Search

Your search keyword '"Chang, A. Y"' showing total 20,662 results
Start Over Author "Chang, A. Y"

Refine your results

more »

more »

more »

more »

more »

more »
20,662 results on '"Chang, A. Y"'

1. The Detection of H$_2$O Maser Emission from mid-IR Red Galaxies

Author

Kuo, C. Y., Tai, C. Y., Constantin, A., Braatz, J. A., Chung, H. H., Chen, B. Y., Pesce, D. W., Impellizzeri, C. M. V., Gao, F., and Chang, Y. Y.

Subjects
Astrophysics - Astrophysics of Galaxies, Astrophysics - Cosmology and Nongalactic Astrophysics
Abstract

We report the detection of H$_2$O maser emission in 4 out of 77 (5.2%) mid-IR red galaxies that meet the color criteria of $W1-W2 > 0.5$ and $W1-W4 > 7$ and are classified as Type-2 AGNs based on optical, near-IR, and mid-IR spectral energy distribution (SED) fitting. Here, $W1$, $W2$, and $W4$ represent the IR magnitudes at 3.4, 4.6, and 22 micron, respectively, as measured by the Wide-field Infrared Survey Explorer. Three of the four newly identified maser galaxies are classified as either Seyfert 2 or LINER systems, but none are disk maser systems. Our analysis indicates that AGN identifications based solely on SED fitting are unreliable, resulting in an unexpectedly low detection rate. By restricting our sample to optically classified Type 2 AGNs that satisfy the mid-IR color criteria, we achieve a maser detection rate of ~13-18%, aligning with previous predictions for mid-IR red sources. These selection criteria are the most effective to date for facilitating new maser detections, particularly in light of the recent identification of additional Type 2 AGNs identified from ongoing galaxy and AGN surveys., Comment: 11 pages, 3 tables, 3 figures

Published
2024

2. Significant age‐related differences between lower leg muscles of older and younger female subjects detected by ultrashort echo time magnetization transfer modeling

Author

Jerban, Saeed, Mohammadi, Hamidreza Shaterian, Athertya, Jiyo S, Afsahi, Amir Masoud, Shojaeiadib, Niloofar, Moazamian, Dina, Ward, Samuel R, Woods, Gina, Chung, Christine B, Du, Jiang, and Chang, Eric Y

Subjects
Biomedical and Clinical Sciences, Clinical Sciences, Clinical Research, Aging, Biomedical Imaging, Musculoskeletal, Humans, Female, Adult, Muscle, Skeletal, Leg, Magnetic Resonance Imaging, Aged, Young Adult, Reproducibility of Results, macromolecular protons, magnetization transfer, MRI, muscle, myotendinous junction, UTE, Medicinal and Biomolecular Chemistry, Biomedical Engineering, Nuclear Medicine & Medical Imaging, Clinical sciences, Biomedical engineering
Abstract

Magnetization transfer (MT) magnetic resonance imaging (MRI) can be used to estimate the fraction of water and macromolecular proton pools in tissues. MT modeling paired with ultrashort echo time acquisition (UTE-MT modeling) has been proposed to improve the evaluation of the myotendinous junction and fibrosis in muscle tissues, which the latter increases with aging. This study aimed to determine if the UTE-MT modeling technique is sensitive to age-related changes in the skeletal muscles of the lower leg. Institutional review board approval was obtained, and all recruited subjects provided written informed consent. The legs of 31 healthy younger (28.1 ± 6.1 years old, BMI = 22.3 ± 3.5) and 20 older (74.7 ± 5.5 years old, BMI = 26.7 ± 5.9) female subjects were imaged using UTE sequences on a 3 T MRI scanner. MT ratio (MTR), macromolecular fraction (MMF), macromolecular T2 (T2-MM), and water T2 (T2-W) were calculated using UTE-MT modeling for the anterior tibialis (ATM), posterior tibialis (PTM), soleus (SM), and combined lateral muscles. Results were compared between groups using the Wilcoxon rank sum test. Three independent observers selected regions of interest (ROIs) and processed UTE-MRI images separately, and the intraclass correlation coefficient (ICC) was calculated for a reproducibility study. Significantly lower mean MTR and MMF values were present in the older compared with the younger group in all studied lower leg muscles. T2-MM showed significantly lower values in the older group only for PTM and SM muscles. In contrast, T2-W showed significantly higher values in the older group. The age-related differences were more pronounced for MMF (-17 to -19%) and T2-W (+20 to 47%) measurements in all muscle groups compared with other investigated MR measures. ICCs were higher than 0.93, indicating excellent consistency between the ROI selection and MRI measurements of independent readers. As demonstrated by significant differences between younger and older groups, this research emphasizes the potential of UTE-MT MRI techniques in evaluating age-related skeletal muscle changes.

Published
2024

3. Quantitative ultrashort echo time MR imaging of knee osteochondral junction: An ex vivo feasibility study

Author

Athertya, Jiyo S, Suprana, Arya, Lo, James, Lombardi, Alecio F, Moazamian, Dina, Chang, Eric Y, Du, Jiang, and Ma, Yajun

Subjects
Engineering, Biomedical Engineering, Arthritis, Aging, Clinical Research, Osteoarthritis, Biomedical Imaging, 4.2 Evaluation of markers and technologies, Musculoskeletal, Humans, Feasibility Studies, Magnetic Resonance Imaging, Knee Joint, Male, Female, Middle Aged, Cartilage, Articular, Aged, Time Factors, Osteoarthritis, Knee, Adult, Cadaver, knee joint, osteochondral junction, quantitative, ultrashort echo time, Medicinal and Biomolecular Chemistry, Clinical Sciences, Nuclear Medicine & Medical Imaging, Clinical sciences, Biomedical engineering
Abstract

Compositional changes can occur in the osteochondral junction (OCJ) during the early stages and progressive disease evolution of knee osteoarthritis (OA). However, conventional magnetic resonance imaging (MRI) sequences are not able to image these regions efficiently because of the OCJ region's rapid signal decay. The development of new sequences able to image and quantify OCJ region is therefore highly desirable. We developed a comprehensive ultrashort echo time (UTE) MRI protocol for quantitative assessment of OCJ region in the knee joint, including UTE variable flip angle technique for T1 mapping, UTE magnetization transfer (UTE-MT) modeling for macromolecular proton fraction (MMF) mapping, UTE adiabatic T1ρ (UTE-AdiabT1ρ) sequence for T1ρ mapping, and multi-echo UTE sequence for T2* mapping. B1 mapping based on the UTE actual flip angle technique was utilized for B1 correction in T1, MMF, and T1ρ measurements. Ten normal and one abnormal cadaveric human knee joints were scanned on a 3T clinical MRI scanner to investigate the feasibility of OCJ imaging using the proposed protocol. Volumetric T1, MMF, T1ρ, and T2* maps of the OCJ, as well as the superficial and full-thickness cartilage regions, were successfully produced using the quantitative UTE imaging protocol. Significantly lower T1, T1ρ, and T2* relaxation times were observed in the OCJ region compared with those observed in both the superficial and full-thickness cartilage regions, whereas MMF showed significantly higher values in the OCJ region. In addition, all four UTE biomarkers showed substantial differences in the OCJ region between normal and abnormal knees. These results indicate that the newly developed 3D quantitative UTE imaging techniques are feasible for T1, MMF, T1ρ, and T2* mapping of knee OCJ, representative of a promising approach for the evaluation of compositional changes in early knee OA.

