Using Large Language Models for Efficient Cancer Registry Coding in the Real Hospital Setting: A Feasibility Study.

The primary challenge in reporting cancer cases lies in the labor-intensive and time-consuming process of manually reviewing numerous reports. Current methods predominantly rely on rule-based approaches or custom-supervised learning models, which predict diagnostic codes based on a single pathology report per patient. Although these methods show promising evaluation results, their biased outcomes in controlled settings may hinder adaption to real-world reporting workflows.

Self-improving generative foundation model for synthetic medical image generation and clinical applications.

In many clinical and research settings, the scarcity of high-quality medical imaging datasets has hampered the potential of artificial intelligence (AI) clinical applications. This issue is particularly pronounced in less common conditions, underrepresented populations and emerging imaging modalities, where the availability of diverse and comprehensive datasets is often inadequate. To address this challenge, we introduce a unified medical image-text generative model called MINIM that is capable of synthesizing medical images of various organs across various imaging modalities based on textual instructions.

Changes in Staging and Management of Non-Small Cell Lung Cancer (NSCLC) Patients Following the Implementation of Low-Dose Chest Computed Tomography (LDCT) Screening at Kaohsiung Medical University Hospital.

Background: Low-dose computed tomography (LDCT) has been widely adopted for lung cancer screening due to its proven ability to reduce lung cancer mortality, especially among high-risk populations.

Methods: This retrospective study aims to evaluate the impact of LDCT screening on non-small cell lung cancer (NSCLC) staging at Kaohsiung Medical University Hospital (KMUH) from 2011 to 2020, following the introduction of LDCT in 2013. The study examines the correlation between LDCT screening volume and changes in the distribution of NSCLC stages, particularly early-stage (stages 0 and I) and late-stage (stage IV) diagnoses.

Continuous intraocular pressure variations following positional transitions in normal subjects and open angle glaucoma patients using a contact lens sensor system.

Purpose: To evaluate the efficacy of a novel contact lens sensor system (CLS) in detecting continuous intraocular pressure (IOP) variations following positional transitions, and to explore these IOP variation patterns among normal subjects, untreated glaucoma patients, and untreated ocular hypertension (OHT) subjects.

Methods: This prospective comparative study included 20 normal subjects, 14 high tension glaucoma (HTG) patients, 16 normal tension glaucoma (NTG) patients, and 14 OHT subjects. Participants wore the CLS for continuous IOP monitoring while undergoing sequential body position changes: from sitting to supine to a 10° head-down tilt (HDT), with each position maintained for ten minutes.