Publications by authors named "O Gevaert"
We have developed the regionalpcs method, an approach for summarizing gene-level methylation. regionalpcs addresses the challenge of deciphering complex epigenetic mechanisms in diseases like Alzheimer's disease. In contrast to averaging, regionalpcs uses principal components analysis to capture complex methylation patterns across gene regions.
View Article and Find Full Text PDFBackground: Non-small-cell lung cancer (NSCLC) remains a global health challenge, driving morbidity and mortality. The emerging field of radiogenomics utilizes statistical methods to correlate radiographic tumor features with genomic characteristics from biopsy samples. Radiomic techniques automate the precise extraction of imaging features from tumor regions in radiographic scans, which are subjected to machine learning (ML) to predict genomic attributes.
View Article and Find Full Text PDFArticle Synopsis
- Cancer is complex and expensive to analyze, often requiring genetic profiling for better treatment strategies.
- Recent advancements in deep learning have improved the prediction of genetic changes from whole slide images, but transformers have been challenging to implement due to their complexity and data limitations.
- The new SEQUOIA model, which uses a simplified transformer approach, successfully predicts cancer-related gene information from vast numbers of tumor samples and has shown promise in aiding breast cancer risk assessment and understanding gene expression in specific regions.
Article Synopsis
- Researchers developed a tri-light warning system to classify COVID-19 patients based on their risk levels (high, uncertain, low) using data from 3,646 patients in China.
- The system analyzes CT scans and clinical records, successfully predicting ICU admissions with strong performance metrics (AUROC and AUPRC) even with data variations.
- This tool aims to optimize medical resources and improve responses to both original and emerging COVID-19 variants.