A5031131124- Cutaneous Melanoma Detection and Management
- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
- Nonmelanoma Skin Cancer Studies
- Remote-Sensing Image Classification
- Lung Cancer Diagnosis and Treatment
- Medical Image Segmentation Techniques
- Breast Cancer Treatment Studies
- COVID-19 diagnosis using AI
- Anomaly Detection Techniques and Applications
- Human Mobility and Location-Based Analysis
- Medical Imaging Techniques and Applications
- Image Processing Techniques and Applications
- Remote Sensing in Agriculture
- Remote Sensing and Land Use
- Advanced Image Processing Techniques
- Optical Coherence Tomography Applications
- Digital Media Forensic Detection
- Machine Learning in Bioinformatics
- E-Government and Public Services
- Advanced Data Compression Techniques
- Chemokine receptors and signaling
- PARP inhibition in cancer therapy
- Machine Fault Diagnosis Techniques
- Gait Recognition and Analysis
Technical University of Munich
2021-2025
Mohamed bin Zayed University of Artificial Intelligence
2024
Southeast University
2023
Hunan University of Science and Technology
2022
Chengdu University
2022
Southwest Medical University
2022
Hunan University
2019-2021
Central South University
2021
Wuhu Hit Robot Technology Research Institute
2019
Southwest Hospital
2013-2018
Abstract Background Accurate and robust pathological image analysis for colorectal cancer (CRC) diagnosis is time-consuming knowledge-intensive, but essential CRC patients’ treatment. The current heavy workload of pathologists in clinics/hospitals may easily lead to unconscious misdiagnosis based on daily analyses. Methods Based a state-of-the-art transfer-learned deep convolutional neural network artificial intelligence (AI), we proposed novel patch aggregation strategy clinic using weakly...
Precise skin lesion classification is still challenging due to two problems, i.e., (1) inter-class similarity and intra-class variation of images, (2) the weak generalization ability single Deep Convolutional Neural Network trained with limited data. Therefore, we propose a Global-Part (GP-CNN) model, which treats fine-grained local information global context equal importance. The model consists Global (G-CNN) Part (P-CNN). Specifically, G-CNN downscaled dermoscopy used extract global-scale...
Airway segmentation is crucial for the examination, diagnosis, and prognosis of lung diseases, while its manual delineation unduly burdensome. To alleviate this time-consuming potentially subjective procedure, researchers have proposed methods to automatically segment airways from computerized tomography (CT) images. However, some small-sized airway branches (e.g., bronchus terminal bronchioles) significantly aggravate difficulty automatic by machine learning models. In particular, variance...
With the advent of very-high-resolution remote sensing images, semantic change detection (SCD) based on deep learning has become a research hotspot in recent years. SCD aims to observe Earth’s land surface and plays vital role monitoring ecological environment, use cover. Existing mainly focus single-task detection; problem they face is that existing methods are incapable identifying which type occurred each multi-temporal image. In addition, few binary region help train SCD-based network....
Automated breast ultrasound image segmentation is essential in a computer-aided diagnosis (CAD) system for tumors. In this article, we present feature pyramid nonlocal network (FPNN) with transform modal ensemble learning (TMEL) accurate tumor images. Specifically, the FPNN fuses multilevel features under special consideration of long-range dependencies by combining module and network. Additionally, TMEL introduced to guide two iFPNNs extract different details. Two publicly available...
Anomaly detection (AD) in medical applications is a promising field, offering cost-effective alternative to labor-intensive abnormal data collection and labeling. However, the success of feature reconstruction-based methods AD often hindered by two critical factors: domain gap pre-trained encoders exploration decoder potential. The EA2D method we propose overcomes these challenges, paving way for more effective imaging. In this paper, present encoder-attention-2decoder (EA2D), novel tailored...
The objective of this study was to investigate the clinical characteristics and surgical modality plasma cell mastitis (PCM). A total 93 breasts 91 female patients with PCM from June 2003 2010 (unilateral in 89 bilateral two patients) were investigated study. All divided into groups: direct excision group (DE group) received focused nipple retraction correction; incision drainage (ID these procedures only event failing at least drainages. Clinical characteristics, extent excision, prognosis...
For nearly 2000 years, Eucommia ulmoides Oliver (EUO) has been utilized in traditional Chinese medicine (TCM) throughout China. Flavonoids present bark and leaves of EUO are responsible for their antioxidant, anti-inflammatory, antitumor, anti-osteoporosis, hypoglycemic, hypolipidemic, antibacterial, antiviral properties, but the main bioactive compound not established yet. In this study, we isolated identified quercetin glycoside (QAG) from (EUOL) preliminarily explored its molecular...
High-ratio image compression has always been a hotspot for remote sensing satellite transmission. Especially resource-limited environment on board, plays an important role in data storage and This article proposes novel method integrating information extraction network into comprehensive framework order to achieve high-ratio codec. To reconstruct region-of-interest (ROI) latent representations, we propose feature selection (LFS) module. Some of the channel representations are removed...
The main purpose of cloud detection is to estimate coverage and thus determine whether transmit remote sensing images earth or execute subsequent tasks based on coverage. Fast accurate estimation a necessary preprocessing step board. Therefore, we propose new approach for using regression network directly predict the A network, which termed <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math...
Tissue segmentation is the mainstay of pathological examination, whereas manual delineation unduly burdensome. To assist this time-consuming and subjective step, researchers have devised methods to automatically segment structures in images. Recently, automated machine deep learning based dominate tissue research studies. However, most approaches are supervised developed using a large number training samples, which pixel-wise annotations expensive sometimes can be impossible obtain. This...
Defocus blur, due to spatially-varying sizes and shapes, is hard remove. Existing methods either are unable effectively handle irregular defocus blur or fail generalize well on other datasets. In this work, we propose a divide-and-conquer approach tackling issue, which gives rise novel end-to-end deep learning method, called prior-and-prediction inverse kernel transformer (P2IKT), for single image deblurring. Since most can be approximated as Gaussian its variants, construct an module in our...
Artificial intelligence (AI) techniques are promising in early diagnosis of skin diseases. However, a precondition for their success is the access to large-scaled annotated data. Until now, obtaining this data has only been feasible with very high personnel and financial resources.The aim study was overcome obstacle caused by scarcity labelled data.To simulate scenario label shortage, we discarded proportion labels training set. The set consisted both unlabelled images. We then leveraged...
Urban vitality is the driving force behind sustainable urban development.As most frequently used public space in cities, enhancement of street great significance for improving human-centred habitats.Based on multi-source big data, this study uses spatial and statistical analysis methods to explore impact factors vitality.Through quantitative evaluation these factors, we propose corresponding strategies enhance street.Firstly, elements streets are extracted using deep learning algorithm based...