Quantitative ultrasound (QUS) analysis can differentiate between gingival and alveolar mucosal tissues based on distinct Burr and Nakagami model parameters, suggesting QUS's potential as an objective and quantitative diagnostic tool for periodontal disease assessment.
Three-photon excited fluorescence microscopy enables high-contrast, high-resolution imaging of blood flow, neural structure, and inflammatory response in the mouse spinal cord up to 550 μm in depth, providing a powerful tool to study diverse cellular dynamics in the spinal cord in vivo.
A deep learning-based approach to accurately estimate the femur caput-collum-diaphyseal (CCD) angle from X-ray images, which is crucial for diagnosing and managing hip problems.
The paper proposes a reconstructive source imaging procedure for diffusive ultrasound modulated bioluminescence tomography (UMBLT) in optically anisotropic media with partial data and uncertain optical parameters.
nnMamba bietet eine robuste Lösung für die Modellierung von Langstreckenabhängigkeiten in der medizinischen Bildanalyse.
Deep learning GANs enhance biological image quality across microscopy systems.
Proposing BVDM for generating synthetic live cell microscopy videos to enhance deep learning methods in the biomedical domain.
Deep learning models in image-based drug discovery may extract morphological features from single cells using Grad-CAM, but may include irrelevant pixels, leading to doubts about the fidelity of learned representations. Grad-CAMO is introduced to measure interpretability and guide model design.
Residual-based large language models (LLMs) can significantly enhance performance in biomedical imaging tasks, setting new benchmarks in 2D and 3D classification.
ULM imaging in awake mice reveals significant vascular changes from anesthesia to awake states, enabling longitudinal studies.