A5101947891- Image Enhancement Techniques
- Photoacoustic and Ultrasonic Imaging
- Image Processing Techniques and Applications
- Thermography and Photoacoustic Techniques
- Cell Image Analysis Techniques
- Radiomics and Machine Learning in Medical Imaging
- Advanced Fluorescence Microscopy Techniques
- Radiopharmaceutical Chemistry and Applications
- Mental Health Research Topics
- Image and Signal Denoising Methods
- Technology and Human Factors in Education and Health
- Esophageal Cancer Research and Treatment
- Monoclonal and Polyclonal Antibodies Research
- Advanced Neural Network Applications
- Photodynamic Therapy Research Studies
- Multimodal Machine Learning Applications
- Nanoplatforms for cancer theranostics
- Domain Adaptation and Few-Shot Learning
- Anorectal Disease Treatments and Outcomes
- Medical Image Segmentation Techniques
- Pelvic floor disorders treatments
- Medical Imaging and Analysis
- Growth Hormone and Insulin-like Growth Factors
- Enhanced Recovery After Surgery
- Advanced X-ray and CT Imaging
University of Pennsylvania
2024
Shanghai Jiao Tong University
2020-2024
University of Michigan
2024
California University of Pennsylvania
2023-2024
Sun Yat-sen University
2015
The First Affiliated Hospital, Sun Yat-sen University
2015
The point-by-point scanning mechanism of photoacoustic microscopy (PAM) results in low-speed imaging, limiting the application PAM. In this work, we propose a method to improve quality sparse PAM images using convolutional neural networks (CNNs), thereby speeding up image acquisition while maintaining good quality. CNN model utilizes attention modules, residual blocks, and perceptual losses reconstruct image, which is mapping from 1/4 or 1/16 low-sampling latent fully-sampled one. trained...
As a hybrid imaging technology, photoacoustic microscopy (PAM) suffers from noise due to the maximum permissible exposure of laser intensity, attenuation ultrasound in tissue, and inherent transducer. De-noising is an image processing method reduce noise, PAM quality can be recovered. However, previous de-noising techniques usually heavily rely on manually selected parameters, resulting unsatisfactory slow performance for different noisy images, which greatly hinders practical clinical...
Photoacoustic microscopy (PAM) is a promising imaging modality because it able to reveal optical absorption contrast in high resolution on the order of micrometer. It can be applied an endoscopic approach by implementing PAM into miniature probe, termed photoacoustic endoscopy (PAE). Here we develop focus-adjustable PAE (FA-PAE) probe characterized both (in micrometers) and large depth focus (DOF) via novel optomechanical design for adjustment. To realize DOF 2-mm plano-convex lens specially...
Image quality is degraded in the out-of-focus region because of depth-variant (DV) point spread function (DV-PSF) a fluorescence microscope. Either non-blind or blind deconvolution for restoration results limited improvement. In this work, we propose two-step learning-based DV (LB-DVD) to restore image. first step, DV-PSF predicted by defocus level prediction convolutional neural network (DelpNet). second extracted used deconvolution. To our knowledge, LB-DVD proposed and demonstrated time....
Auto-segmentation of medical images is critical to boost precision radiology and radiation oncology efficiency, thereby improving quality for both health care practitioners patients. An appropriate metric evaluate auto-segmentation results one the significant tools necessary building an effective, robust, practical technique. However, by comparing predicted segmentation with ground truth, currently widely-used metrics usually focus on overlapping area (Dice Coefficient) or most severe...
Medical image auto-segmentation techniques are basic and critical for numerous image-based analysis applications that play an important role in developing advanced personalized medicine. Compared with manual segmentations, auto-segmentations expected to contribute a more efficient clinical routine workflow by requiring fewer human interventions or revisions auto-segmentations. However, current methods usually developed the help of some popular segmentation metrics do not directly consider...
Abstract Auto-segmentation is one of the critical and foundational steps for medical image analysis. The quality auto-segmentation techniques influences efficiency precision radiology radiation oncology since high-quality auto-segmentations usually require limited manual correction. Segmentation metrics are necessary important to evaluate results guide development techniques. Currently widely applied segmentation compare with ground truth in terms overlapping area (e.g., Dice Coefficient...
Fluorescence microscopy plays an important role in biomedical research. The depth-variant point spread function (PSF) of a fluorescence microscope produces low-quality images especially the out-of-focus regions thick specimens. Traditional deconvolution to restore is usually insufficient since depth-invariant PSF assumed. This article aims at handling by learning-based and reducing artifacts. We propose adaptive weighting (AWDVD) with defocus level prediction convolutional neural network...