A5049790626- AI in cancer detection
- Medical Image Segmentation Techniques
- Radiomics and Machine Learning in Medical Imaging
- Video Surveillance and Tracking Methods
- Human Pose and Action Recognition
- Brain Tumor Detection and Classification
- Face and Expression Recognition
- Lung Cancer Diagnosis and Treatment
- Medical Imaging Techniques and Applications
- Advanced Neural Network Applications
- Prostate Cancer Diagnosis and Treatment
- Gait Recognition and Analysis
- Alzheimer's disease research and treatments
- Advanced X-ray and CT Imaging
- Image Processing Techniques and Applications
- Medical Imaging and Analysis
- Neural Networks and Applications
- Robotics and Sensor-Based Localization
- Advanced Image and Video Retrieval Techniques
- Advanced Neuroimaging Techniques and Applications
- Anomaly Detection Techniques and Applications
- Sparse and Compressive Sensing Techniques
- Face recognition and analysis
- Morphological variations and asymmetry
- Neuroinflammation and Neurodegeneration Mechanisms
Qilu Hospital of Shandong University
2021-2023
Advanced Imaging Research (United States)
2017-2020
Duke University
2016-2020
Siemens (United States)
2015-2020
Siemens Healthcare (United States)
2019-2020
Arizona State University
2018
Duke University Hospital
2016-2017
Duke Medical Center
2016-2017
Ohio University
2011-2016
Alzheimer’s Disease Neuroimaging Initiative
2014
<h3>Importance</h3> Mammography screening currently relies on subjective human interpretation. Artificial intelligence (AI) advances could be used to increase mammography accuracy by reducing missed cancers and false positives. <h3>Objective</h3> To evaluate whether AI can overcome interpretation limitations with a rigorous, unbiased evaluation of machine learning algorithms. <h3>Design, Setting, Participants</h3> In this diagnostic study conducted between September 2016 November 2017, an...
Detecting an anomaly such as a malignant tumor or nodule from medical images including mammogram, CT PET is still ongoing research problem drawing lot of attention with applications in diagnosis. A conventional way to address this learn discriminative model using training datasets negative and positive samples. The learned can be used classify testing sample into class. However, applications, the high unbalance between samples poses difficulty for learning algorithms, they will biased...
Automated segmentation of brain structures from MR images is an important practice in many neuroimage studies. In this paper, we explore the utilization a multi-view ensemble approach that relies on neural networks (NN) to combine multiple decision maps achieving accurate hippocampus segmentation. Constructed under general convolutional NN structure, our Ensemble-Net different convolution configurations capture complementary information residing label probabilities produced by U-Seg-Net (a...
Opportunistic prostate cancer (PCa) screening is a controversial topic. Magnetic resonance imaging (MRI) has proven to detect with high sensitivity and specificity, leading the idea perform an image-guided screening; Methods: We evaluated prospectively enrolled cohort of 49 healthy men participating in dedicated PCa trial employing biparametric MRI (bpMRI) protocol consisting T2-weighted (T2w) diffusion weighted (DWI) sequences. Datasets were analyzed both by human readers fully automated...
Prostate cancer (PCa) is the most prevalent and one of leading causes death among men. Multi-parametric MRI (mp-MRI) a prominent diagnostic scan, which could help in avoiding unnecessary biopsies for men screened PCa. Artificial intelligence (AI) systems radiologists to be more accurate consistent diagnosing clinically significant from mp-MRI scans. Lack specificity has been identified recently as weak points such assistance systems. In this paper, we propose novel false positive reduction...
To investigate if supervised machine learning (ML) classifiers would be able to predict clinically significant cancer (sPC) from a set of quantitative image-features and compare these results with established PI-RADS v2 assessment scores.We retrospectively included 201, histopathologically-proven, peripheral zone (PZ) prostate lesions. Gleason scores ≤3+3 were considered as insignificant (inPC) ≥3+4 sPC encoded in binary fashion, serving ground-truth. MRI was performed at 3T high...
This study aimed to investigate the effects of thrombin released in hematoma after intracerebral hemorrhage (ICH) on cerebral injury perihematomal tissues and evaluate protection effect hirudin ICH.We used autologous uncoagulated blood injection method prepare ICH rat model, all rats were randomly divided into a normal group, an or group.At different time points, heads cut harvest brain sections.Immunohistochemical staining, histochemical hematoxylin eosin staining conducted for CD34,...
Detailed analysis of brain structures is essential in identifying anatomical biomarkers Alzheimer's disease (AD). In this paper, we develop a new radial distance model to compare different hippocampal shapes and measure their atrophies over time. Using harmonic mappings, project surfaces onto cylinders obtain evenly-spaced quadrilateral meshes. Surface distances estimated via the quad-meshes are invariant global shifts surrounding tissues, leading powerful way detect localized progressions....
Purpose: To determine whether domain transfer learning can improve the performance of deep features extracted from digital mammograms using a pre-trained convolutional neural network (CNN) in prediction occult invasive disease for patients with ductal carcinoma situ (DCIS) on core needle biopsy. Method: In this study, we collected mammography magnification views 140 DCIS at biopsy, 35 which were subsequently upstaged to cancer. We utilized CNN model that was two natural image data sets...
Identifying intermediate biomarkers of Alzheimer's disease (AD) is great importance for diagnosis and prognosis the disease. In this study, we develop a new AD staging method to classify patients into Normal Controls (NC), Mild Cognitive Impairment (MCI), groups. Our solution employs novel metric learning technique that improves classification rates through guidance some weak supervisory information in progression. More specifically, those are form pairwise constraints specify relative Mini...
In this paper, we propose a nonlinear metric learning framework to boost the performance of semi-supervised (SSL) algorithms. Constructed on top Laplacian SVM (LapSVM), proposed method learns smooth feature space transformation that makes input data points more linearly separable. Coherent point drifting (CPD) is utilized as geometric model with consideration its remarkable expressive power in generating sophisticated yet deformations. Our has broad applicability, and it can be integrated...
Predicting the risk of occult invasive disease in ductal carcinoma <i>in situ </i>(DCIS) is an important task to help address overdiagnosis and overtreatment problems associated with breast cancer. In this work, we investigated feasibility using computer-extracted mammographic features predict patients biopsy proven DCIS. We proposed a computer-vision algorithm based approach extract from magnification views full field digital mammography (FFDM) for After expert radiologist provided region...
In this work we revisit TV filter and propose an improved version that is tailored to diagnostic CT purposes. We revise cost function, which results in symmetric gradient function leads more natural noise texture. apply a multi-scale approach resolve grain issue images. examine texture, granularity, loss of low contrast the test also discuss potential acceleration by Nesterov Conjugate Gradient methods.
Motion estimation is a very important method for improving image quality by compensating the cardiac motion at best phase reconstructed. We tackle problem using an registration approach. compare performance of three gradient-based methods on clinical data. In addition to simple gradient descent, we test Nesterov accelerated descent and conjugate algorithms. The results show that provide significant speedup over conventional with no loss quality.
A new motion estimation and compensation method for cardiac computed tomography (CT) was developed. By combining two (ME) approaches the proposed estimates local global then preforms compensated reconstruction. The combined has parts: one is estimation, which coronary artery by using tree tracking registration; other entire image registration. final linear combination of the. We use backproject-then-warp Pack et al. to perform reconstruction (MCR). evaluated with 5 patient data improvements...