A5050726436- Biomedical Text Mining and Ontologies
- Topic Modeling
- Bioinformatics and Genomic Networks
- Machine Learning in Healthcare
- Artificial Intelligence in Healthcare and Education
- Computational Drug Discovery Methods
- Natural Language Processing Techniques
- Artificial Intelligence in Healthcare
- Pharmacovigilance and Adverse Drug Reactions
- Protein Structure and Dynamics
- Colorectal Cancer Screening and Detection
- Vitamin C and Antioxidants Research
- Biosimilars and Bioanalytical Methods
- Machine Learning in Bioinformatics
- Heme Oxygenase-1 and Carbon Monoxide
- Folate and B Vitamins Research
- Genomics and Phylogenetic Studies
- Liver Disease Diagnosis and Treatment
- Porphyrin Metabolism and Disorders
- Bacterial Genetics and Biotechnology
- Text Readability and Simplification
- Medical Coding and Health Information
- Genetics, Bioinformatics, and Biomedical Research
- Computational and Text Analysis Methods
- Ion Transport and Channel Regulation
Florida State University
2024-2025
University of California, San Francisco
2023-2024
Bharathiar University
2016-2023
Although patients have easy access to their electronic health records and laboratory test result data through patient portals, results are often confusing hard understand. Many turn web-based forums or question-and-answer (Q&A) sites seek advice from peers. The quality of answers social Q&A on health-related questions varies significantly, not all responses accurate reliable. Large language models (LLMs) such as ChatGPT opened a promising avenue for answered.
Outpatient clinical notes are a rich source of information regarding drug safety. However, data in these currently underutilized for pharmacovigilance due to methodological limitations text mining. Large language models (LLMs) like Bidirectional Encoder Representations from Transformers (BERT) have shown progress range natural processing tasks but not yet been evaluated on adverse event (AE) detection. We adapted new LLM, University California – San Francisco (UCSF)‐BERT, identify serious...
A wealth of knowledge concerning relations between genes and its associated diseases is present in biomedical literature. Mining these biological associations from literature can provide immense support to research ranging drug-targetable pathways biomarker discovery. However, time cost manual curation heavily slows it down. In this current scenario one the crucial technologies text mining, relation extraction shows promising result explore with diseases. By developing automatic gene-disease...
A novel coronavirus (SARS-CoV-2) has caused a major outbreak in human all over the world. There are several proteins interplay during entry and replication of this virus human. Here, we have used text mining named entity recognition method to identify co-occurrence important COVID 19 genes/proteins interaction network based on frequency interaction. Network analysis revealed set genes/proteins, highly dense genes/protein clusters sub-networks Angiotensin-converting enzyme 2 (ACE2), Helicase,...
Vedolizumab (VDZ) and ustekinumab (UST) are second-line treatments in pediatric patients with ulcerative colitis (UC) refractory to antitumor necrosis factor (anti-TNF) therapy. Pediatric studies comparing the effectiveness of these medications lacking. Using a registry from ImproveCareNow (ICN), global research network inflammatory bowel disease, we compared UST VDZ anti-TNF UC.
Differential diagnosis (DDx) is crucial for medicine as it helps healthcare providers systematically distinguish between conditions that share similar symptoms. This study evaluates the influence of lab test results on DDx accuracy generated by large language models (LLMs). Clinical vignettes from 50 randomly selected case reports PMC-Patients were created, incorporating demographics, symptoms, and data. Five LLMs-GPT-4, GPT-3.5, Llama-2-70b, Claude-2, Mixtral-8x7B-were tested to generate...
The present study aimed to reveal the molecular mechanism of T-2 toxin-induced cerebral edema by aquaporin-4 (AQP4) blocking and permeation. AQP4 is a class aquaporin channels that mainly expressed in brain, its structural changes lead life-threatening complications such as cardio-respiratory arrest, nephritis, irreversible brain damage. We employed dynamics simulation, text mining, vitro vivo analysis functional induced toxin on AQP4. action leads disrupted permeation water coefficients are...