Published
2024

4. Deep Convolutional Neural Network for Dedicated Regions-of-Interest Based Multi-Parameter Quantitative Ultrashort Echo Time (UTE) Magnetic Resonance Imaging of the Knee Joint

Author

Lu, Xing, Ma, Yajun, Chang, Eric Y, Athertya, Jiyo, Jang, Hyungseok, Jerban, Saeed, Covey, Dana C, Bukata, Susan, Chung, Christine B, and Du, Jiang

Subjects
Biomedical and Clinical Sciences, Clinical Sciences, Arthritis, Clinical Research, Biomedical Imaging, Bioengineering, Networking and Information Technology R&D (NITRD), Machine Learning and Artificial Intelligence, Osteoarthritis, Health Disparities, Musculoskeletal, Humans, Magnetic Resonance Imaging, Knee Joint, Neural Networks, Computer, Male, Female, Deep Learning, Middle Aged, Adult, Osteoarthritis, Knee, Image Processing, Computer-Assisted, Cartilage, Articular, Aged, Image Interpretation, Computer-Assisted, Quantitative MRI, Automated segmentation, DCNN, RMQ-Net, UTE, Knee joint, OA
Abstract

We proposed an end-to-end deep learning convolutional neural network (DCNN) for region-of-interest based multi-parameter quantification (RMQ-Net) to accelerate quantitative ultrashort echo time (UTE) MRI of the knee joint with automatic multi-tissue segmentation and relaxometry mapping. The study involved UTE-based T1 (UTE-T1) and Adiabatic T1ρ (UTE-AdiabT1ρ) mapping of the knee joint of 65 human subjects, including 20 normal controls, 29 with doubtful-minimal osteoarthritis (OA), and 16 with moderate-severe OA. Comparison studies were performed on UTE-T1 and UTE-AdiabT1ρ measurements using 100%, 43%, 26%, and 18% UTE MRI data as the inputs and the effects on the prediction quality of the RMQ-Net. The RMQ-net was modified and retrained accordingly with different combinations of inputs. Both ROI-based and voxel-based Pearson correlation analyses were performed. High Pearson correlation coefficients were achieved between the RMQ-Net predicted UTE-T1 and UTE-AdiabT1ρ results and the ground truth for segmented cartilage with acceleration factors ranging from 2.3 to 5.7. With an acceleration factor of 5.7, the Pearson r-value achieved 0.908 (ROI-based) and 0.945 (voxel-based) for UTE-T1, and 0.733 (ROI-based) and 0.895 (voxel-based) for UTE-AdiabT1ρ, correspondingly. The results demonstrated that RMQ-net can significantly accelerate quantitative UTE imaging with automated segmentation of articular cartilage in the knee joint.

Published
2024

5. Unlocking the Wisdom of Large Language Models: An Introduction to The Path to Artificial General Intelligence

Author

Chang, Edward Y.

Subjects
Computer Science - Artificial Intelligence, I.2.7
Abstract

This booklet, "Unlocking the Wisdom of LLM Collaborative Intelligence," introduces the comprehensive work "The Path to Artificial General Intelligence." Through ten aphorisms, it distills the core principles of LLM Collaborative Intelligence (LCI) as a promising framework toward achieving AGI. The booklet also offers titles, abstracts, and introductions from the main chapters, along with the first two chapters in full. The second edition, released this week, includes significant enhancements to Chapters 6 to 9 and a revised preface addressing Yann LeCun's skepticism about AGI. LeCun argues that LLMs lack memory, planning, and grounding, but we propose that LCI's collaborative architecture, involving multimodal LLMs with executive, legislative, and judicial roles, overcomes these limitations. Chapters on SocraSynth, EVINCE, consciousness modeling, and behavior modeling demonstrate that collaborative LLMs with checks and balances can achieve intelligence beyond any single model's capability. By combining complementary strengths, such as world modeling and advanced sensory capabilities, LCI enables models to work together and perceive reality beyond human limitations. As with human institutions, progress depends on cooperation, not isolation. Collaborative LLMs may unlock new levels of intelligence, paving the way toward AGI., Comment: 153 pages, 5 figures

Published
2024

6. EVINCE: Optimizing Adversarial LLM Dialogues via Conditional Statistics and Information Theory

Author

Chang, Edward Y.

Subjects
Computer Science - Artificial Intelligence, I.2.7
Abstract

This paper introduces EVINCE (Entropy and Variation IN Conditional Exchanges), a framework that optimizes multi-LLM dialogues using conditional statistics and information theory. EVINCE introduces dual entropy optimization to balance perspective diversity with prior knowledge, providing quantitative measures for modulating LLM interactions. Through information-theoretic metrics and mutual information optimization, the framework demonstrates consistent improvement over single-LLM performance in applications ranging from disease diagnosis to news debiasing. We present theoretical foundations and empirical validation for this structured approach to LLM collaboration., Comment: 32 pages, 9 figures, 10 tables. arXiv admin note: substantial text overlap with arXiv:2405.15808

Published
2024

7. Uncovering Biases with Reflective Large Language Models

Author

Chang, Edward Y.

Subjects
Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Computer Science - Machine Learning, I.2.7
Abstract

Biases and errors in human-labeled data present significant challenges for machine learning, especially in supervised learning reliant on potentially flawed ground truth data. These flaws, including diagnostic errors and societal biases, risk being propagated and amplified through models trained using maximum likelihood estimation. We present the Reflective LLM Dialogue Framework RLDF, which leverages structured adversarial dialogues between multiple instances of a single LLM or different LLMs to uncover diverse perspectives and correct inconsistencies. By conditioning LLMs to adopt opposing stances, RLDF enables systematic bias detection through conditional statistics, information theory, and divergence metrics. Experiments show RLDF successfully identifies potential biases in public content while exposing limitations in human-labeled data. Our framework supports measurable progress tracking and explainable remediation actions, offering a scalable approach for improving content neutrality through transparent, multi-perspective analysis., Comment: 18 pages, 4 figures, 9 tables

Published
2024

8. Robotically adjustable kinematics in a wrist-driven orthosis eases grasping across tasks

Author

Chang, Erin Y., McPherson, Andrew I. W., and Stuart, Hannah S.

Subjects
Computer Science - Robotics
Abstract

Without finger function, people with C5-7 spinal cord injury (SCI) regularly utilize wrist extension to passively close the fingers and thumb together for grasping. Wearable assistive grasping devices often focus on this familiar wrist-driven technique to provide additional support and amplify grasp force. Despite recent research advances in modernizing these tools, people with SCI often abandon such wearable assistive devices in the long term. We suspect that the wrist constraints imposed by such devices generate undesirable reach and grasp kinematics. Here we show that using continuous robotic motor assistance to give users more adaptability in their wrist posture prior to wrist-driven grasping reduces task difficulty and perceived exertion. Our results demonstrate that more free wrist mobility allows users to select comfortable and natural postures depending on task needs, which improves the versatility of the assistive grasping device for easier use across different hand poses in the arm's workspace. This behavior holds the potential to improve ease of use and desirability of future device designs through new modes of combining both body-power and robotic automation., Comment: 6 pages, 8 figures. Presented at the 2024 International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

Published
2024

9. Light Dark Matter Constraints from SuperCDMS HVeV Detectors Operated Underground with an Anticoincidence Event Selection

