6 papers · 1 filter
VIVID-Med: LLM-Supervised Structured Pretraining for Deployable Medical ViTs
Xiyao Wang, Xiaoyu Tan, Yang Dai +3
Vision-language pretraining has driven significant progress in medical image analysis. However, current methods typically supervise visual encoders using one-hot labels or free-for…
Beyond Benchmarks of IUGC: Rethinking Requirements of Deep Learning Methods for Intrapartum Ultrasound Biometry from Fetal Ultrasound Videos
Jieyun Bai, Zihao Zhou, Yitong Tang +60
A substantial proportion (45\%) of maternal deaths, neonatal deaths, and stillbirths occur during the intrapartum phase, with a particularly high burden in low- and middle-income c…
Vector Contrastive Learning For Pixel-Wise Pretraining In Medical Vision
Yuting He, Shuo Li
Contrastive learning (CL) has become a cornerstone of self-supervised pretraining (SSP) in foundation models, however, extending CL to pixel-wise representation, crucial for medica…
Gaze-Assisted Human-Centric Domain Adaptation for Cardiac Ultrasound Image Segmentation
Ruiyi Li, Yuting He, Rongjun Ge +4
Domain adaptation (DA) for cardiac ultrasound image segmentation is clinically significant and valuable. However, previous domain adaptation methods are prone to be affected by the…
DuSSS: Dual Semantic Similarity-Supervised Vision-Language Model for Semi-Supervised Medical Image Segmentation
Qingtao Pan, Wenhao Qiao, Jingjiao Lou +2
Semi-supervised medical image segmentation (SSMIS) uses consistency learning to regularize model training, which alleviates the burden of pixel-wise manual annotations. However, it…
TASL-Net: Tri-Attention Selective Learning Network for Intelligent Diagnosis of Bimodal Ultrasound Video
Chengqian Zhao, Zhao Yao, Zhaoyu Hu +8
In the intelligent diagnosis of bimodal (gray-scale and contrast-enhanced) ultrasound videos, medical domain knowledge such as the way sonographers browse videos, the particular ar…