A5039661451- AI in cancer detection
- Head and Neck Cancer Studies
- Oral Health Pathology and Treatment
- COVID-19 diagnosis using AI
- Biosensors and Analytical Detection
- Photodynamic Therapy Research Studies
- Radiomics and Machine Learning in Medical Imaging
- Optical Imaging and Spectroscopy Techniques
- Energetic Materials and Combustion
- Dental Research and COVID-19
- High-Velocity Impact and Material Behavior
- Electromagnetic Launch and Propulsion Technology
- Advanced Fluorescence Microscopy Techniques
- Nanoplatforms for cancer theranostics
- Image Processing Techniques and Applications
- Optical measurement and interference techniques
- Imbalanced Data Classification Techniques
- Telemedicine and Telehealth Implementation
- Innovative Microfluidic and Catalytic Techniques Innovation
- Photoacoustic and Ultrasonic Imaging
- Brain Tumor Detection and Classification
- Dental Radiography and Imaging
- Oral microbiology and periodontitis research
- Microfluidic and Capillary Electrophoresis Applications
- Advanced optical system design
Jinan University
2025
University of Arizona
2003-2024
Dongguan University of Technology
2024
State Key Laboratory of Industrial Control Technology
2011-2015
Zhejiang University
2014-2015
Zhejiang University of Technology
2011
Oral cancer, a pervasive and rapidly growing malignant disease, poses significant global health concern. Early accurate diagnosis is pivotal for improving patient outcomes. Automatic methods based on artificial intelligence have shown promising results in the oral cancer field, but accuracy still needs to be improved realistic diagnostic scenarios. Vision Transformers (ViT) outperformed learning CNN models recently many computer vision benchmark tasks. This study explores effectiveness of...
Oral cancer is a growing health issue in number of low- and middle-income countries (LMIC), particularly South Southeast Asia. The described dual-modality, dual-view, point-of-care oral screening device, developed for high-risk populations remote regions with limited infrastructure, implements autofluorescence imaging (AFI) white light (WLI) on smartphone platform, enabling early detection pre-cancerous cancerous lesions the cavity potential to reduce morbidity, mortality, overall healthcare...
With the goal to screen high-risk populations for oral cancer in low- and middle-income countries (LMICs), we have developed a low-cost, portable, easy use smartphone-based intraoral dual-modality imaging platform. In this paper present an image classification approach based on autofluorescence white light images using deep learning methods. The information from pair is extracted, calculated, fused feed neural networks. We investigated compared performance of different convolutional...
Significance: Convolutional neural networks (CNNs) show the potential for automated classification of different cancer lesions. However, their lack interpretability and explainability makes CNNs less than understandable. Furthermore, may incorrectly concentrate on other areas surrounding salient object, rather network’s attention focusing directly object to be recognized, as network has no incentive focus solely correct subjects detected. This inhibits reliability CNNs, especially biomedical...
Abstract Sensitive and specific blood-based assays for the detection of pulmonary extrapulmonary tuberculosis would reduce mortality associated with missed diagnoses, particularly in children. Here we report a nanoparticle-enhanced immunoassay read by dark-field microscopy that detects two Mycobacterium virulence factors (the glycolipid lipoarabinomannan its carrier protein) on surface circulating extracellular vesicles. In cohort study 147 hospitalized severely immunosuppressed children...
Abstract Early detection of oral cancer in low-resource settings necessitates a Point-of-Care screening tool that empowers Frontline-Health-Workers (FHW). This study was conducted to validate the accuracy Convolutional-Neural-Network (CNN) enabled m(mobile)-Health device deployed with FHWs for delineation suspicious lesions (malignant/potentially-malignant disorders). The effectiveness tested tertiary-care hospitals and India. subjects were screened independently, either by alone or along...
Significance: Oral cancer is among the most common cancers globally, especially in low- and middle-income countries. Early detection effective way to reduce mortality rate. Deep learning-based image classification models usually need be hosted on a computing server. However, internet connection unreliable for screening low-resource settings. Aim: To develop mobile-based dual-mode method customized Android application point-of-care oral detection. Approach: The dataset used our study was...
In medical imaging, deep learning-based solutions have achieved state-of-the-art performance. However, reliability restricts the integration of learning into practical workflows since conventional frameworks cannot quantitatively assess model uncertainty. this work, we propose to address shortcoming by utilizing a Bayesian network capable estimating uncertainty oral cancer image classification reliability. We evaluate using large intraoral cheek mucosa dataset captured our customized device...
Oral cancer is a growing health issue in low- and middle-income countries due to betel quid, tobacco, alcohol use younger populations of middle- high-income communities the prevalence human papillomavirus. The described point-of-care, smartphone-based intraoral probe enables autofluorescence imaging polarized white light compact geometry through USB-connected camera module. small size flexible head improves on previous designs allows cheek pockets, tonsils, base tongue, areas greatest risk...
Phase unwrapping is a very important step in fringe projection 3D imaging. In this paper, we propose new neural network for accurate phase to address the special needs Instead of labeling wrapped with integers directly, two-step training process same configuration proposed. first step, (network I) trained label only four key features phase. second another II) segments. The advantages are that dimension can be much larger from data, and serious Gaussian noise correctly unwrapped. We...
India has one of the highest rates oral squamous cell carcinoma (OSCC) in world, with an incidence 15 per 100,000 and more than 70,000 deaths year. The problem is exacerbated by a lack medical infrastructure routine screening, especially rural areas. New technologies for cancer detection timely treatment at point care are urgently needed.
Significance: Early detection of oral cancer is vital for high-risk patients, and machine learning-based automatic classification ideal disease screening. However, current datasets collected from populations are unbalanced often have detrimental effects on the performance classification. Aim: To reduce class bias caused by data imbalance. Approach: We 3851 polarized white light cheek mucosa images using our customized screening device. use weight balancing, augmentation, undersampling, focal...
Significance: The rates of melanoma and nonmelanoma skin cancer are rising across the globe. Due to a shortage board-certified dermatologists, burden dermal lesion screening erythema monitoring has fallen primary care physicians (PCPs). An adjunctive device for would be beneficial because PCPs not typically extensively trained in dermatological care. Aim: We aim examine feasibility using smartphone-camera-based dermascope USB-camera-based utilizing polarized white-light imaging (PWLI)...
Convolutional neural networks have demonstrated excellent performance in oral cancer detection and classification. However, the end-to-end learning strategy makes CNNs hard to interpret, it can be challenging fully understand decision-making procedure. Additionally, reliability is also a significant challenge for CNN based approaches. In this study, we proposed network called attention branch (ABN), which combines visual explanation mechanisms improve recognition interpret simultaneously. We...
In this Letter, a microLED-based chromatic confocal microscope with virtual slit is proposed and demonstrated for three-dimensional (3D) profiling without any mechanical scanning or external light source. the method, micro-scale light-emitting diode (microLED) panel works as point source array to achieve lateral scanning. Axial realized through aberration of an aspherical objective. A pinhole technique utilized improve contrast precision depth reconstruction. The system performance has been...