Author

SuperCDMS Collaboration, Albakry, M. F., Alkhatib, I., Alonso-González, D., Amaral, D. W. P., Anczarski, J., Aralis, T., Aramaki, T., Arnquist, I. J., Langroudy, I. Ataee, Azadbakht, E., Bathurst, C., Bhattacharyya, R., Biffl, A. J., Brink, P. L., Buchanan, M., Bunker, R., Cabrera, B., Calkins, R., Cameron, R. A., Cartaro, C., Cerdeño, D. G., Chang, Y. -Y., Chaudhuri, M., Chen, J. -H., Chen, R., Chott, N., Cooley, J., Coombes, H., Cushman, P., Cyna, R., Das, S., De Brienne, F., Dharani, S., di Vacri, M. L., Diamond, M. D., Elwan, M., Fascione, E., Figueroa-Feliciano, E., Fouts, K., Fritts, M., Germond, R., Ghaith, M., Golwala, S. R., Hall, J., Harms, S. A. S., Harris, K., Hassan, N., Hong, Z., Hoppe, E. W., Hsu, L., Huber, M. E., Iyer, V., Jardin, D., Kashyap, V. K. S., Keller, S. T. D., Kelsey, M. H., Kennard, K. T., Kubik, A., Kurinsky, N. A., Lee, M., Leyva, J., Liu, J., Liu, Y., Loer, B., Asamar, E. Lopez, Lukens, P., MacFarlane, D. B., Mahapatra, R., Mammo, J. S., Mast, N., Mayer, A. J., Theenhausen, H. Meyer zu, Michaud, É., Michielin, E., Mirabolfathi, N., Mirzakhani, M., Mohanty, B., Monteiro, D., Nelson, J., Neog, H., Novati, V., Orrell, J. L., Osborne, M. D., Oser, S. M., Pandey, L., Pandey, S., Partridge, R., Pedreros, D. S., Peng, W., Perna, L., Perry, W. L., Podviianiuk, R., Poudel, S. S., Pradeep, A., Pyle, M., Rau, W., Reid, E., Ren, R., Reynolds, T., Rios, M., Roberts, A., Robinson, A. E., Ryan, J. L., Saab, T., Sadek, D., Sadoulet, B., Sahoo, S. P., Saikia, I., Sander, J., Sattari, A., Schmidt, B., Schnee, R. W., Scorza, S., Serfass, B., Simchony, A., Sincavage, D. J., Sinervo, P., Street, J., Sun, H., Tanner, E., Terry, G. D., Toback, D., Verma, S., Villano, A. N., von Krosigk, B., Watkins, S. L., Wen, O., Williams, Z., Wilson, M. J., Winchell, J., Wykoff, K., Yellin, S., Young, B. A., Yu, T. C., Zatschler, B., Zatschler, S., Zaytsev, A., Zhang, E., Zheng, L., Zuniga, A., and Zurowski, M. J.

Subjects
High Energy Physics - Experiment, Astrophysics - Cosmology and Nongalactic Astrophysics, Physics - Instrumentation and Detectors
Abstract

This article presents constraints on dark-matter-electron interactions obtained from the first underground data-taking campaign with multiple SuperCDMS HVeV detectors operated in the same housing. An exposure of 7.63 g-days is used to set upper limits on the dark-matter-electron scattering cross section for dark matter masses between 0.5 and 1000 MeV/$c^2$, as well as upper limits on dark photon kinetic mixing and axion-like particle axioelectric coupling for masses between 1.2 and 23.3 eV/$c^2$. Compared to an earlier HVeV search, sensitivity was improved as a result of an increased overburden of 225 meters of water equivalent, an anticoincidence event selection, and better pile-up rejection. In the case of dark-matter-electron scattering via a heavy mediator, an improvement by up to a factor of 25 in cross-section sensitivity was achieved., Comment: 7 pages + title and references, 4 figures, and 1 table

Published
2024

10. Qualitative and Quantitative MR Imaging of the Cartilaginous Endplate: A Review

Author

Wei, Zhao, Athertya, Jiyo S, Chung, Christine B, Bydder, Graeme M, Chang, Eric Y, Du, Jiang, Wenhui, Yang, and Ma, Yajun

Subjects
Biomedical and Clinical Sciences, Clinical Sciences, Biomedical Imaging, cartilaginous endplate, MRI, morphology, quantification, Physical Sciences, Engineering, Medical and Health Sciences, Nuclear Medicine & Medical Imaging, Clinical sciences
Abstract

The cartilaginous endplate (CEP) plays a pivotal role in facilitating the supply of nutrients and, transport of metabolic waste, as well as providing mechanical support for the intervertebral disc (IVD). Recent technological advances have led to a surge in MR imaging studies focused on the CEP. This article describes the anatomy and functions of the CEP as well as MRI techniques for both qualitative and quantitative assessment of the CEP. Effective CEP MR imaging sequences require two key features: high spatial resolution and relatively short echo time. High spatial resolution spoiled gradient echo (SPGR) and ultrashort echo time (UTE) sequences, fulfilling these requirements, are the basis for most of the sequences employed in CEP imaging. This article reviews existing sequences for qualitative CEP imaging, such as the fat-suppressed SPGR and UTE, dual-echo subtraction UTE, inversion recovery prepared and fat-suppressed UTE, and dual inversion recovery prepared UTE sequences. These sequences are employed together with other techniques for quantitative CEP imaging, including measurements of T2*, T2, T1, T1ρ, magnetization transfer, perfusion, and diffusion tensor parameters. EVIDENCE LEVEL: 1 TECHNICAL EFFICACY: Stage 2.

Published
2024

11. Towards assessing and improving the reliability of ultrashort echo time quantitative magnetization transfer (UTE-qMT) MRI of cortical bone: In silico and ex vivo study

Author

Shin, Soo Hyun, Moazamian, Dina, Tang, Qingbo, Jerban, Saeed, Ma, Yajun, Du, Jiang, and Chang, Eric Y

Subjects
Engineering, Physical Sciences, Biomedical Engineering, Bioengineering, Biomedical Imaging, Clinical Research, Ultrashort echo time, Quantitative magnetization transfer, MRI, Cortical bone, Digital phantom, Animals, Rats, Magnetic Resonance Imaging, Reproducibility of Results, Phantoms, Imaging, Algorithms, Computer Simulation, Image Processing, Computer-Assisted, Signal-To-Noise Ratio, Cortical Bone, Nuclear Medicine & Medical Imaging
Abstract

ObjectiveTo assess and improve the reliability of the ultrashort echo time quantitative magnetization transfer (UTE-qMT) modeling of the cortical bone.Materials and methodsSimulation-based digital phantoms were created that mimic the UTE-qMT properties of cortical bones. A wide range of SNR from 25 to 200 was simulated by adding different levels of noise to the synthesized MT-weighted images to assess the effect of SNR on UTE-qMT fitting results. Tensor-based denoising algorithm was applied to improve the fitting results. These results from digital phantom studies were validated via ex vivo rat leg bone scans.ResultsThe selection of initial points for nonlinear fitting and the number of data points tested for qMT analysis have minimal effect on the fitting result. Magnetization exchange rate measurements are highly dependent on the SNR of raw images, which can be substantially improved with an appropriate denoising algorithm that gives similar fitting results from the raw images with an 8-fold higher SNR.DiscussionThe digital phantom approach enables the assessment of the reliability of bone UTE-qMT fitting by providing the known ground truth. These findings can be utilized for optimizing the data acquisition and analysis pipeline for UTE-qMT imaging of cortical bones.

