LLM-IE: a python package for biomedical generative information extraction with large language models.

Objectives: Despite the recent adoption of large language models (LLMs) for biomedical information extraction (IE), challenges in prompt engineering and algorithms persist, with no dedicated software available. To address this, we developed : a Python package for building complete IE pipelines.

Materials And Methods: The supports named entity recognition, entity attribute extraction, and relation extraction tasks.

Leveraging large language models for knowledge-free weak supervision in clinical natural language processing.

The performance of deep learning-based natural language processing systems is based on large amounts of labeled training data which, in the clinical domain, are not easily available or affordable. Weak supervision and in-context learning offer partial solutions to this issue, particularly using large language models (LLMs), but their performance still trails traditional supervised methods with moderate amounts of gold-standard data. In particular, inferencing with LLMs is computationally heavy.

Multimodal prehabilitation and postoperative outcomes in upper abdominal surgery: systematic review and meta-analysis.

The impact of multimodal prehabilitation on postoperative complications in upper abdominal surgeries is understudied. This review analyzes randomized trials on multimodal prehabilitation with patient and hospital outcomes. MEDLINE, Embase, CINAHL, and Cochrane CENTRAL were searched for trials on prehabilitation before elective (non-emergency) abdominal surgery.

Leveraging Large Language Models for Knowledge-free Weak Supervision in Clinical Natural Language Processing.

The performance of deep learning-based natural language processing systems is based on large amounts of labeled training data which, in the clinical domain, are not easily available or affordable. Weak supervision and in-context learning offer partial solutions to this issue, particularly using large language models (LLMs), but their performance still trails traditional supervised methods with moderate amounts of gold-standard data. In particular, inferencing with LLMs is computationally heavy.

Extraction of Radiological Characteristics From Free-Text Imaging Reports Using Natural Language Processing Among Patients With Ischemic and Hemorrhagic Stroke: Algorithm Development and Validation.

Background: Neuroimaging is the gold-standard diagnostic modality for all patients suspected of stroke. However, the unstructured nature of imaging reports remains a major challenge to extracting useful information from electronic health records systems. Despite the increasing adoption of natural language processing (NLP) for radiology reports, information extraction for many stroke imaging features has not been systematically evaluated.

Bloodstream infections after solid organ transplantation: clinical epidemiology and antimicrobial resistance (2016-21).

Background: Solid organ transplant (SOT) recipients are at risk of bloodstream infections (BSIs) with MDR organisms (MDROs).

Objectives: To describe the epidemiology of BSI in the year after several types of SOT, as well as the prevalence of MDRO infections in this population.

Methods: We conducted a single-centre, retrospective study of kidney, liver, heart, and multi-organ transplantation patients.

A Neuro-Informatics Pipeline for Cerebrovascular Disease: Research Registry Development.

Background: Although stroke is well recognized as a critical disease, treatment options are often limited. Inpatient stroke encounters carry critical information regarding the mechanisms of stroke and patient outcomes; however, these data are typically formatted to support administrative functions instead of research. To support improvements in the care of patients with stroke, a substantive research data platform is needed.

INTRAVITREAL ANTI-VASCULAR ENDOTHELIAL GROWTH FACTOR PHARMACEUTICAL UTILIZATION AND FINANCIAL IMPACT IN A VALUE-BASED CARE PROGRAM.

Purpose: To investigate trends and the potential impact of the COVID-19 pandemic on the utilization of intravitreal antivascular endothelial growth factor (anti-VEGF) pharmaceuticals in an accountable care organization (ACO).

Methods: We retrospectively analyzed the Centers for Medicare and Medicaid Services beneficiary claims for all patients in the Houston Methodist Coordinated Care ACO registry during the years 2018, 2019, and 2020.

Results: Across the 3 years studied, a mean of 708 patients received anti-VEGF injections per year.

The Association between Serum Albumin and Post-Operative Outcomes among Patients Undergoing Common Surgical Procedures: An Analysis of a Multi-Specialty Surgical Cohort from the National Surgical Quality Improvement Program (NSQIP).

