Publications by authors named "Kevin Bronik"
Article Synopsis
- * This study introduces a novel approach using two coupled convolutional neural networks (CNNs) to learn the reliability of individual annotators and the consensus label from noisy observations, aiming to differentiate between unreliable behaviors of annotators and accurate data representations.
- * The proposed method was validated on various datasets, including simulated and real-world medical images, consistently outperforming other strategies, especially when faced with fewer annotations and greater disagreements among annotators.
This paper presents a framework for automatically learning shape and appearance models for medical (and certain other) images. The algorithm was developed with the aim of eventually enabling distributed privacy-preserving analysis of brain image data, such that shared information (shape and appearance basis functions) may be passed across sites, whereas latent variables that encode individual images remain secure within each site. These latent variables are proposed as features for privacy-preserving data mining applications.
View Article and Find Full Text PDF