A5043696543- Advanced MRI Techniques and Applications
- MRI in cancer diagnosis
- Advanced Neuroimaging Techniques and Applications
- Radiomics and Machine Learning in Medical Imaging
- Advanced Radiotherapy Techniques
- Medical Imaging and Analysis
- Photoacoustic and Ultrasonic Imaging
- Image and Signal Denoising Methods
- Advanced Data Compression Techniques
- Ultrasound and Hyperthermia Applications
- Magnetic Properties and Applications
- Ultrasound Imaging and Elastography
- Thermal Regulation in Medicine
- Infrared Thermography in Medicine
- Hepatocellular Carcinoma Treatment and Prognosis
- Historical and Literary Analyses
- Advanced Image Processing Techniques
- Medical Imaging Techniques and Applications
- Shakespeare, Adaptation, and Literary Criticism
University of Wisconsin–Madison
2021-2024
Google (United States)
2023
To assess the effects of cardiovascular-induced motion on conventional DWI pancreas and to evaluate motion-robust methods in a phantom healthy volunteers.3T was acquired using standard monopolar motion-compensated gradient waveforms, including an anatomically accurate with controllable compressive volunteers (n = 8, 10). In volunteers, highly controlled single-slice breath-holding cardiac gating whole-pancreas respiratory-triggered were acquired. For each acquisition, ADC variability across...
To design a b value and first-order motion moment (M
Purpose To evaluate feasibility and reproducibility of liver diffusion‐weighted (DW) MRI using cardiac‐motion‐robust, blood‐suppressed, reduced‐distortion techniques. Methods DW‐MRI data were acquired at 3T in an anatomically accurate phantom including controlled pulsatile motion, eight healthy volunteers four patients with known or suspected metastases. Standard monopolar motion‐robust (M1‐nulled, M1‐optimized) DW gradient waveforms each single‐shot echo‐planar imaging (ssEPI) multishot EPI...
Background There is an unmet need for fully automated image prescription of the liver to enable efficient, reproducible MRI. Purpose To develop and evaluate artificial intelligence (AI)‐based prescription. Study Type Prospective. Population A total 570 female/469 male patients (age: 56 ± 17 years) with 72%/8%/20% assigned randomly training/validation/testing; two female/four healthy volunteers 31 6 years). Field Strength/Sequence 1.5 T, 3. 0 T ; spin echo, gradient bSSFP . Assessment 1039...
In this work we report initial results evaluating the imaging artifacts associated with different candidate materials for use in needle-based interventions. The impact of each material was assessed using MRT and DWI ADC as well compared local Bo field maps.
This work developed a novel automated AI-based method for liver image prescription from localizer and evaluated it in large retrospective patient cohort (1,039 patients training/testing), across pathologies, field strengths, against radiologists’ inter-reader reproducibility performance. 3D axial achieved S/I shift of <2.3 cm compared to manual 99.5% test dataset. The AI performed well all sub-cohorts better than We successfully implemented the on clinical MR system, which demonstrated...
An automated AI-based method for liver image prescription from a localizer was recently proposed. In this work, AI further evaluated in larger retrospective patient cohort (1,039 patients training/testing), across pathologies, field strengths, and against radiologists&rsquo; inter-reader reproducibility performance. 3D axial achieved S/I shift of &lt;2.3 cm compared to manual 99.5% test dataset. The performed well all sub-cohorts better than successfully implemented on clinical MR...
A major limitation of standard diffusion-weighted imaging (DWI) in the various organs, including liver, is presence artifacts, bias, and poor reproducibility resulting from physiologic cardiovascular-induced motion. Recent efforts to develop motion-robust DWI sequences show promise, but there an important unmet need for highly controlled validation motion phantoms. This study presents fabrication anthropomorphic liver flow phantom that can produce compressive generate control artifact seen...
There is growing evidence of tri-exponential IVIM-DWI signal behavior in the liver. However, estimates often suffer from instability associated with separating three decaying components sampled along a single dimension (i.e., b-value). In this work, standard monopolar and noise-optimized 2D (b-M 1 ) data sampling schemes were acquired. From each acquisition, ROI-based fitting blood velocity deviation distribution (BVD) method performed. The estimated parameters compared across acquisitions...
Conventional liver diffusion MRI acquisitions suffer from several challenges including pulsatile motion-induced signal voids and B0-induced distortions. In this study, we validated motion-robust, blood-suppressed, reduced-distortion techniques in an anatomically accurate phantom controlled motion. By combining optimized motion-compensated, blood-suppressed waveforms (i.e. MODI) with msEPI acquisitions, superior image quality quantitative performance can be achieved the presence of various...
Neural network based image compression has made significant progress in recent years. The learned codecs are commonly reported to outperform their conventional counterparts perceptual quality. Despite the superior performance, much more complex decode, which hinders usage practice. Without a advance hardware capability, codec will likely remain primary component for large scale services. It is therefore desirable improve quality of codecs. In this paper, we present an adaptive quantization...