set of clinical abbreviations that were manually sense annotated. “Sense Completeness” and “Annotation Cost” were used to measure the performance of different methods. Clustering error rates were also reported for different clustering algorithms. ResultsA clustering-based semi-automated method was developed to build corpus-specific sense inventories for abbreviations in hospital admission notes. Evaluation demonstrated that this method could largely reduce manual annotation cost and increase the completeness of sense inventories when compared with a manual annotation method using random samples. ConclusionThe authors developed an effective clustering-based method for building corpus-specific sense inventories for abbreviations in a clinical corpus. To the best of the authors knowledge, this is the first time clustering technologies have been used to help building sense inventories of abbreviations in clinical text. The results demonstrated that the clustering-based method performed better than the manual annotation method using random samples for the task of building sense inventories of clinical abbreviations.

Bringing Histology to the Bedside

Bringing Histology to the Bedside