A5028959249- Medical Image Segmentation Techniques
- Advanced Neural Network Applications
- Retinal Imaging and Analysis
- Image Retrieval and Classification Techniques
- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
- Face and Expression Recognition
- Trauma, Hemostasis, Coagulopathy, Resuscitation
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Research on Leishmaniasis Studies
- Hip disorders and treatments
- Medical Imaging and Analysis
- COVID-19 diagnosis using AI
- Acute Kidney Injury Research
- Advanced SAR Imaging Techniques
- Hip and Femur Fractures
- Multiple Myeloma Research and Treatments
- Geophysical Methods and Applications
- Traumatic Brain Injury and Neurovascular Disturbances
- Cardiac Valve Diseases and Treatments
Harbin Institute of Technology
2024
Shenzhen University
2023-2024
Shenzhen University Health Science Center
2022-2023
The Affiliated Yongchuan Hospital of Chongqing Medical University
2023
Chongqing Medical University
2023
The Segment Anything Model (SAM) is the first foundation model for general image segmentation. It has achieved impressive results on various natural segmentation tasks. However, medical (MIS) more challenging because of complex modalities, fine anatomical structures, uncertain and object boundaries, wide-range scales. To fully validate SAM's performance data, we collected sorted 53 open-source datasets built a large dataset with 18 84 objects, 125 object-modality paired targets, 1050K 2D...
As artificial intelligence technology advances, the application of object detection in field SAR (synthetic aperture radar) imagery is becoming increasingly widespread. However, it also faces challenges such as resource limitations spaceborne environments and significant uncertainty intensity interference scenarios. These factors make performance evaluation key to ensuring smooth execution tasks. In face complex harsh scenarios, methods that rely on single-dimensional assess models have had...
Ultrasound (US) is important for diagnosing infant developmental dysplasia of the hip (DDH). However, accuracy diagnosis depends heavily on expertise. We aimed to develop a novel automatic system (DDHnet) accurate, fast, and robust DDH.An system, DDHnet, was proposed diagnose DDH by analyzing static ultrasound images. A five-fold cross-validation experiment conducted using dataset containing 881 patients verify performance DDHnet. In addition, blind test 209 (158 normal 51 abnormal cases)....
Interactive medical image segmentation refers to the accurate of target interest through interaction (e.g., click) between user and image. It has been widely studied in recent years as it is less dependent on abundant annotated data more flexible than fully automated segmentation. However, current studies have not explored user-provided prompt information points), including knowledge mined one interaction, relationship multiple interactions. Thus, this paper, we introduce a novel framework...
Two-stage image inpainting method based on transformer and Fast Fourier convolution