Published
2024

12. Cohesin prevents cross-domain gene coactivation

Author

Dong, Peng, Zhang, Shu, Gandin, Valentina, Xie, Liangqi, Wang, Lihua, Lemire, Andrew L, Li, Wenhong, Otsuna, Hideo, Kawase, Takashi, Lander, Arthur D, Chang, Howard Y, and Liu, Zhe J

Subjects
Biochemistry and Cell Biology, Bioinformatics and Computational Biology, Genetics, Biological Sciences, Biotechnology, Human Genome, 1.1 Normal biological development and functioning, Generic health relevance, Cohesins, Cell Cycle Proteins, Chromosomal Proteins, Non-Histone, Chromatin, Gene Expression Regulation, Promoter Regions, Genetic, Single-Cell Analysis, Saccharomyces cerevisiae, Transcriptome, Medical and Health Sciences, Developmental Biology, Agricultural biotechnology, Bioinformatics and computational biology
Abstract

The contrast between the disruption of genome topology after cohesin loss and the lack of downstream gene expression changes instigates intense debates regarding the structure-function relationship between genome and gene regulation. Here, by analyzing transcriptome and chromatin accessibility at the single-cell level, we discover that, instead of dictating population-wide gene expression levels, cohesin supplies a general function to neutralize stochastic coexpression tendencies of cis-linked genes in single cells. Notably, cohesin loss induces widespread gene coactivation and chromatin co-opening tens of million bases apart in cis. Spatial genome and protein imaging reveals that cohesin prevents gene co-bursting along the chromosome and blocks spatial mixing of transcriptional hubs. Single-molecule imaging shows that cohesin confines the exploration of diverse enhancer and core promoter binding transcriptional regulators. Together, these results support that cohesin arranges nuclear topology to control gene coexpression in single cells.

Published
2024

13. Ultrashort echo time MRI detects significantly lower collagen but higher pore water in the tibial cortex of female patients with osteopenia and osteoporosis

Author

Jerban, Saeed, Ma, Yajun, Wei, Zhao, Shen, Meghan, Ibrahim, Zubaid, Jang, Hyungseok, Lu, Pengzhe, Chang, Douglas G, Woods, Gina, Chung, Christine B, Chang, Eric Y, and Du, Jiang

Subjects
Biomedical and Clinical Sciences, Clinical Sciences, Aging, Osteoporosis, Clinical Research, Musculoskeletal, Humans, Female, Tibia, Magnetic Resonance Imaging, Bone Diseases, Metabolic, Adult, Aged, Collagen, Middle Aged, Cortical Bone, Water, Porosity, MRI, cortical bone, ultrashort echo time, pore water, magnetization transfer, collagen proton density, Biological Sciences, Engineering, Medical and Health Sciences, Anatomy & Morphology, Biological sciences, Biomedical and clinical sciences
Abstract

Ultrashort echo time (UTE) MRI can quantify the major proton pool densities in cortical bone, including total (TWPD), bound (BWPD), and pore water (PWPD) proton densities, as well as the macromolecular proton density (MMPD), associated with the collagen content, which is calculated using macromolecular fraction (MMF) from UTE magnetization transfer (UTE-MT) modeling. This study aimed to investigate the differences in water and collagen contents in tibial cortical bone, between female osteopenia (OPe) patients, osteoporosis (OPo) patients, and young participants (Young). Being postmenopausal and above 55 yr old were the inclusion criteria for OPe and OPo groups. The tibial shaft of 14 OPe (72.5 ± 6.8 yr old), 31 OPo (72.0 ± 6.4 yr old), and 31 young subjects (28.0 ± 6.1 yr old) were scanned using a knee coil on a clinical 3T scanner. Basic UTE, inversion recovery UTE, and UTE-MT sequences were performed. Investigated biomarkers were compared between groups using Kruskal-Wallis test. Spearman's correlation coefficients were calculated between the TH DXA T-score and UTE-MRI results. MMF, BWPD, and MMPD were significantly lower in OPo patients than in the young group, whereas T1, TWPD, and PWPD were significantly higher in OPo patients. The largest OPo/Young average percentage differences were found in MMF (41.9%), PWPD (103.5%), and MMPD (64.0%). PWPD was significantly higher (50.7%), while BWPD was significantly lower (16.4%) in OPe than the Young group on average. MMF was found to be significantly lower (27%) in OPo patients compared with OPe group. T1, MMF, TWPD, PWPD, and MMPD values significantly correlated with the TH DXA T-scores (provided by the patients and only available for OPe and OPo patients). DXA T-score showed the highest correlations with PWPD (R = 0.55) and MMF (R = 0.56) values. TWPD, PWPD, and MMF estimated using the UTE-MRI sequences were recommended to evaluate individuals with OPe and OPo.

Published
2024

14. More accurate trabecular bone imaging using UTE MRI at the resonance frequency of fat

Author

Jerban, Saeed, Moazamian, Dina, Mohammadi, Hamidreza Shaterian, Ma, Yajun, Jang, Hyungseok, Namiranian, Behnam, Shin, Soo Hyun, Alenezi, Salem, Shah, Sameer B, Chung, Christine B, Chang, Eric Y, and Du, Jiang

Subjects
Biomedical and Clinical Sciences, Clinical Sciences, Biomedical Imaging, Clinical Research, 4.2 Evaluation of markers and technologies, Humans, Magnetic Resonance Imaging, Cancellous Bone, Aged, Male, Female, Adipose Tissue, MRI, Fat peak frequency, Trabecular bone, UTE, Microstructure, Biological Sciences, Engineering, Medical and Health Sciences, Endocrinology & Metabolism, Clinical sciences
Abstract

High-resolution magnetic resonance imaging (HR-MRI) has been increasingly used to assess the trabecular bone structure. High susceptibility at the marrow/bone interface may significantly reduce the marrow's apparent transverse relaxation time (T2*), overestimating trabecular bone thickness. Ultrashort echo time MRI (UTE-MRI) can minimize the signal loss caused by susceptibility-induced T2* shortening. However, UTE-MRI is sensitive to chemical shift artifacts, which manifest as spatial blurring and ringing artifacts partially due to non-Cartesian sampling. In this study, we proposed UTE-MRI at the resonance frequency of fat to minimize marrow-related chemical shift artifacts and the overestimation of trabecular thickness. Cubes of trabecular bone from six donors (75 ± 4 years old) were scanned using a 3 T clinical scanner at the resonance frequencies of fat and water, respectively, using 3D UTE sequences with five TEs (0.032, 1.1, 2.2, 3.3, and 4.4 ms) and a clinical 3D gradient echo (GRE) sequence at 0.2 × 0.2 × 0.4 mm3 voxel size. Trabecular bone thickness was measured in 30 regions of interest (ROIs) per sample. MRI results were compared with thicknesses obtained from micro-computed tomography (μCT) at 50 μm3 voxel size. Linear regression models were used to calculate the coefficient of determination between MRI- and μCT-based trabecular thickness. All MRI-based trabecular thicknesses showed significant correlations with μCT measurements. The correlations were higher (examined with paired Student's t-test, P

Published
2024

15. Realization of a crosstalk-free multi-ion node for long-distance quantum networking

Author

Lai, P. -C., Wang, Y., Shi, J. -X., Cui, Z. -B., Wang, Z. -Q., Zhang, S., Liu, P. -Y., Tian, Z. -C., Sun, Y. -D., Chang, X. -Y., Qi, B. -X., Huang, Y. -Y., Zhou, Z. -C., Wu, Y. -K., Xu, Y., Pu, Y. -F., and Duan, L. -M.

