A5048981603- AI in cancer detection
- Image Processing Techniques and Applications
- Advanced Image Processing Techniques
- Advanced Vision and Imaging
- Biomedical Text Mining and Ontologies
- Cervical Cancer and HPV Research
- Biomedical and Engineering Education
- Multimodal Machine Learning Applications
- Topic Modeling
- Radiomics and Machine Learning in Medical Imaging
- Cancer Genomics and Diagnostics
- Image Retrieval and Classification Techniques
- Digital Holography and Microscopy
- Higher Education Learning Practices
- Medical Image Segmentation Techniques
- Gene expression and cancer classification
- Advanced Image Fusion Techniques
- Digital Image Processing Techniques
- Advances in Oncology and Radiotherapy
- Higher Education Practises and Engagement
- Virus-based gene therapy research
- Machine Learning in Healthcare
- Cell Image Analysis Techniques
Wuhan National Laboratory for Optoelectronics
2019-2023
Huazhong University of Science and Technology
2019-2023
Hong Kong University of Science and Technology
2023
University of Hong Kong
2023
Computer-assisted diagnosis is key for scaling up cervical cancer screening. However, current recognition algorithms perform poorly on whole slide image (WSI) analysis, fail to generalize diverse staining and imaging, show sub-optimal clinical-level verification. Here, we develop a progressive lesion cell method combining low- high-resolution WSIs recommend cells recurrent neural network-based WSI classification model evaluate the degree of WSIs. We train validate our analysis system 3,545...
In the cytopathology screening of cervical cancer, high-resolution digital cytopathological slides are critical for interpretation lesion cells. However, acquisition requires high-end imaging equipment and long scanning time. study, we propose a GAN-based progressive multisupervised super-resolution model called PathSRGAN (pathology GAN) to learn mapping real low-resolution images. With respect characteristics images, design new two-stage generator architecture with two supervision terms....
Whole slide image (WSI) analysis presents significant computational challenges due to the massive number of patches in gigapixel images. While transformer architectures excel at modeling long-range correlations through self-attention, their quadratic complexity makes them impractical for pathology applications. Existing solutions like local-global or linear self-attention reduce costs but compromise strong capabilities full self-attention. In this work, we propose Querent, i.e., query-aware...
Cervical cancer is a leading malignancy in female reproductive system. While AI-assisted cytology offers cost-effective and non-invasive screening solution, current systems struggle with generalizability complex clinical scenarios. To address this issue, we introduced Smart-CCS, generalizable Cancer Screening paradigm based on pretraining adaptation to create robust systems. develop validate first curated large-scale, multi-center dataset named CCS-127K, which comprises total of 127,471...
We introduce LYSTO, the Lymphocyte Assessment Hackathon, which was held in conjunction with MICCAI 2019 Conference Shenzhen (China). The competition required participants to automatically assess number of lymphocytes, particular T-cells, images colon, breast, and prostate cancer stained CD3 CD8 immunohistochemistry. Differently from other challenges setup medical image analysis, LYSTO were solely given a few hours address this problem. In paper, we describe goal multi-phase organization...
Remarkable strides in computational pathology have been made the task-agnostic foundation model that advances performance of a wide array downstream clinical tasks. Despite promising performance, there are still several challenges. First, prior works resorted to either vision-only or vision-captions data, disregarding invaluable reports and gene expression profiles which respectively offer distinct knowledge for versatile applications. Second, current progress FMs predominantly concentrates...
Histopathology serves as the gold standard in cancer diagnosis, with clinical reports being vital interpreting and understanding this process, guiding treatment patient care. The automation of histopathology report generation deep learning stands to significantly enhance efficiency lessen labor-intensive, time-consuming burden on pathologists writing. In pursuit advancement, we introduce HistGen, a multiple instance learning-empowered framework for together first benchmark dataset...
Foundation models pretrained on large-scale datasets are revolutionizing the field of computational pathology (CPath). The generalization ability foundation is crucial for success in various downstream clinical tasks. However, current have only been evaluated a limited type and number tasks, leaving their overall performance unclear. To address this gap, we established most comprehensive benchmark to evaluate off-the-shelf across six distinct task types, encompassing total 39 specific Our...
Fluorescence staining is an important technique in life science for labeling cellular constituents. However, it also suffers from being time-consuming, having difficulty simultaneous labeling, etc. Thus, virtual staining, which does not rely on chemical has been introduced. Recently, deep learning models such as transformers have applied to tasks. their performance relies large-scale pretraining, hindering development the field. To reduce reliance large amounts of computation and data, we...
Abstract Computer-assisted diagnosis is key for popularizing cervical cancer screening. However, current recognition algorithms are insufficient in accuracy and generalization lesion cells, especially when facing diversity data clinical applications. Inspired by manual reading slide under microscopes, we develop a progressive cell method combing low high resolutions WSIs to recommend cells recurrent neural network-based WSI classification model evaluate the degree of WSIs. After validating...
Biomedical microscopy images with high-resolution (HR) and axial information can help analysis diagnosis. However, obtaining such usually takes more time economic costs, which makes it impractical in most scenarios. In this paper, we first propose a novel Self-texture Transfer Super-resolution Refocusing Network (STSRNet) to reconstruct HR multi-focal plane (MFP) from single 2D low-resolution (LR) wide field image without relying on scanning or any special devices. The proposed STSRNet...
Multi-focus image fusion technologies compress different focus depth images into an in which most objects are focus. However, although existing techniques, including traditional algorithms and deep learning-based algorithms, can generate high-quality fused images, they need multiple with depths the same field of view. This criterion may not be met some cases where time efficiency is required or hardware insufficient. The problem especially prominent large-size whole slide images. paper...
Few-shot learning presents a critical solution for cancer diagnosis in computational pathology (CPath), addressing fundamental limitations data availability, particularly the scarcity of expert annotations and patient privacy constraints. A key challenge this paradigm stems from inherent disparity between limited training set whole slide images (WSIs) enormous number contained patches, where significant portion these patches lacks diagnostically relevant information, potentially diluting...
Computer-assisted cervical screening is an effective method to save the doctors’ workload and improve their work efficiency. Usually, correct classification of cells depends on nuclear segmentation effect extraction features. However, precise nucleus remains a huge challenge, especially for densely distributed nucleus. Moreover, previous cellular methods are mostly based morphological features size or color. Those individual can make accurate severe lesions, but not mild lesions. In this...
High-resolution 3D medical images are important for analysis and diagnosis, but axial scanning to acquire them is very time-consuming. In this paper, we propose a fast end-to-end multi-focal plane imaging network (MFPINet) reconstruct high-resolution from single 2D low-resolution wild filed image without relying on scanning. To realistic MFP fast, the proposed MFPINet adopts generative adversarial framework strategies of post-sampling refocusing all focal planes at one time. We conduct...
We introduce LYSTO, the Lymphocyte Assessment Hackathon, which was held in conjunction with MICCAI 2019 Conference Shenzen (China). The competition required participants to automatically assess number of lymphocytes, particular T-cells, histopathological images colon, breast, and prostate cancer stained CD3 CD8 immunohistochemistry. Differently from other challenges setup medical image analysis, LYSTO were solely given a few hours address this problem. In paper, we describe goal multi-phase...
M. Macintyre, S. Street, Y. Chen, H. Gao, Liu, R. Lu, J. Ma, Z. Meng, Pan, L. Samuel, Y-H. Tsai, Wang, Wei, W.K.E. Wong, X. Wu, Yang, B. Zhan, Zhang, ZhangUniversity of Warwick (UNITED KINGDOM)