Clinical text is rich in information, with mentions of treatment, medication and anatomy among many other clinical terms. Multiple terms can refer to the same core concepts which can be referred as a clinical entity. Ontologies like the Unified Medical Language System (UMLS) are developed and maintained to store millions of clinical entities including the definitions, relations and other corresponding information. These ontologies are used for standardization of clinical text by normalizing varying surface forms of a clinical term through Biomedical entity linking. With the introduction of transformer-based language models, there has been significant progress in Biomedical entity linking. In this work, we focus on learning through synonym pairs associated with the entities. As compared to the existing approaches, our approach significantly reduces the training data and resource consumption. Moreover, we propose a suite of context-based and context-less reranking techniques for performing the entity disambiguation. Overall, we achieve similar performance to the state-of-the-art zero-shot and distant supervised entity linking techniques on the Medmentions dataset, the largest annotated dataset on UMLS, without any domain-based training. Finally, we show that retrieval performance alone might not be sufficient as an evaluation metric and introduce an article level quantitative and qualitative analysis to reveal further insights on the performance of entity linking methods.

Extracting information from genomic reports of cancer patients is crucial for both healthcare professionals and cancer research. While Large Language Models (LLMs) have shown promise in extracting information, their potential for handling genomic reports remains unexplored. These reports are complex, multi-page documents that feature a variety of visually rich, structured layouts and contain many domain-specific terms. Two primary challenges complicate the process: (i) extracting data from PDFs with intricate layouts and domain-specific terminology and (ii) dealing with variations in report layouts from different laboratories, making extraction layout-dependent and posing challenges for subsequent data processing. To tackle these issues, we propose GR-PROMPT, a prompt-based technique, and GR-FORMAT, a standardized format. Together, these two convert a genomic report in PDF format into GR-FORMAT as a JSON file using a multimodal LLM. To address the lack of available datasets for this task, we introduce GR-DATASET, a synthetic collection of 100 cancer genomic reports in PDF format. Each report is accompanied by key-value information presented in a layout-specific format, as well as structured key-value information in GR-FORMAT. This is the first dataset in this domain to promote further research for the task. We performed our experiment on this dataset.