Subjects
Quantum Physics
Abstract

Trapped atomic ions constitute one of the leading physical platforms for building the quantum repeater nodes to realize large-scale quantum networks. In a long-distance trapped-ion quantum network, it is essential to have crosstalk-free dual-type qubits: one type, called the communication qubit, to establish entangling interface with telecom photons; and the other type, called the memory qubit, to store quantum information immune from photon scattering under entangling attempts. Here, we report the first experimental implementation of a telecom-compatible and crosstalk-free quantum network node based on two trapped $^{40}$Ca$^{+}$ ions. The memory qubit is encoded on a long-lived metastable level to avoid crosstalk with the communication qubit encoded in another subspace of the same ion species, and a quantum wavelength conversion module is employed to generate ion-photon entanglement over a $12\,$km fiber in a heralded style. Our work therefore constitutes an important step towards the realization of quantum repeaters and long-distance quantum networks., Comment: 12 pages, 12 figures

Published
2024

16. Ensuring Ground Truth Accuracy in Healthcare with the EVINCE framework

Author

Chang, Edward Y.

Subjects
Computer Science - Artificial Intelligence, I.2.7
Abstract

Misdiagnosis is a significant issue in healthcare, leading to harmful consequences for patients. The propagation of mislabeled data through machine learning models into clinical practice is unacceptable. This paper proposes EVINCE, a system designed to 1) improve diagnosis accuracy and 2) rectify misdiagnoses and minimize training data errors. EVINCE stands for Entropy Variation through Information Duality with Equal Competence, leveraging this novel theory to optimize the diagnostic process using multiple Large Language Models (LLMs) in a structured debate framework. Our empirical study verifies EVINCE to be effective in achieving its design goals., Comment: 23 pages, 4 tables, 4 figures

Published
2024

17. Integrating Emotional and Linguistic Models for Ethical Compliance in Large Language Models

Author

Chang, Edward Y.

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence, I.2.7
Abstract

This research develops advanced methodologies for Large Language Models (LLMs) to better manage linguistic behaviors related to emotions and ethics. We introduce DIKE, an adversarial framework that enhances the LLMs' ability to internalize and reflect global human values, adapting to varied cultural contexts to promote transparency and trust among users. The methodology involves detailed modeling of emotions, classification of linguistic behaviors, and implementation of ethical guardrails. Our innovative approaches include mapping emotions and behaviors using self-supervised learning techniques, refining these guardrails through adversarial reviews, and systematically adjusting outputs to ensure ethical alignment. This framework establishes a robust foundation for AI systems to operate with ethical integrity and cultural sensitivity, paving the way for more responsible and context-aware AI interactions., Comment: 29 pages, 10 tables, 6 figures

Published
2024

18. Modeling Emotions and Ethics with Large Language Models

Author

Chang, Edward Y.

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence, I.2.0
Abstract

This paper explores the integration of human-like emotions and ethical considerations into Large Language Models (LLMs). We first model eight fundamental human emotions, presented as opposing pairs, and employ collaborative LLMs to reinterpret and express these emotions across a spectrum of intensity. Our focus extends to embedding a latent ethical dimension within LLMs, guided by a novel self-supervised learning algorithm with human feedback (SSHF). This approach enables LLMs to perform self-evaluations and adjustments concerning ethical guidelines, enhancing their capability to generate content that is not only emotionally resonant but also ethically aligned. The methodologies and case studies presented herein illustrate the potential of LLMs to transcend mere text and image generation, venturing into the realms of empathetic interaction and principled decision-making, thereby setting a new precedent in the development of emotionally aware and ethically conscious AI systems., Comment: 8 pages, 4 figures, 3 tables

Published
2024

19. Engineered extrachromosomal oncogene amplifications promote tumorigenesis

Author

Pradella, Davide, Zhang, Minsi, Gao, Rui, Yao, Melissa A., Gluchowska, Katarzyna M., Cendon-Florez, Ylenia, Mishra, Tanmay, La Rocca, Gaspare, Weigl, Moritz, Jiao, Ziqi, Nguyen, Hieu H. M., Lisi, Marta, Ozimek, Mateusz M., Mastroleo, Chiara, Chen, Kevin, Grimm, Felix, Luebeck, Jens, Zhang, Shu, Zolli, Andrea Alice, Sun, Eric G., Dameracharla, Bhargavi, Zhao, Zhengqiao, Pritykin, Yuri, Sigel, Carlie, Chang, Howard Y., Mischel, Paul S., Bafna, Vineet, Antonescu, Cristina R., and Ventura, Andrea

Published
2024
Full Text
View/download PDF

22. Coordinated inheritance of extrachromosomal DNAs in cancer cells

Author

Hung, King L., Jones, Matthew G., Wong, Ivy Tsz-Lo, Curtis, Ellis J., Lange, Joshua T., He, Britney Jiayu, Luebeck, Jens, Schmargon, Rachel, Scanu, Elisa, Brückner, Lotte, Yan, Xiaowei, Li, Rui, Gnanasekar, Aditi, Chamorro González, Rocío, Belk, Julia A., Liu, Zhonglin, Melillo, Bruno, Bafna, Vineet, Dörr, Jan R., Werner, Benjamin, Huang, Weini, Cravatt, Benjamin F., Henssen, Anton G., Mischel, Paul S., and Chang, Howard Y.

Published
2024
Full Text
View/download PDF

23. Origins and impact of extrachromosomal DNA

Author

Bailey, Chris, Pich, Oriol, Thol, Kerstin, Watkins, Thomas B. K., Luebeck, Jens, Rowan, Andrew, Stavrou, Georgia, Weiser, Natasha E., Dameracharla, Bhargavi, Bentham, Robert, Lu, Wei-Ting, Kittel, Jeanette, Yang, S. Y. Cindy, Howitt, Brooke E., Sharma, Natasha, Litovchenko, Maria, Salgado, Roberto, Hung, King L., Cornish, Alex J., Moore, David A., Houlston, Richard S., Bafna, Vineet, Chang, Howard Y., Nik-Zainal, Serena, Kanu, Nnennaya, McGranahan, Nicholas, Flanagan, Adrienne M., Mischel, Paul S., Jamal-Hanjani, Mariam, and Swanton, Charles

Published
2024
Full Text
View/download PDF

24. Enhancing transcription–replication conflict targets ecDNA-positive cancers

Author

Tang, Jun, Weiser, Natasha E., Wang, Guiping, Chowdhry, Sudhir, Curtis, Ellis J., Zhao, Yanding, Wong, Ivy Tsz-Lo, Marinov, Georgi K., Li, Rui, Hanoian, Philip, Tse, Edison, Mojica, Salvador Garcia, Hansen, Ryan, Plum, Joshua, Steffy, Auzon, Milutinovic, Snezana, Meyer, S. Todd, Luebeck, Jens, Wang, Yanbo, Zhang, Shu, Altemose, Nicolas, Curtis, Christina, Greenleaf, William J., Bafna, Vineet, Benkovic, Stephen J., Pinkerton, Anthony B., Kasibhatla, Shailaja, Hassig, Christian A., Mischel, Paul S., and Chang, Howard Y.

