A5043281384- Radiomics and Machine Learning in Medical Imaging
- Advanced X-ray and CT Imaging
- Lung Cancer Diagnosis and Treatment
- Medical Imaging Techniques and Applications
- AI in cancer detection
- COVID-19 diagnosis using AI
- Colorectal Cancer Screening and Detection
- MRI in cancer diagnosis
- Digital Radiography and Breast Imaging
- Radiation Dose and Imaging
- Advanced X-ray Imaging Techniques
- Advanced Memory and Neural Computing
- Advanced MEMS and NEMS Technologies
- Hepatocellular Carcinoma Treatment and Prognosis
- Laser Design and Applications
- Esophageal Cancer Research and Treatment
- Ferroelectric and Negative Capacitance Devices
- Ocular and Laser Science Research
- Gas Sensing Nanomaterials and Sensors
- Pancreatic and Hepatic Oncology Research
- Colorectal Cancer Treatments and Studies
- Laser Material Processing Techniques
- Brain Tumor Detection and Classification
- Advanced Sensor and Energy Harvesting Materials
- Acute Ischemic Stroke Management
Columbia University Irving Medical Center
2019-2024
Memorial Sloan Kettering Cancer Center
2024
Xi'an Technological University
2024
Columbia University
2020-2021
NewYork–Presbyterian Hospital
2020-2021
New York Hospital Queens
2020-2021
Shanghai Jiao Tong University
2015-2019
Lung cancer is the first killer among deaths. Malignant lung nodules have extremely high mortality while some of benign don't need any treatment .Thus, accuracy diagnosis between or malignant necessary. Notably, although currently additional invasive biopsy second CT scan in 3 months later may help radiologists to make judgments, easier approaches are imminently needed. In this paper, we propose a novel CAD method distinguish and from images directly, which can not only improve efficiency...
This study was designed to evaluate the predictive performance of 18F-fluorodeoxyglucose positron emission tomography (PET)-based radiomic features for local control esophageal cancer treated with concurrent chemoradiotherapy (CRT). For each 30 patients enrolled, 440 were extracted from both pre-CRT and mid-CRT PET images. The top 25 highest areas under receiver operating characteristic curve identifying status selected as discriminative features. Four machine-learning methods, random forest...
Lung cancer is the leading cause of deaths worldwide. Early diagnosis critical in increasing 5-year survival rate lung cancer, so efficient and accurate detection nodules, potential precursors to evermore important. In this paper, a computer-aided nodule system using convolution neural networks (CNN) handcrafted features for false positive reduction developed. The CNNs were trained with three types images: CT images, their nodule-enhanced blood vessel-enhanced images. For each candidate,...
Lung cancer is a major cause of deaths, and the 5-year survival rate stage IV lung patients only 2%. However, I significantly increases to 50%. As such, spiral computed tomography (CT) scans are necessary diagnose high-risk in early stages. In this study, computer-aided detection (CAD) system with radiomics was proposed. This could automatically detect pulmonary nodules reduce radiologists' workloads human errors.In proposed scheme, nodular enhancement filter used segment nodule candidates...
Computer-aided detection (CAD) systems can assist radiologists in reducing the interpretation time and improving results computed tomographic colonography (CTC). However, existing false positives (FPs) impair advantages of CAD systems. This study aims to develop new morphological features for FP reduction while maintaining high sensitivity. Volumetric feature maps are each polyp candidate by using three-dimensional (3-D) geodesic distance transformation, circular transformation (CcT),...
Computer-aided detection (CAD) systems for computed tomography colonography (CTC) can automatically detect colorectal polyps. The main problem of currently developed CAD-CTC is the numerous false positives (FPs) caused by existence complicated colon structures (e.g., haustral fold, residual fecal material, inflation tube, and ileocecal valve). This study proposes a scheme using shape index, multiscale enhancement filters, radiomic features to address FP issue.Shape index filter calculated in...
