Segmentation Accuracy
Segmentation accuracy, the precision of delineating objects or regions within images, is a crucial aspect of many fields, particularly medical image analysis and remote sensing. Current research focuses on improving accuracy through advanced model architectures like U-Net and its variants, Transformers, and novel loss functions designed to address challenges such as class imbalance and small object detection. These advancements are driving improvements in diagnostic accuracy, treatment planning, and automated analysis across diverse applications, impacting both scientific understanding and practical outcomes.
Papers
Localise to segment: crop to improve organ at risk segmentation accuracy
Abraham George Smith, Denis Kutnár, Ivan Richter Vogelius, Sune Darkner, Jens Petersen
Self-training with dual uncertainty for semi-supervised medical image segmentation
Zhanhong Qiu, Haitao Gan, Ming Shi, Zhongwei Huang, Zhi Yang