Tagging biomedical entities such as gene, protein, cell, and cell-line is the first step an important pre-requisite in literature mining. In this paper, we describe our hybrid named entity tagging approach namely BCC-NER (bidirectional, contextual clues tagger for gene/protein mention recognition). deployed with three modules. The module text processing which includes basic NLP pre-processing, feature extraction, selection. second training model building bidirectional conditional random...
Abstract Background The Mayo endoscopic subscore (MES) is an important quantitative measure of disease activity in ulcerative colitis. Colonoscopy reports routine clinical care usually characterize colitis using free text description, limiting their utility for research and quality improvement. We sought to develop algorithms classify colonoscopy according MES. Methods annotated 500 from 2 health systems. trained evaluated 4 classes algorithms. Our primary outcome was accuracy identifying...
Abstract Background Acute hepatic porphyria (AHP) is a group of rare but treatable conditions associated with diagnostic delays 15 years on average. The advent electronic health records (EHR) data and machine learning (ML) may improve the timely recognition diseases like AHP. However, prediction models can be difficult to train given limited case numbers, unstructured EHR data, selection biases intrinsic healthcare delivery. We sought characterize for identifying patients Methods This study...
Acute Hepatic Porphyria (AHP) is a group of rare but treatable conditions associated with diagnostic delays fifteen years on average. The advent electronic health records (EHR) data and machine learning (ML) may improve the timely recognition diseases like AHP. However, prediction models can be difficult to train given limited case numbers, unstructured EHR data, selection biases intrinsic healthcare delivery.
Abstract Background and Aims Outpatient clinical notes are a rich source of information regarding drug safety. However, data in these currently underutilized for pharmacovigilance due to methodological limitations text mining. Large language models (LLM) like BERT have shown progress range natural processing tasks but not yet been evaluated on adverse event detection. Methods We adapted new LLM, UCSF BERT, identify serious events (SAEs) occurring after treatment with non-steroid...
Breast cancer is still a major worldwide health con- cern, which emphasises how important it to have accurate and trustworthy diagnostic techniques. Our goal improve breast identification in this scenario by utilizing mammography pictures from the RSNA Cancer dataset. Leveraging power of deep learning, our methodology integrates SEResNet ConvNeXtV2 models, simultaneously exploring comparing various methods. To enhance performance we incorporated changes activation functions replacing ReLU...
<p></p><p></p><p>A novel coronavirus (SARS-CoV-2) has caused a major outbreak in human all over the world. There are several proteins interplay during entry and replication of this virus human. Here, we have used text mining named entity recognition method to identify co-occurrence important COVID 19 genes/proteins interaction network based on frequency interaction. Network analysis revealed set genes/proteins, highly dense genes/protein clusters sub-networks...
Abstract About 1 in 9 older adults over 65 has Alzheimer’s disease (AD), many of whom also have multiple other chronic conditions such as hypertension and diabetes, necessitating careful monitoring through laboratory tests. Understanding the patterns tests this population aids our understanding management these along with AD. In study, we used an unimodal cosinor model to assess seasonality lab using electronic health record (EHR) data from 34,303 AD patients OneFlorida+ Clinical Research...
<sec> <title>BACKGROUND</title> Although patients have easy access to their electronic health records and laboratory test result data through patient portals, results are often confusing hard understand. Many turn web-based forums or question-and-answer (Q&amp;A) sites seek advice from peers. The quality of answers social Q&amp;A on health-related questions varies significantly, not all responses accurate reliable. Large language models (LLMs) such as ChatGPT opened a promising...
microRNA (miRNA)–messenger RNA (mRNA or gene) interactions are pivotal in various biological processes, including the regulation of gene expression, cellular differentiation, proliferation, apoptosis, and development, as well maintenance homeostasis pathogenesis numerous diseases, such cancer, cardiovascular neurological disorders, metabolic conditions. Understanding mechanisms miRNA–mRNA can provide insights into disease potential therapeutic targets. However, extracting these efficiently...