Abstract Background Data collected from hospitals are usually partially annotated by radiologists due to time constraints. Developing and evaluating deep learning models on these data may result in over or under estimation Purpose We aimed quantitatively investigate how the percentage of lesions CT images will influence performance universal lesion detection (ULD) algorithms. Methods trained a multi‐view feature pyramid network with position‐aware attention (MVP‐Net) perform ULD. Three...
Scatter reduces the image quality in computed tomography (CT), but scatter correction remains a challenge. A previously proposed primary modulation method simultaneously obtains and single scan. However, separating is challenging because it an underdetermined problem. In this study, optimization-based estimation (OSE) algorithm to estimate correct scatter.In concept of modulation, modulated, smooth by inserting modulator between x-ray source object. algorithm, objective function designed for...
Lung field segmentation for chest radiography is critical to pulmonary disease diagnosis. In this paper, we propose a new deformable model using weighted sparse shape composition with robust initialization achieve and accurate lung segmentation.Our method consists of three steps: initialization, deformation regularization. The steps regularization are iteratively employed until convergence. First, since sensitive the initial shape, obtained by novel voting strategy, which allows reliable...
Repeated CT scans are known to increase the risk of cancer; thus, it is paradoxical use multiple follow-up monitor development a lung nodule and conduct early treatment nodule. In case solitary nodule, regional scanning region interest (ROI) reconstruction likely restore internal area at A limited-range few-view proposed in this paper for follow-ups with extremely reduced X-radiation. For planned an ROI, where positioned, can be employed, less tissue exposed X-radiation per view. An ROI...
Tumor assessment through imaging is crucial for diagnosing and treating cancer. Lesions in the liver, a common site metastatic disease, are particularly challenging to accurately detect segment. This labor-intensive task subject individual variation, which drives interest automation using artificial intelligence (AI). Evaluate AI lesion detection segmentation CT context of human performance on same task. Use internal testing determine how an AI-developed model (ScaleNAS) trained lesions...
Contrast-enhanced computed tomography scans (CECT) are routinely used in the evaluation of different clinical scenarios, including detection and characterization hepatocellular carcinoma (HCC). Quantitative medical image analysis has been an exponentially growing scientific field. A number studies reported on effects variations contrast enhancement phase reproducibility quantitative imaging features extracted from CT scans. The identification labeling is a time-consuming task, with current...
To address the measuring requirements for ultra-large diameter and ultra-high laser energy in laser-driven inertial confinement fusion, a calorimetric method was employed measurement. NB10 neutral absorption glass used as material, temperature measurement system based on thermoelectric array designed to establish basic parts Additionally, thermal effects of pulse continuous were simulated using finite element analysis, theoretically demonstrating two forms lasers are equivalent. Through...
Abstract Background Convolutional neural networks (CNNs) have achieved great success in pulmonary nodules detection, which plays an important role lung cancer screening. Purpose In this paper, we proposed a novel strategy for nodule detection by learning it from harder task, was to transform images into normal images. We named as with image category transformation (PUNDIT). Methods There were two steps candidate and false positive (FP) reduction. step, segmentation‐based framework built...
To correct the scatter in cone beam CT, an optimization-based estimation method (OPMSE) is proposed. Like previous Fourier-based primary modulation (FPMSE), semi-transparent absorber array (SBAA) placed between X-ray source and imaging target, thus projection data behind can still be obtained, which meaningful for dose saving image quality improving. In proposed algorithm, estimated by solving optimization problem, makes use of prior information that are both smooth not only each frame, but...
We propose a one-shot thickness measurement method for sponge-like structures using propagation-based X-ray phase-contrast imaging (P-PCI) method. In P-PCI, the air-material interface refracts incident X-ray. Refracted many times along their paths by such structure, X-rays propagate randomly within small divergent angle range, resulting in speckle pattern captured image. found structure and contrast of projection are directly related images. This relationship can be described natural...