A5102839444- Generative Adversarial Networks and Image Synthesis
- Cell Image Analysis Techniques
- AI in cancer detection
- EEG and Brain-Computer Interfaces
- Advanced Vision and Imaging
- Medical Image Segmentation Techniques
- Neuroscience and Neural Engineering
- Tactile and Sensory Interactions
RWTH Aachen University
2024
Recent advances in computer vision have led to significant progress the generation of realistic image data, with denoising diffusion probabilistic models proving be a particularly effective method. In this study, we demonstrate that can effectively generate fully-annotated microscopy data sets through an unsupervised and intuitive approach, using rough sketches desired structures as starting point. The proposed pipeline helps reduce reliance on manual annotations when training deep...
The segmentation and tracking of living cells play a vital role within the biomedical domain, particularly in cancer research, drug development, developmental biology. These are usually tedious time-consuming tasks that traditionally done by experts. Recently, to automatize these processes, deep learning based methods have been proposed. require large-scale datasets their full potential is constrained scarcity annotated data imaging domain. To address this limitation, we propose Biomedical...
This study proposes a retinal prosthetic simulation framework driven by visual fixations, inspired the saccade mechanism, and assesses performance improvements through end-to-end optimization in classification task. Salient patches are predicted from input images using self-attention map of vision transformer to mimic fixations. These then encoded trainable U-Net simulated pulse2percept predict percepts. By incorporating learnable encoder, we aim optimize information transmitted implant,...
Automated cell segmentation in microscopy images is essential for biomedical research, yet conventional methods are labor-intensive and prone to error. While deep learning-based approaches have proven effective, they often require large annotated datasets, which scarce due the challenges of manual annotation. To overcome this, we propose a novel framework synthesizing densely 2D 3D using cascaded diffusion models. Our method synthesizes masks from sparse annotations multi-level models NeuS,...