A5081179416- Radiomics and Machine Learning in Medical Imaging
- COVID-19 diagnosis using AI
- AI in cancer detection
- Lung Cancer Diagnosis and Treatment
- Retinal Imaging and Analysis
- Advanced Neural Network Applications
- Medical Imaging and Analysis
- Medical Image Segmentation Techniques
- Human Pose and Action Recognition
- Glaucoma and retinal disorders
- Brain Tumor Detection and Classification
- Advanced biosensing and bioanalysis techniques
- Computational Drug Discovery Methods
- EEG and Brain-Computer Interfaces
- Image Retrieval and Classification Techniques
- Medical Imaging Techniques and Applications
- Colorectal Cancer Screening and Detection
- Retinal Diseases and Treatments
- Gastric Cancer Management and Outcomes
- Advanced Vision and Imaging
- Dental Radiography and Imaging
- ECG Monitoring and Analysis
- Digital Imaging for Blood Diseases
- Acute Ischemic Stroke Management
- Video Surveillance and Tracking Methods
Shanghai Jiao Tong University
2017-2025
Ruijin Hospital
2020-2021
Shanghai Academy of Environmental Sciences
2020
Deepblue Technology (China)
2020
Zhejiang University
2014-2017
Hospital Universitario Quirónsalud Madrid
2016
Ospedale Policlinico San Martino
2016
We propose a conceptually simple framework for fast COVID-19 screening in 3D chest CT images. The can efficiently predict whether or not scan contains pneumonia while simultaneously identifying types between and Interstitial Lung Disease (ILD) caused by other viruses. In the proposed method, two 3D-ResNets are coupled together into single model above-mentioned tasks via novel prior-attention strategy. extend residual learning with mechanism design new so-called (PARL) block. be easily built...
Many common eye diseases and cardiovascular can be diagnosed through retinal imaging. However, due to uneven illumination, image blurring, low contrast, images with poor quality are not useful for diagnosis, especially in automated analyzing systems. Here, we propose a new enhancement method improve color luminosity contrast.A luminance gain matrix, which is obtained by gamma correction of the value channel HSV (hue, saturation, value) space, used enhance R, G, B (red, green blue) channels,...
Telemedicine and the medical "big data" era in ophthalmology highlight use of non-mydriatic ocular fundus photography, which has given rise to indispensable applications portable cameras. However, case image quality is more vulnerable distortions, such as uneven illumination, color distortion, blur, low contrast. Such distortions are called generic distortions. This paper proposes an algorithm capable selecting images fair that would be especially useful assist inexperienced individuals...
Identifying drug–protein interactions (DPIs) is crucial in drug discovery, and a number of machine learning methods have been developed to predict DPIs. Existing usually use unrealistic data sets with hidden bias, which will limit the accuracy virtual screening methods. Meanwhile, most DPI prediction pay more attention molecular representation but lack effective research on protein high-level associations between different instances. To this end, we present novel structure-aware multimodal...
We modified and reconstructed a high image quality portable non-mydriatic fundus camera compared it with the tabletop to evaluate efficacy of new in detecting retinal diseases. designed built novel handheld telemedicine system. The cameras was that existing commercial by taking photographs 364 eyes from 254 patients. In all 800 images taken two types, 400 per camera, were graded four clarity classifications. Using 63% (252/400) as excellent overall quality, 20.5% (82/400) good, 11.75%...
Abstract A simple, convenient, and highly sensitive bio-interface for graphene field-effect transistors (GFETs) based on multifunctional nano-denatured bovine serum albumin (nano-dBSA) functionalization was developed to target cancer biomarkers. The novel graphene–protein bioelectronic interface constructed by heating denature native BSA the substrate surface. formed nano-dBSA film served as cross-linker immobilize monoclonal antibody against carcinoembryonic antigen (anti-CEA mAb) channel...
Introduction: The loss of nigrosome-1 which is also referred to as the swallow tail sign (STS) in T2* weighted iron-sensitive magnetic resonance imaging (MRI) has recently emerged a new biomarker for idiopathic Parkinson's disease (IPD). However, consistent recognition STS difficult due individual variations and different parameters. Radiomics might have potential overcome these shortcomings. Therefore, we chose explore whether radiomic features substantia nigra (SN) based on quantitative...
In computed tomography, automated detection of pulmonary nodules with a broad spectrum appearance is still challenge, especially, in the small nodules. An system usually contains two major steps: candidate and false positive (FP) reduction. We propose novel strategy for fast from volumetric chest CT scans, which can minimize negatives (FNs) positives (FPs). The core nodule-size-adaptive deep model that detect various types, locations, sizes 3D images. After detection, each result located...
