Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.2214/AJR.23.30573 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Article Synopsis
- Deep learning methods show strong potential for predicting lung cancer risk from CT scans, but there's a need for more comprehensive comparisons and validations of these models in real-world settings.
- The study reviews 21 state-of-the-art deep learning models, analyzing their performance using CT scans from a subset of the National Lung Screening Trial, with a focus on malignant versus benign classification.
- Results reveal that 3D deep learning models generally outperformed 2D models, with the best 3D model achieving an AUROC of 0.86 compared to 0.79 for the best 2D model, emphasizing the need to choose appropriate pretrained datasets and model types for effective lung cancer risk prediction.
Learning the language of antibody hypervariability.
Proc Natl Acad Sci U S A
January 2025
Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, MA 02139.
Protein language models (PLMs) have demonstrated impressive success in modeling proteins. However, general-purpose "foundational" PLMs have limited performance in modeling antibodies due to the latter's hypervariable regions, which do not conform to the evolutionary conservation principles that such models rely on. In this study, we propose a transfer learning framework called Antibody Mutagenesis-Augmented Processing (AbMAP), which fine-tunes foundational models for antibody-sequence inputs by supervising on antibody structure and binding specificity examples.
View Article and Find Full Text PDFThe advantages of lexicon-based sentiment analysis in an age of machine learning.
PLoS One
January 2025
Department of Political Science, Middlebury College, Middlebury, Vermont, United States of America.
Assessing whether texts are positive or negative-sentiment analysis-has wide-ranging applications across many disciplines. Automated approaches make it possible to code near unlimited quantities of texts rapidly, replicably, and with high accuracy. Compared to machine learning and large language model (LLM) approaches, lexicon-based methods may sacrifice some in performance, but in exchange they provide generalizability and domain independence, while crucially offering the possibility of identifying gradations in sentiment.
View Article and Find Full Text PDFA dataset and benchmark for hospital course summarization with adapted large language models.
J Am Med Inform Assoc
December 2024
Department of Radiology, Stanford University, Stanford, CA 94304, United States.
Objective: Brief hospital course (BHC) summaries are clinical documents that summarize a patient's hospital stay. While large language models (LLMs) depict remarkable capabilities in automating real-world tasks, their capabilities for healthcare applications such as synthesizing BHCs from clinical notes have not been shown. We introduce a novel preprocessed dataset, the MIMIC-IV-BHC, encapsulating clinical note and BHC pairs to adapt LLMs for BHC synthesis.
View Article and Find Full Text PDFPublic Health Discussions on Social Media: Evaluating Automated Sentiment Analysis Methods.
JMIR Form Res
January 2025
Department of Health Administration, The College of Health Professions, Central Michigan University, Mt Pleasant, MI, United States.
Article Synopsis
- Sentiment analysis is a key method for analyzing text, especially in social media research, where the choice between manual and automated methods is crucial.
- The study compared several sentiment analysis tools, including VADER, TEXT2DATA, LIWC-22, and ChatGPT 4.0, against manually coded sentiment scores from YouTube comments on the opioid crisis, assessing factors like ease of use and cost.
- Findings revealed that LIWC-22 excelled in identifying sentiment patterns, while VADER was best at classifying negative comments, but overall, automated tools showed only fair agreement with manual coding, with ChatGPT performing poorly.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!