A5031013744- AI in cancer detection
- Advanced Graph Neural Networks
- Radiomics and Machine Learning in Medical Imaging
- Tensor decomposition and applications
- Thermal Regulation in Medicine
- Advanced Image Fusion Techniques
- Bioinformatics and Genomic Networks
- Topic Modeling
- Machine Learning in Healthcare
- Phonocardiography and Auscultation Techniques
- Biomedical Text Mining and Ontologies
- Ultrasound Imaging and Elastography
- Ultrasound in Clinical Applications
- Multimodal Machine Learning Applications
Northwestern Polytechnical University
2020-2022
Segmentation of the breast ultrasound (BUS) image is an important step for subsequent assessment and diagnosis lesions. Recently, Deep-learning-based methods have achieved satisfactory performance in many computer vision tasks, especially medical segmentation. Nevertheless, those always require a large number pixel-wise labeled data that expensive practices. In this study, we propose new segmentation method by dense prediction local fusion superpixels anatomy with scarce data. First,...
The advantage of Knowledge Graph (KG) can greatly prompt the interpretability artificial intelligence diagnosis. For breast ultrasound, KG be built through BI-RADS semantic descriptions, and diagnosis achieved by link reconstruction between patients outcomes. However, existing analysis methods consider only linked neighbors entities relations during embedding, but not whole in KG, which reduces power for case a small fraction labeled patients. In this paper, we present transductive learning...
Knowledge graph has drawn increasingly attention in medical artificial intelligence recent years. In the task of knowledge inference, most existing approaches are focusing on single disciplinary, leaving idea multidisciplinary behind. consideration distinct advantages clinical decision, simulating by alignment can also prompt inference. To introduce into we design a preliminary pipeline called compensating generalized which consists and embedding based link prediction. comparison study,...