Published
2024
Full Text
View/download PDF

30. Evaluating Atypical Gaze Patterns through Vision Models: The Case of Cortical Visual Impairment

Author

Avramidis, Kleanthis, Chang, Melinda Y., Sharma, Rahul, Borchert, Mark S., and Narayanan, Shrikanth

Subjects
Electrical Engineering and Systems Science - Signal Processing, Electrical Engineering and Systems Science - Image and Video Processing
Abstract

A wide range of neurological and cognitive disorders exhibit distinct behavioral markers aside from their clinical manifestations. Cortical Visual Impairment (CVI) is a prime example of such conditions, resulting from damage to visual pathways in the brain, and adversely impacting low- and high-level visual function. The characteristics impacted by CVI are primarily described qualitatively, challenging the establishment of an objective, evidence-based measure of CVI severity. To study those characteristics, we propose to create visual saliency maps by adequately prompting deep vision models with attributes of clinical interest. After extracting saliency maps for a curated set of stimuli, we evaluate fixation traces on those from children with CVI through eye tracking technology. Our experiments reveal significant gaze markers that verify clinical knowledge and yield nuanced discriminability when compared to those of age-matched control subjects. Using deep learning to unveil atypical visual saliency is an important step toward establishing an eye-tracking signature for severe neurodevelopmental disorders, like CVI., Comment: 5 pages, 4 figures, submitted to IEEE EMBC 2024

Published
2024

31. Proton Pencil-Beam Scanning Stereotactic Body Radiation Therapy and Hypofractionated Radiation Therapy for Thoracic Malignancies: Patterns of Practice Survey and Recommendations for Future Development from NRG Oncology and PTCOG

Author

Liu, Wei, Feng, Hongying, Taylor, Paige A., Kang, Minglei, Shen, Jiajian, Saini, Jatinder, Zhou, Jun, Giap, Huan B., Yu, Nathan Y., Sio, Terence S., Mohindra, Pranshu, Chang, Joe Y., Bradley, Jeffrey D., Xiao, Ying, Simone II, Charles B., and Lin, Liyong

Subjects
Physics - Medical Physics
Abstract

Stereotactic body radiation therapy (SBRT) and hypofractionation using pencil-beam scanning (PBS) proton therapy (PBSPT) is an attractive option for thoracic malignancies. Combining the advantages of target coverage conformity and critical organ sparing from both PBSPT and SBRT, this new delivery technique has great potential to improve the therapeutic ratio, particularly for tumors near critical organs. Safe and effective implementation of PBSPT SBRT/hypofractionation to treat thoracic malignancies is more challenging than the conventionally-fractionated PBSPT due to concerns of amplified uncertainties at the larger dose per fraction. NRG Oncology and Particle Therapy Cooperative Group (PTCOG) Thoracic Subcommittee surveyed US proton centers to identify practice patterns of thoracic PBSPT SBRT/hypofractionation. From these patterns, we present recommendations for future technical development of proton SBRT/hypofractionation for thoracic treatment. Amongst other points, the recommendations highlight the need for volumetric image guidance and multiple CT-based robust optimization and robustness tools to minimize further the impact of uncertainties associated with respiratory motion. Advances in direct motion analysis techniques are urgently needed to supplement current motion management techniques., Comment: 36 pages, 4 figures, 4 tables

Published
2024

32. SocraSynth: Multi-LLM Reasoning with Conditional Statistics

Author

Chang, Edward Y.

Subjects
Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Computer Science - Machine Learning, I.2.7
Abstract

Large language models (LLMs), while promising, face criticisms for biases, hallucinations, and a lack of reasoning capability. This paper introduces SocraSynth, a multi-LLM agent reasoning platform developed to mitigate these issues. SocraSynth utilizes conditional statistics and systematic context enhancement through continuous arguments, alongside adjustable debate contentiousness levels. The platform typically involves a human moderator and two LLM agents representing opposing viewpoints on a given subject. SocraSynth operates in two main phases: knowledge generation and reasoning evaluation. In the knowledge generation phase, the moderator defines the debate topic and contentiousness level, prompting the agents to formulate supporting arguments for their respective stances. The reasoning evaluation phase then employs Socratic reasoning and formal logic principles to appraise the quality of the arguments presented. The dialogue concludes with the moderator adjusting the contentiousness from confrontational to collaborative, gathering final, conciliatory remarks to aid in human reasoning and decision-making. Through case studies in three distinct application domains, this paper showcases SocraSynth's effectiveness in fostering rigorous research, dynamic reasoning, comprehensive assessment, and enhanced collaboration. This underscores the value of multi-agent interactions in leveraging LLMs for advanced knowledge extraction and decision-making support., Comment: 1 figure, 6 tables, 6 appendices

Published
2024

33. Non-Fermi-liquid behavior in a ferromagnetic heavy fermion system CeTi$_{1-x}$V$_{x}$Ge$_{3}$

Author

Lin, R. -Z., Jin, H., Klavins, P., Chen, W. -T., Chang, Y. -Y., Chung, C. -H., Taufour, V., and Huang, C. -L.

Subjects
Condensed Matter - Strongly Correlated Electrons
Abstract

An investigation of the thermodynamic and electrical transport properties of the isoelectronic chemical substitution series CeTi$_{1-x}$V$_{x}$Ge$_{3}$ (CTVG) single crystals is reported. As x increases, the ferromagnetic (FM) transition temperature is suppressed, reaching absolute zero at the critical concentration x = 0.4, where a non-Fermi-liquid low-temperature specific heat and electrical resistivity, as well as the hyperscaling of specific heat and magnetization are found. Our study clearly identifies an FM quantum critical point (QCP) in CTVG. The obtained critical exponents suggest that CTVG falls in the preasymptotic region of the disorder-tuned FM QCP predicted by the Belitz-Kirkpatrick-Vojta theory.

Published
2024
Full Text
View/download PDF

35. Shoulder Bone Segmentation with DeepLab and U-Net

Author

Carl, Michael, Lall, Kaustubh, Pai, Darren, Chang, Eric Y, Statum, Sheronda, Brau, Anja, Chung, Christine B, Fung, Maggie, and Bae, Won C

Subjects
Engineering, Biomedical and Clinical Sciences, Clinical Sciences, Biomedical Engineering, Clinical Research, Bioengineering, Networking and Information Technology R&D (NITRD), Machine Learning and Artificial Intelligence, Biomedical Imaging, Musculoskeletal, DeepLab, MRI, U-Net, ZTE, glenohumeral, glenoid, humeral head, image processing
Abstract

Evaluation of 3D bone morphology of the glenohumeral joint is necessary for pre-surgical planning. Zero echo time (ZTE) magnetic resonance imaging (MRI) provides excellent bone contrast and can potentially be used in place of computed tomography. Segmentation of shoulder anatomy, particularly humeral head and acetabulum, is needed for detailed assessment of each anatomy and for pre-surgical preparation. In this study we compared performance of two popular deep learning models based on Google's DeepLab and U-Net to perform automated segmentation on ZTE MRI of human shoulders. Axial ZTE images of normal shoulders (n=31) acquired at 3-Tesla were annotated for training with a DeepLab and 2D U-Net, and the trained model was validated with testing data (n=13). While both models showed visually satisfactory results for segmenting the humeral bone, U-Net slightly over-estimated while DeepLab under-estimated the segmented area compared to the ground truth. Testing accuracy quantified by Dice score was significantly higher (p

Published
2024

36. Fast dual-echo estimation of apparent long T2 fraction using ultrashort echo time magnetic resonance imaging in tibialis tendons and its osteoporosis-related differences in women