Abstract Flexible, material‐based, artificial muscles enable compliant and safe technologies for human–machine interaction devices adaptive soft robots, yet there remain long‐term challenges in the development of capable mimicking flexible, controllable, multifunctional human activity. Inspired by limb's activity strategy, combining muscles' adjustable stiffness joints' origami folding, controllable “skeletons,” which are created laminar jamming folding multiple layers flexible sandpaper,...
Abstract Background One of the major challenges in precision medicine is accurate prediction individual patient’s response to drugs. A great number computational methods have been developed predict compounds activity using genomic profiles or chemical structures, but more exploration yet be done combine genetic mutation, gene expression, and cheminformatics one machine learning model. Results We presented here a novel deep-learning model that integrates structure multi-task convolutional...
Person re-identification (ReID) plays an important role in applications such as public security and video surveillance. Recently, learning from synthetic data [ 9 ], which benefits the popularity of engine, has attracted great attention public. However, existing datasets are limited quantity, diversity, realisticity, cannot be efficiently used for ReID problem. To address this challenge, we manually construct a large-scale person dataset named FineGPR with fine-grained attribute annotations....
Objective: To assess the performance of pretreatment 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) radiomics features for predicting EGFR mutation status in patients with non-small cell lung cancer (NSCLC). Patients and Methods: We enrolled total 173 histologically proven NSCLC who underwent preoperative 18F-FDG PET/CT. Tumor tissues all were tested status. A PET/CT prediction model was established through multi-step feature selection. The...
Pneumothorax is a life-threatening emergency that requires immediate treatment. Frontal-view chest X-ray images are typically used for pneumothorax detection in clinical practice. However, manual review of radiographs time-consuming, labor-intensive, and highly dependent on the experience radiologists, which may lead to misdiagnosis. Here, we aim develop reliable automatic classification method assist radiologists rapidly accurately diagnosing frontal radiographs.A novel residual neural...
A smart temperature sensor has been realized using a Taiwan Semiconductor Manufacturing Company (TSMC) 0.18-μm standard CMOS process. Substrate positive-negative- positive (PNP) transistors are used to extract the information. Using techniques of dynamic element matching, single-transistor approach, and offset cancellation, process spread between batches on becomes main inaccuracy sources sensor. The uses single-point calibration method that reduces simplifies trimming. can realize an (±3δ)...
To develop a deep-learning-based automatic diagnosis system for identifying nasopharyngeal carcinoma (NPC) from noncancer (inflammation and hyperplasia), using both white light imaging (WLI) narrow-band (NBI) nasopharyngoscopy images.Retrospective study.A total of 4,783 images (2,898 WLI 1,885 NBI) 671 patients were collected novel deep convolutional neural network (DCNN) framework was developed named Siamese (S-DCNN), which can simultaneously utilize NBI to improve the classification...
Karyotyping is the gold standard in detection of chromosomal abnormalities. To facilitate diagnostic process, this paper, a method for chromosome classification and straightening based on an interleaved multi-task network proposed. This consists three stages. In first stage, multi-scale features are learned via network. second high-resolution from stage input to convolution neural subnetwork joint detection, other fused fed two multi-layer perceptron subnetworks type polarity classification....
Accurate segmentation of craniomaxillofacial (CMF) structures and individual teeth is essential for advancing computer-assisted CMF surgery. This study developed CMF-ELSeg, a novel fully automatic multi-structure model based on deep ensemble learning. A total 143 computed tomography (CT) scans were retrospectively collected manually annotated by experts training validation. Three 3D U-Net-based learning models (V-Net, nnU-Net, UX-Net) benchmarked. CMF-ELSeg employed coarse-to-fine cascaded...
Purpose: To propose a novel approach for automated measurements of margin reflex distance (MRD) using digital image analysis and to evaluate the agreement between manual MRD.Materials Methods: This observational study included 132 eyes 66 volunteers referred an oculoplastic clinic blepharoptosis. Preoperative facial photographs participants were taken. MRD1 MRD2 measured manually by senior surgeon automatically our image-based algorithm. Correlation analyses Bland–Altman performed two...
Purpose In clinical practice, invasiveness is an important reference indicator for differentiating the malignant degree of subsolid pulmonary nodules. These nodules can be classified as atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS), minimally invasive (MIA), or (IAC). The automatic determination a nodule's based on chest CT scans guide treatment planning. However, it challenging, owing to insufficiency training data and their interclass similarity intraclass variation....