Generative AI for Bone Scintigraphy Image Synthesis and Enhanced Deep Learning Model Generalization in Data-Constrained Settings

Information Science, Artificial Intelligence, Medicine, Medical Imaging,
generative AI   bone scintigraphy   deep learning   data augmentation   multicenter   molecular imaging   medical imaging  

Breakthrough Applications of Generative Artificial Intelligence in Nuclear Medicine: Exploring the Potential of Synthetic Bone Scintigraphy Images and Their Application in Deep Learning Background and Research Questions In recent years, the rapid development of Artificial Intelligence (AI) has revolutionized medical imaging analysis. For instance, ...

RADIFF: Controllable Diffusion Models for Radio Astronomical Maps Generation

Physics, Astrophysics, Information Science, Computer Science, Artificial Intelligence,
diffusion models   radio astronomy   data augmentation   generative models   semantic segmentation  

RaDiff: Controllable Diffusion Models for Radio Astronomical Map Generation” Comprehensive Academic News Analysis Background Introduction With the near completion of the Square Kilometer Array (SKA) telescope, radio astronomy is poised for revolutionary advancements in the study of the universe. Boasting unprecedented sensitivity and spatial resolu...

EfficientDeRain+: Learning Uncertainty-Aware Filtering via RainMix Augmentation for High-Efficiency Deraining

Information Science, Electronic Information, Computer Science, Automation, Artificial Intelligence,
deraining   predictive filtering   cascaded dilated filtering   multi-scale filtering   data augmentation   RainMix  

EfficientDeRain+: A High-Efficiency Image Deraining Method Enhanced by RainMix Augmentation Background Rain significantly affects the quality of images and videos captured by computer vision systems, with raindrops and streaks impairing clarity and degrading performance in tasks like pedestrian detection, object tracking, and semantic segmentation....

Mitigating Social Biases of Pre-trained Language Models via Contrastive Self-Debiasing with Double Data Augmentation

Information Science, Computer Science, Artificial Intelligence,
pre-trained language models   social bias   contrastive learning   prompt learning   data augmentation  

Introduction: Currently, pre-trained language models (PLMs) are widely applied in the field of natural language processing, but they have the problem of inheriting and amplifying social biases present in the training corpora. Social biases may lead to unpredictable risks in real-world applications of PLMs, such as automatic job screening systems te...