Publications by authors named "G Baumgartner"
Analyzing the TotalSegmentator for facial feature removal in head CT scans.
Radiography (Lond)
January 2025
Background: Facial recognition technology in medical imaging, particularly with head scans, poses privacy risks due to identifiable facial features. This study evaluates the use of facial recognition software in identifying facial features from head CT scans and explores a defacing pipeline using TotalSegmentator to reduce re-identification risks while preserving data integrity for research.
Methods: 1404 high-quality renderings from the UCLH EIT Stroke dataset, both with and without defacing were analysed.
Background: The Prostate Imaging-Reporting and Data System (PI-RADS) calls for reporting the prostate index lesion and the location within the transition (TZ) or peripheral zone (PZ) and location on a corresponding sector map. The aim of this study was to train a deep learning DL-based algorithm for automatic prostate sector mapping and to validate its' performance.
Methods: An automatic 24-sector grid-map (ASG) of the prostate was developed, based on an automatic zone-specific deep learning segmentation of the prostate.
While gilts and sows are regularly vaccinated against the porcine parvovirus (PPV), little is known on the presence of antibodies in vaccinated sows nor the decline of maternally derived antibodies (MDA) in their offspring. On twelve farms serum samples were taken from 180 gilts and sows vaccinated at least twice with one of three different commercial PPV vaccines. On nine farms, additional 270 serum samples were collected from growing pigs of three different age categories.
View Article and Find Full Text PDFArticle Synopsis
- The study aimed to improve prostate cancer biopsy decision strategies by combining zone-specific prostate-specific antigen density (sPSAD) with the PI-RADS system.
- Using a deep learning system, researchers segmented MRI images to calculate sPSAD for the whole gland and transition zone, yielding better detection rates for significant prostate cancer compared to traditional methods.
- The findings indicated that adopting sPSAD reduced false positives and enhanced specificity in biopsy decisions, particularly in cases classified as PI-RADS 3, while still maintaining high sensitivity for detecting more serious cancer cases.
Article Synopsis
- The study addresses challenges in organizing and retrieving large volumes of medical imaging data, focusing on developing a CNN that classifies images based on voxel data without relying on traditional naming and metadata.
- A 3D CNN was trained on over 31,000 prostate MRI volumes using patient-based cross-validation, demonstrating high accuracy (99.88%) even when trained with a reduced dataset (10%).
- The CNN's capabilities in automatic sequence identification of prostate MRIs could improve the management of medical imaging data and support more efficient clinical workflows using AI.