A5100705184- Medical Imaging Techniques and Applications
- Radiomics and Machine Learning in Medical Imaging
- Advanced X-ray and CT Imaging
- Advanced MRI Techniques and Applications
- Cardiac Imaging and Diagnostics
- Advanced Radiotherapy Techniques
- Radiopharmaceutical Chemistry and Applications
- Lung Cancer Diagnosis and Treatment
- Image and Signal Denoising Methods
- Radiation Detection and Scintillator Technologies
- Medical Imaging and Pathology Studies
- MRI in cancer diagnosis
- Medical Image Segmentation Techniques
- Radiation Dose and Imaging
- Advanced Semiconductor Detectors and Materials
- Generative Adversarial Networks and Image Synthesis
- Cancer, Hypoxia, and Metabolism
- Chemotherapy-induced cardiotoxicity and mitigation
- Atomic and Subatomic Physics Research
- Advanced Image Processing Techniques
- Artificial Intelligence in Healthcare and Education
- Effects of Vibration on Health
- Ergonomics and Musculoskeletal Disorders
- Advanced Neural Network Applications
- Hydrogen's biological and therapeutic effects
Yale University
2016-2025
North China Electric Power University
2023
Chinese General Hospital College of Nursing and Liberal Arts
2023
Chinese People's Liberation Army
2023
Yale Cancer Center
2012-2021
Jaguar Land Rover (United Kingdom)
2021
University of New Haven
2015-2020
University of Southampton
2020
Lantheus Medical Imaging (United States)
2014
VA Connecticut Healthcare System
2014
Multi-contrast MRI (MC-MRI) captures multiple complementary imaging modalities to aid in radiological decision-making. Given the need for lowering time cost of acquisitions, current deep accelerated reconstruction networks focus on exploiting redundancy between contrasts. However, existing works are largely supervised with paired data and/or prohibitively expensive fully-sampled sequences. Further, typically rely convolutional architectures which limited their capacity model long-range...
The collaboration of Yale, the University California, Davis, and United Imaging Healthcare has successfully developed NeuroEXPLORER, a dedicated human brain PET imager with high spatial resolution, sensitivity, built-in 3-dimensional camera for markerless continuous motion tracking. It depth-of-interaction time-of-flight resolutions, along 52.4-cm transverse field view (FOV) an extended axial FOV (49.5 cm) to enhance sensitivity. Here, we present physical characterization, performance...
Our aim is to investigate the impact of respiratory motion on tumor quantification and delineation in static PET/CT imaging using a population patient traces. A total 1295 traces acquired during whole body were classified into three types according qualitative shape their signal histograms. Each trace was scaled diaphragm amplitudes (6 mm, 11 mm 16 mm) drive computer simulation that validated with physical phantom experiment. Three lung lesions one liver lesion simulated diameters 1 cm 2 cm....
Reducing radiation dose is important for PET imaging. However, reducing injection doses causes increased image noise and low signal-to-noise ratio (SNR), subsequently affecting diagnostic quantitative accuracies. Deep learning methods have shown a great potential to reduce the improve SNR in data. In this work, we comprehensively investigated accuracy of small lung nodules, addition visual quality, using deep based denoising oncological We applied optimized an advanced method on U-net...
Hydrogen exerts beneficial effects in disease animal models of ischemia-reperfusion injury as well inflammatory and neurological disease. Additionally, molecular hydrogen is useful for various novel medical therapeutic applications the clinical setting. In present study, concentration rat blood tissue was estimated. Wistar rats were orally administered super-rich water (HSRW), intraperitoneal intravenous administration saline (HSRS), inhalation gas. A new method determining then applied...
The nuclear medicine field has seen a rapid expansion of academic and commercial interest in developing artificial intelligence (AI) algorithms. Users developers can avoid some the pitfalls AI by recognizing following best practices algorithm development. In this article, recommendations on technical for algorithms are provided, beginning with general then continuing descriptions how one might practice these principles specific topics within medicine. This report was produced Task Force...