Author

Jerban, Saeed, Moazamian, Dina, Ma, Yajun, Afsahi, Amir Masoud, Dwek, Sophia, Athertya, Jiyo, Malhi, Bhavsimran, Jang, Hyungseok, Woods, Gina, Chung, Christine B, Du, Jiang, and Chang, Eric Y

Subjects
Engineering, Atomic, Molecular and Optical Physics, Physical Sciences, Biomedical Engineering, Osteoporosis, Biomedical Imaging, Clinical Research, Musculoskeletal, Tibialis tendon, tendon quality, osteoporosis, magnetic resonance imaging, ultrashort echo time, Condensed Matter Physics, Optical Physics, Other Physical Sciences, Biomedical engineering, Atomic, molecular and optical physics
Abstract

BackgroundTendon and bone comprise a critical interrelating unit. Bone loss, including that seen with osteopenia (OPe) or osteoporosis (OPo), may be associated with a reduction in tendon quality, though this remains incompletely investigated. Clinical magnetic resonance imaging (MRI) sequences cannot directly detect signals from tendons because of the very short T2. Clinical MRI may detect high-graded abnormalities by changes in the adjacent structures like bone. However, ultrashort echo time MRI (UTE-MRI) can capture high signals from all tendons. To determine if the long T2 fraction, as measured by a dual-echo UTE-MRI sequence, is a sensitive quantitative technique to the age- and bone-loss-related changes of the lower leg tendons.MethodsThis is a cross-sectional study conducted between January 2018 to February 2020 in the lower legs of 14 female patients with OPe [72±6 years old, body mass index (BMI) =25.8±6.2 kg/m2] and 31 female patients with OPo (73±6 years old, BMI=22.0±3.8 kg/m2), as well as 30 female subjects with normal bone (Normal, 35±18 years old, BMI =23.2±4.3 kg/m2), were imaged on a 3T clinical scanner using a dual-echo 3D Cones UTE sequence. We defined the apparent long T2 signal fraction (aFrac-LongT2) of tendons as the ratio between the signal at the second echo time (TE =2.2 ms) to the UTE signal. The average aFrac-LongT2 and the cross-sectional area were calculated for the anterior tibialis tendons (ATTs) and the posterior tibialis tendons (PTTs). The Kruskal-Wallis rank test was used to compare the differences in aFrac-LongT2 and the cross-sectional area of the tendons between the groups.ResultsThe aFrac-LongT2 of the ATTs and PTTs were significantly higher in the OPo group compared with the Normal group (22.2% and 34.8% in the ATT and PTT, respectively, P

Published
2024

40. A human autoimmune organoid model reveals IL-7 function in coeliac disease

Author

Santos, António J. M., van Unen, Vincent, Lin, Zhongqi, Chirieleison, Steven M., Ha, Nhi, Batish, Arpit, Chan, Joshua E., Cedano, Jose, Zhang, Elisa T., Mu, Qinghui, Guh-Siesel, Alexander, Tomaske, Madeline, Colburg, Deana, Varma, Sushama, Choi, Shannon S., Christophersen, Asbjørn, Baghdasaryan, Ani, Yost, Kathryn E., Karlsson, Kasper, Ha, Andrew, Li, Jing, Dai, Hongjie, Sellers, Zachary M., Chang, Howard Y., Dunn, James C. Y., Zhang, Bing M., Mellins, Elizabeth D., Sollid, Ludvig M., Fernandez-Becker, Nielsen Q., Davis, Mark M., and Kuo, Calvin J.

Published
2024
Full Text
View/download PDF

46. The geometry of flow: Advancing predictions of river geometry with multi-model machine learning

Author

Chang, Shuyu Y, Ghahremani, Zahra, Manuel, Laura, Erfani, Mohammad, Shen, Chaopeng, Cohen, Sagy, Van Meter, Kimberly, Pierce, Jennifer L, Meselhe, Ehab A, and Goharian, Erfan

Subjects
Physics - Geophysics, Computer Science - Machine Learning
Abstract

Hydraulic geometry parameters describing river hydrogeomorphic is important for flood forecasting. Although well-established, power-law hydraulic geometry curves have been widely used to understand riverine systems and mapping flooding inundation worldwide for the past 70 years, we have become increasingly aware of the limitations of these approaches. In the present study, we have moved beyond these traditional power-law relationships for river geometry, testing the ability of machine-learning models to provide improved predictions of river width and depth. For this work, we have used an unprecedentedly large river measurement dataset (HYDRoSWOT) as well as a suite of watershed predictor data to develop novel data-driven approaches to better estimate river geometries over the contiguous United States (CONUS). Our Random Forest, XGBoost, and neural network models out-performed the traditional, regionalized power law-based hydraulic geometry equations for both width and depth, providing R-squared values of as high as 0.75 for width and as high as 0.67 for depth, compared with R-squared values of 0.57 for width and 0.18 for depth from the regional hydraulic geometry equations. Our results also show diverse performance outcomes across stream orders and geographical regions for the different machine-learning models, demonstrating the value of using multi-model approaches to maximize the predictability of river geometry. The developed models have been used to create the newly publicly available STREAM-geo dataset, which provides river width, depth, width/depth ratio, and river and stream surface area (%RSSA) for nearly 2.7 million NHDPlus stream reaches across the rivers and streams across the contiguous US., Comment: 30 pages, 10 figures

Published
2023

47. MedEval: A Multi-Level, Multi-Task, and Multi-Domain Medical Benchmark for Language Model Evaluation

Author

He, Zexue, Wang, Yu, Yan, An, Liu, Yao, Chang, Eric Y., Gentili, Amilcare, McAuley, Julian, and Hsu, Chun-Nan

Subjects
Computer Science - Computation and Language
Abstract

Curated datasets for healthcare are often limited due to the need of human annotations from experts. In this paper, we present MedEval, a multi-level, multi-task, and multi-domain medical benchmark to facilitate the development of language models for healthcare. MedEval is comprehensive and consists of data from several healthcare systems and spans 35 human body regions from 8 examination modalities. With 22,779 collected sentences and 21,228 reports, we provide expert annotations at multiple levels, offering a granular potential usage of the data and supporting a wide range of tasks. Moreover, we systematically evaluated 10 generic and domain-specific language models under zero-shot and finetuning settings, from domain-adapted baselines in healthcare to general-purposed state-of-the-art large language models (e.g., ChatGPT). Our evaluations reveal varying effectiveness of the two categories of language models across different tasks, from which we notice the importance of instruction tuning for few-shot usage of large language models. Our investigation paves the way toward benchmarking language models for healthcare and provides valuable insights into the strengths and limitations of adopting large language models in medical domains, informing their practical applications and future advancements., Comment: Accepted to EMNLP 2023. Camera-ready version: updated IRB, added more evaluation results on LLMs such as GPT4, LLaMa2, and LLaMa2-chat

Published
2023

48. Whole knee joint mapping using a phase modulated UTE adiabatic T1ρ (PM‐UTE‐AdiabT1ρ) sequence

Author

Ma, Yajun, Carl, Michael, Tang, Qingbo, Moazamian, Dina, Athertya, Jiyo S, Jang, Hyungseok, Bukata, Susan V, Chung, Christine B, Chang, Eric Y, and Du, Jiang