While studies have demonstrated an association between preoperative hypoalbuminemia and adverse clinical outcomes, the optimal serum albumin threshold for risk-stratification in the broader surgical population remains poorly defined. We sought define the optimal threshold of preoperative serum albumin concentration for risk-stratification of adverse post-operative outcomes. Using the American College of Surgeons National Surgical Quality Improvement Program Database, we identified 842,672 patients that had undergone a common surgical procedure in one of eight surgical specialties.

Association between isolated abnormal coagulation profile on transfusion following major surgery: A NSQIP analysis of individuals without bleeding disorders.

Introduction: Preoperative coagulation screening for patients without bleeding disorders remains controversial. The combinatorial risk of INR, aPTT, and platelet count (PLT) abnormalities leading to bleeding requiring transfusion is not known in these patients. We examined the association between abnormal coagulation profile and the risk of transfusion following common elective surgery in patients without bleeding disorders.

Deep learning-based NLP data pipeline for EHR-scanned document information extraction.

Objective: Scanned documents in electronic health records (EHR) have been a challenge for decades, and are expected to stay in the foreseeable future. Current approaches for processing include image preprocessing, optical character recognition (OCR), and natural language processing (NLP). However, there is limited work evaluating the interaction of image preprocessing methods, NLP models, and document layout.

Determining Cause of Death During Coronavirus Disease 2019 Pandemic.

Unlabelled: Controversy exists whether the cause of death due to severe acute respiratory syndrome coronavirus 2 is directly related to the infection or to underlying conditions. The purpose of this study is to assess the relationship of severe acute respiratory syndrome coronavirus 2 infection with the cause of death in hospitalized patients.

Design: Retrospective observational study; deidentified discharge summaries of deceased patients were reviewed by two intensivists and classified as coronavirus disease 2019-related (caused by severe acute respiratory syndrome coronavirus 2) or coronavirus disease 2019-unrelated (not caused by severe acute respiratory syndrome coronavirus 2 or indeterminate) deaths.

Factors associated with endometrial cancer and hyperplasia among middle-aged and older Hispanics.

Objective: While disparities in endometrial hyperplasia and endometrial cancer are well documented in Blacks and Whites, limited information exists for Hispanics. The objective is to describe the patient characteristics associated with endometrial hyperplasia symptoms, endometrial hyperplasia with atypia and endometrial cancer, and assess factors contributing to racial/ethnic differences in disease outcomes.

Methods: This single-center, retrospective study included women aged ≥50 years with ≥ two encounters for endometrial hyperplasia symptoms, endometrial hyperplasia with atypia and endometrial cancer between 2012 and 2016.

Trends and Factors Associated with Nebulized Therapy Prescription in Older Adults with Chronic Obstructive Pulmonary Disease from 2008 to 2015.

Medical management of patients with chronic obstructive pulmonary disease (COPD) includes nebulized therapy as an option for inhalational drug delivery. A broad variety of short- and long-acting bronchodilators and inhaled corticosteroids in the nebulized form are available. Despite this, limited information exists on the pattern and predictors of nebulized prescription.

School-based human papillomavirus vaccination program for increasing vaccine uptake in an underserved area in Texas.

Objective: Compare the effectiveness of community-based HPV-related education and onsite school-based vaccination versus community-based education only for increasing HPV vaccine uptake in a rural, medically underserved area.

Methods: Our cohort included 2307 Rio Grande City Consolidated Independent School District (RGCISD) middle school students from 3 schools enrolled in August 2016 and followed until April 2018. Using a quasi-experimental design, this study implemented an onsite school-based vaccination program and physician-led education on HPV and HPV vaccines for parents/guardians, school nurses/staff, and pediatric/family providers in the surrounding community (15-mile radius of RGCCISD) at 1 middle school ("intervention school"), and education-only for the remaining 2 schools ("comparison schools").