Kalyan KS

provides a comprehensive survey of various transformer-based biomedical pretrained language models.

In the last few years, people started to share lots of information related to health in the form of tweets, reviews and blog posts. All these user generated clinical texts can be mined to generate useful insights. However, automatic analysis of clinical text requires identification of standard medical concepts. Most of the existing deep learning based medical concept normalization systems are based on CNN or RNN. Performance of these models is limited as they have to be trained from scratch… Show more

In the last few years, people started to share lots of information related to health in the form of tweets, reviews and blog posts. All these user generated clinical texts can be mined to generate useful insights. However, automatic analysis of clinical text requires identification of standard medical concepts. Most of the existing deep learning based medical concept normalization systems are based on CNN or RNN. Performance of these models is limited as they have to be trained from scratch (except embeddings). In this work, we propose a medical concept normalization system based on BERT and highway layer. BERT, a pre-trained context sensitive deep language representation model advanced state-of-the-art performance in many NLP tasks and gating mechanism in highway layer helps the model to choose only important information. Experimental results show that our model outperformed all existing methods on two standard datasets. Further, we conduct a series of experiments to study the impact of different learning rates and batch sizes, noise and freezing encoder layers on our model. Show less

Pattisapu et al. (2020) formulate medical concept normalization (MCN) as text similarity problem and propose a model based on RoBERTa and graph embedding based target concept vectors. However, graph embedding techniques ignore valuable information available in the clinical ontology like concept description and synonyms. In this work, we enhance the model of Pattisapu et al. (2020) with two novel changes. First, we use retrofitted target concept vectors instead of graph embedding based vectors… Show more

Pattisapu et al. (2020) formulate medical concept normalization (MCN) as text similarity problem and propose a model based on RoBERTa and graph embedding based target concept vectors. However, graph embedding techniques ignore valuable information available in the clinical ontology like concept description and synonyms. In this work, we enhance the model of Pattisapu et al. (2020) with two novel changes. First, we use retrofitted target concept vectors instead of graph embedding based vectors. It is the first work to leverage both concept description and synonyms to represent concepts in the form of retrofitted target concept vectors in text similarity framework based social media MCN. Second, we generate both concept and concept mention vectors with same size which eliminates the need of dense layers to project concept mention vectors into the target concept embedding space. Our model outperforms existing methods with improvements up to 3.75% on two standard datasets. Further when trained only on mapping lexicon synonyms, our model outperforms existing methods with significant improvements up to 14.61%. We attribute these significant improvements to the two novel changes introduced. Show less

This paper presents our approach for task 2 and task 3 of Social Media Mining for Health (SMM4H) 2020 shared tasks. In task 2, we have to differentiate adverse drug reaction (ADR) tweets from nonADR tweets and is treated as binary classification. Task 3 involves extracting ADR mentions and then mapping them to MedDRA codes. Extracting ADR mentions is treated as sequence labeling and normalizing ADR mentions is treated as multi-class classification. Our system is based on pre-trained language… Show more

This paper presents our approach for task 2 and task 3 of Social Media Mining for Health (SMM4H) 2020 shared tasks. In task 2, we have to differentiate adverse drug reaction (ADR) tweets from nonADR tweets and is treated as binary classification. Task 3 involves extracting ADR mentions and then mapping them to MedDRA codes. Extracting ADR mentions is treated as sequence labeling and normalizing ADR mentions is treated as multi-class classification. Our system is based on pre-trained language model RoBERTa and it achieves a) F1-score of 58% in task 2 which is 12% more than the average score b) relaxed F1-score of 70.1% in ADR extraction of task 3 which is 13.7% more than the average score and relaxed F1-score of 35% in ADR extraction + normalization of task 3 which is 5.8% more than the average score. Overall, our models achieve promising results in both the tasks with significant improvements over average scores. Show less

Medical concept normalization helps in discovering standard concepts in free-form text i.e., maps health-related mentions to standard concepts in a clinical knowledge base. It is much beyond simple string matching and requires a deep semantic understanding of concept mentions. Recent research approach concept normalization as either text classification or text similarity. The main drawback in existing a) text classification approach is ignoring valuable target concepts information in learning… Show more

Medical concept normalization helps in discovering standard concepts in free-form text i.e., maps health-related mentions to standard concepts in a clinical knowledge base. It is much beyond simple string matching and requires a deep semantic understanding of concept mentions. Recent research approach concept normalization as either text classification or text similarity. The main drawback in existing a) text classification approach is ignoring valuable target concepts information in learning input concept mention representation b) text similarity approach is the need to separately generate target concept embeddings which is time and resource consuming. Our proposed model overcomes these drawbacks by jointly learning the representations of input concept mention and target concepts. First, we learn input concept mention representation using RoBERTa. Second, we find cosine similarity between embeddings of input concept mention and all the target concepts. Here, embeddings of target concepts are randomly initialized and then updated during training. Finally, the target concept with maximum cosine similarity is assigned to the input concept mention. Our model surpasses all the existing methods across three standard datasets by improving accuracy up to 2.31%. Show less

Medical concept normalization (MCN) i.e., mapping of colloquial medical phrases to standard concepts is an essential step in analysis of medical social media text. The main drawback in existing state-of-the-art approach (Kalyan and Sangeetha, 2020b) is learning target concept vector representations from scratch which requires more training instances. Our model is based on RoBERTa and target concept embeddings. In our model, we integrate a) target concept information in the form of target… Show more

Medical concept normalization (MCN) i.e., mapping of colloquial medical phrases to standard concepts is an essential step in analysis of medical social media text. The main drawback in existing state-of-the-art approach (Kalyan and Sangeetha, 2020b) is learning target concept vector representations from scratch which requires more training instances. Our model is based on RoBERTa and target concept embeddings. In our model, we integrate a) target concept information in the form of target concept vectors generated by encoding target concept descriptions using SRoBERTa, state-of-the-art RoBERTa based sentence embedding model and b) domain lexicon knowledge by enriching target concept vectors with synonym relationship knowledge using retrofitting algorithm. It is the first attempt in MCN to exploit both target concept information as well as domain lexicon knowledge in the form of retrofitted target concept vectors. Our model outperforms all the existing models with an accuracy improvement up to 1.36% on three standard datasets. Further, our model when trained only on mapping lexicon synonyms achieves up to 4.87% improvement in accuracy Show less

Distributed vector representations or embeddings map variable length text to dense fixed length vectors as well as capture prior knowledge which can transferred to downstream tasks. Even though embeddings have become de facto standard for text representation in deep learning based NLP tasks in both general and clinical domains, there is no survey paper which presents a detailed review of embeddings in Clinical Natural Language Processing. In this survey paper, we discuss various medical corpora… Show more

Distributed vector representations or embeddings map variable length text to dense fixed length vectors as well as capture prior knowledge which can transferred to downstream tasks. Even though embeddings have become de facto standard for text representation in deep learning based NLP tasks in both general and clinical domains, there is no survey paper which presents a detailed review of embeddings in Clinical Natural Language Processing. In this survey paper, we discuss various medical corpora and their characteristics, medical codes and present a brief overview as well as comparison of popular embeddings models. We classify clinical embeddings and discuss each embedding type in detail. We discuss various evaluation methods followed by possible solutions to various challenges in clinical embeddings. Finally, we conclude with some of the future directions which will advance research in clinical embeddings. Show less