Subjects
Engineering, Biomedical Engineering, Clinical Research, Bioengineering, 4.2 Evaluation of markers and technologies, Musculoskeletal, Reproducibility of Results, Knee Joint, Anterior Cruciate Ligament, Patellar Ligament, Meniscus, Magnetic Resonance Imaging, AdiabT(1 rho), T-1 rho, UTE, whole knee imaging, AdiabT1ρ, T1ρ, Nuclear Medicine & Medical Imaging, Biomedical engineering
Abstract

PurposeTo develop a 3D phase modulated UTE adiabatic T1ρ (PM-UTE-AdiabT1ρ ) sequence for whole knee joint mapping on a clinical 3 T scanner.MethodsThis new sequence includes six major features: (1) a magnetization reset module, (2) a train of adiabatic full passage pulses for spin locking, (3) a phase modulation scheme (i.e., RF cycling pair), (4) a fat saturation module, (5) a variable flip angle scheme, and (6) a 3D UTE Cones sequence for data acquisition. A simple exponential fitting was used for T1ρ quantification. Phantom studies were performed to investigate PM-UTE-AdiabT1ρ 's sensitivity to compositional changes and reproducibility as well as its correlation with continuous wave-T1ρ measurement. The PM-UTE-AdiabT1ρ technique was then applied to five ex vivo and five in vivo normal knees to measure T1ρ values of femoral cartilage, meniscus, posterior cruciate ligament, anterior cruciate ligament, patellar tendon, and muscle.ResultsThe phantom study demonstrated PM-UTE-AdiabT1ρ 's high sensitivity to compositional changes, its high reproducibility, and its strong linear correlation with continuous wave-T1ρ measurement. The ex vivo and in vivo knee studies demonstrated average T1ρ values of 105.6 ± 8.4 and 77.9 ± 3.9 ms for the femoral cartilage, 39.2 ± 5.1 and 30.1 ± 2.2 ms for the meniscus, 51.6 ± 5.3 and 29.2 ± 2.4 ms for the posterior cruciate ligament, 79.0 ± 9.3 and 52.0 ± 3.1 ms for the anterior cruciate ligament, 19.8 ± 4.5 and 17.0 ± 1.8 ms for the patellar tendon, and 91.1 ± 8.8 and 57.6 ± 2.8 ms for the muscle, respectively.ConclusionThe 3D PM-UTE-AdiabT1ρ sequence allows volumetric T1ρ assessment for both short and long T2 tissues in the knee joint on a clinical 3 T scanner.

Published
2024

49. Rotator cuff muscle fibrosis can be assessed using ultrashort echo time magnetization transfer MRI with fat suppression

Author

Chang, Eric Y, Suprana, Arya, Tang, Qingbo, Cheng, Xin, Fu, Eddie, Orozco, Elisabeth, Jerban, Saeed, Shah, Sameer B, Du, Jiang, and Ma, Yajun

Subjects
Biomedical and Clinical Sciences, Engineering, Biomedical Engineering, Physical Injury - Accidents and Adverse Effects, Musculoskeletal, Animals, Rats, Rotator Cuff, Tendons, Muscular Atrophy, Magnetic Resonance Imaging, Adipose Tissue, fibrosis, magnetization transfer, rotator cuff muscle, ultrashort TE, Medicinal and Biomolecular Chemistry, Clinical Sciences, Nuclear Medicine & Medical Imaging, Clinical sciences, Biomedical engineering
Abstract

Muscle degeneration following rotator cuff tendon tearing is characterized by fatty infiltration and fibrosis. While tools exist for the characterization of fat, the ability to noninvasively assess muscle fibrosis is limited. The purpose of this study was to evaluate the capability of quantitative ultrashort echo time T1 (UTE-T1) and UTE magnetization transfer (UTE-MT) mapping with and without fat suppression (FS) for the differentiation of injured and control rotator cuff muscles and for the detection of fibrosis. A rat model of chronic massive rotator cuff tearing (n = 12) was used with tenotomy of the right supraspinatus and infraspinatus tendons and silicone implants to prevent healing. Imaging was performed on a 3-T scanner, and UTE-T1 mapping with and without FS and UTE-MT with and without FS for macromolecular fraction (MMF) mapping was performed. At 20 weeks postinjury, T1 and MMF were measured in the supraspinatus and infraspinatus muscles of the injured and contralateral, internal control sides. Histology was performed and connective tissue fraction (CTF) was measured, defined as the area of collagen-rich extracellular matrix divided by the total muscle area. Paired t-tests and correlation analyses were performed. Significant differences between injured and control sides were found for CTF in the supraspinatus (mean ± SD, 14.5% ± 3.9% vs. 11.3% ± 3.7%, p = 0.01) and infraspinatus (17.0% ± 5.4% vs. 12.5% ± 4.6%, p

Published
2024

50. Myelin water quantification in multiple sclerosis using short repetition time adiabatic inversion recovery prepared-fast spin echo (STAIR-FSE) imaging

Author

Moazamian, Dina, Mohammadi, Hamidreza Shaterian, Athertya, Jiyo S, Shin, Soo Hyun, Lo, James, Chang, Eric Y, Du, Jiang, Bydder, Graeme M, and Ma, Yajun

Subjects
Engineering, Atomic, Molecular and Optical Physics, Physical Sciences, Biomedical Engineering, Multiple Sclerosis, Neurosciences, Clinical Research, Brain Disorders, Autoimmune Disease, Biomedical Imaging, Neurodegenerative, 4.2 Evaluation of markers and technologies, Neurological, Myelin water, short repetition time adiabatic inversion recovery, fast spin echo, multiple sclerosis, Condensed Matter Physics, Optical Physics, Other Physical Sciences, Biomedical engineering, Atomic, molecular and optical physics
Abstract

BackgroundMyelin water imaging (MWI) is a myelin-specific technique, which has great potential for the assessment of demyelination and remyelination. This study develops a new MWI method, which employs a short repetition time adiabatic inversion recovery (STAIR) technique in combination with a commonly used fast spin echo (FSE) sequence and provides quantification of myelin water (MW) fractions.MethodWhole-brain MWI was performed using the short repetition time adiabatic inversion recovery prepared-fast spin echo (STAIR-FSE) technique on eight healthy volunteers (mean age: 38±14 years, four-males) and seven patients with multiple sclerosis (MS) (mean age: 53.7±8.7 years, two-males) on a 3T clinical magnetic resonance imaging scanner. To facilitate the quantification of apparent myelin water fraction (aMWF), a proton density-weighted FSE was also used during the scans to allow total water imaging. The aMWF measurements of MS lesions and normal-appearing white matter (NAWM) regions in MS patients were compared with those measured in normal white matter (NWM) regions in healthy volunteers. Both the analysis of variance (ANOVA) test and paired comparison were performed for the comparison.ResultsThe MW in the whole-brain was selectively imaged and quantified using the STAIR-FSE technique in all participants. MS lesions showed much lower signal intensities than NAWM in the STAIR-FSE images. ANOVA analysis revealed a significant difference in the aMWF measurements between the three groups. Moreover, the aMWF measurements in MS lesions were significantly lower than those in both NWM of healthy volunteers and NAWM of MS patients. Lower aMWF measurements in NAWM were also found in comparison with those in NWM.ConclusionsThe STAIR-FSE technique is capable of measuring aMWF values for the indirect detection of myelin loss in MS, thus facilitating clinical translation of whole brain MWI and quantification, which show great potential for the detection and evaluation of changes in myelin in the brain of patients with MS for future larger cohort studies.

Published
2024

Catalog

Books, media, physical & digital resources
See catalog results