An important need exists for strategies to perform rigorous objective clinical-task-based evaluation of artificial intelligence (AI) algorithms nuclear medicine. To address this need, we propose a 4-class framework evaluate AI promise, technical task-specific efficacy, clinical decision making, and postdeployment efficacy. We provide best practices each these classes. Each class yields claim that provides descriptive performance the algorithm. Key are tabulated as RELAINCE (Recommendations...
Purpose: To minimize respiratory motion artifacts, this work proposes quiescent period gating (QPG) methods that extract PET data from the end‐expiration and form a single frame with reduced improved signal‐to‐noise properties. Methods: Two QPG are proposed evaluated. Histogram‐based (H‐QPG) extracts fraction of determined by window displacement signal histogram. Cycle‐based (C‐QPG) below specified threshold maximum amplitude each individual cycle. Performances both were compared to ungated...
Respiratory motion degrades the detection and quantification capabilities of PET/CT imaging. Moreover, mismatch between a fast helical CT image time-averaged PET due to respiratory results in additional attenuation correction artifacts inaccurate localization. Current compensation approaches typically have 3 limitations: among respiration-gated images (CTAC) map can introduce gated reconstructions that subsequently affect accuracy estimation; sinogram-based do not correct for intragate...
Limited view tomographic reconstruction aims to reconstruct a image from limited number of projection views arising sparse or angle acquisitions that reduce radiation dose shorten scanning time. However, such suffers severe artifacts due the incompleteness sinogram. To derive quality reconstruction, previous methods use UNet-like neural architectures directly predict full data; but these leave deep network architecture issue largely intact and cannot guarantee consistency between sinogram...
Dedicated cardiac SPECT scanners with cadmium-zinc-telluride cameras have shown capabilities for shortened scan times or reduced radiation doses, as well improved image quality. Since most dedicated do not integrated CT, quantification attenuation correction (AC) is challenging and artifacts are routinely encountered in daily clinical practice. In this work, we demonstrated a direct AC technique using deep learning (DL) myocardial perfusion imaging (MPI). <b>Methods:</b> an institutional...
Trustworthiness is a core tenet of medicine. The patient–physician relationship evolving from dyad to broader ecosystem health care. With the emergence artificial intelligence (AI) in medicine, elements trust must be revisited. We envision road map for establishment trustworthy AI ecosystems nuclear In this report, contextualized history technologic revolutions. Opportunities applications medicine related diagnosis, therapy, and workflow efficiency, as well emerging challenges critical...
To reduce the potential risk of radiation to patient, low-dose computed tomography (LDCT) has been widely adopted in clinical practice for reconstructing cross-sectional images using sinograms with reduced x-ray flux. The LDCT image quality is often degraded by different levels noise depending on protocols. will be further when patient metallic implants, where suffers from additional streak artifacts along amplified levels, thus affecting medical diagnosis and other CT-related applications....
Purpose: The goal of this work was to investigate the effects MRI surface coils on attenuation-corrected PET emission data. authors studied cases where either an or a CT scan would be used provide attenuation correction (AC). Combined MR/PET scanners that use for AC (MR-AC) face challenge absent in MR images and thus cannot directly account coils. Combining could achieved by transporting subject stereotactically registered table between independent scanners. In case, conventional CT-AC...
Purpose: We present a method to correct respiratory motion blurring in PET/CT imaging using internal–external (INTEX) correlation. The internal of known tumor is derived from respiratory-gated PET images; this then correlated with external signals determine the complete information during scan. Methods: For each data, listmode data were phase-gated into five bins and reconstructed. centroid targeted bin was determined corresponding mean displacement externally monitored signal. Based on...
Data-driven respiratory gating techniques were developed to correct for motion in PET studies, without the help of external tracking systems. Due greatly increased image noise gated reconstructions, it is desirable develop a data-driven event-by-event correction method. In this study, using Centroid-of-distribution (COD) algorithm, we established technique TOF list-mode data, and investigated its performance by comparing with an system-based Ten human scans pancreatic β-cell tracer...