A5002366529- Radiomics and Machine Learning in Medical Imaging
- Nuclear Engineering Thermal-Hydraulics
- Nuclear reactor physics and engineering
- Medical Imaging Techniques and Applications
- Nuclear Materials and Properties
- Nuclear Physics and Applications
- Quantum, superfluid, helium dynamics
- Risk and Safety Analysis
- Functional Brain Connectivity Studies
- Ultrasound and Cavitation Phenomena
- Adversarial Robustness in Machine Learning
- AI in cancer detection
- Neural Networks and Applications
- COVID-19 Clinical Research Studies
- Ethics and Social Impacts of AI
- Neural dynamics and brain function
- Privacy-Preserving Technologies in Data
- Machine Learning and Algorithms
- Statistical Methods and Inference
- Big Data and Business Intelligence
- Artificial Intelligence in Healthcare and Education
- Geochemistry and Geologic Mapping
- Stochastic Gradient Optimization Techniques
- Atomic and Subatomic Physics Research
- Machine Learning in Bioinformatics
Microsoft (United States)
2020-2025
Microsoft Research (United Kingdom)
2024
Xuzhou Medical College
2024
North China Electric Power University
2024
Indiana University School of Medicine
2024
Indiana University – Purdue University Indianapolis
2024
Lianyungang Oriental Hospital
2024
People's Liberation Army 401 Hospital
2023
University of Washington
2022
Yangzhou University
2022
Abstract The majority of proteins must form higher-order assemblies to perform their biological functions, yet few machine learning models can accurately and rapidly predict the symmetry involving multiple copies same protein chain. Here, we address this gap by finetuning several classes foundation models, homo-oligomer symmetry. Our best model named Seq2Symm, which utilizes ESM2, outperforms existing template-based deep methods achieving an average AUC-PR 0.47, 0.44 0.49 across symmetries...
Automatic and accurate segmentation of lesions in images metastatic castration-resistant prostate cancer has the potential to enable personalized radiopharmaceutical therapy advanced treatment response monitoring. The aim this study is develop a convolutional neural networks-based framework for fully-automated detection whole-body PET/CT images. 525 patients with were available study, acquired [18F]DCFPyL radiotracer that targets prostate-specific membrane antigen (PSMA). U-Net (1)-based...
Leprosy is an infectious disease that mostly affects underserved populations. Although it has been largely eliminated, still about 200'000 new patients are diagnosed annually. In the absence of a diagnostic test, clinical diagnosis often delayed, potentially leading to irreversible neurological damage and its resulting stigma, as well continued transmission. Accelerating could significantly contribute advancing global leprosy elimination. Digital Artificial Intelligence (AI) driven...
A unique, new stand-alone acoustic inertial confinement nuclear fusion test device was successfully tested. Experiments using four different liquid types were conducted in which bubbles self-nucleated without the use of external neutrons. Four independent detection systems used (i.e., a neutron track plastic detector to provide unambiguous visible records for fast neutrons, ${\mathrm{BF}}_{3}$ detector, NE-113-type scintillation and NaI $\ensuremath{\gamma}$ ray detector). Statistically...
Abstract Generative Adversarial Networks (GANs) have made releasing of synthetic images a viable approach to share data without the original dataset. It has been shown that such can be used for variety downstream tasks as training classifiers would otherwise require dataset shared. However, recent work GAN models and their synthetically generated infer set membership by an adversary who access entire some auxiliary information. Current approaches mitigate this problem (such DPGAN [1]) lead...
Acquired aplastic anemia (AA) is caused by autoreactive T cell-mediated destruction of early hematopoietic cells. Somatic loss human leukocyte antigen (HLA) class I alleles was identified as a mechanism immune escape in surviving cells some patients with AA. However, pathogenicity, structural characteristics, and clinical impact specific HLA AA remain poorly understood. Here, we evaluated somatic 505 from 2 multi-institutional cohorts. Using combination mutation frequencies, peptide-binding...
To evaluate the generalizability of artificial intelligence (AI) algorithms that use deep learning methods to identify middle ear disease from otoscopic images, between internal external performance. 1842 images were collected three independent sources: (a) Van, Turkey, (b) Santiago, Chile, and (c) Ohio, USA. Diagnostic categories consisted (i) normal or (ii) abnormal. Deep used develop models performance, using area under curve (AUC) estimates. A pooled assessment was performed by combining...
<title>Abstract</title> The majority of proteins must form higher-order assemblies to perform their biological functions. Despite the importance protein quaternary structure, there are few machine learning models that can accurately and rapidly predict symmetry involving multiple copies same chain. Here, we address this gap by training several classes foundation models, including ESM-MSA, ESM2, RoseTTAFold2, homo-oligomer symmetry. Our best model named Seq2Symm, which utilizes outperforms...
Purpose: This study examines the core traits of image-to-image translation (I2I) networks, focusing on their effectiveness and adaptability in everyday clinical settings. Methods: We have analyzed data from 794 patients diagnosed with prostate cancer (PCa), using ten prominent 2D/3D I2I networks to convert ultrasound (US) images into MRI scans. also introduced a new analysis Radiomic features (RF) via Spearman correlation coefficient explore whether high performance (SSIM>85%) could detect...
Presbycusis is characterized by bilateral sensorineural hearing loss at high frequencies and often accompanied cognitive decline. This study aimed to identify the topological reorganization of brain functional network in presbycusis with/without decline using graph theory analysis approaches based on resting-state magnetic resonance imaging (rs-fMRI).Resting-state fMRI scans were obtained from 30 patients with decline, without 50 age-, sex-, education-matched healthy controls. Graph was...
COVID-19 mortality risk stratification tools could improve care, inform accurate and rapid triage decisions, guide family discussions regarding goals of care. A minority prognostic have been tested in external cohorts. Our objective was to compare machine learning algorithms develop a tool for predicting subsequent clinical outcomes COVID-19. We conducted retrospective cohort study that included hospitalized patients with from March 2020 2021. Seven Hundred Twelve consecutive University...
Generative Adversarial Networks (GANs) have made releasing of synthetic images a viable approach to share data without the original dataset. It has been shown that such can be used for variety downstream tasks as training classifiers would otherwise require dataset shared. However, recent work GAN models and their synthetically generated infer set membership by an adversary who access entire some auxiliary information. Current approaches mitigate this problem (such DPGAN) lead dramatically...
AI for good (AI4G) projects involve developing and applying artificial intelligence (AI) based solutions to further goals in areas such as sustainability, health, humanitarian aid, social justice. Developing deploying must be done collaboration with partners who are experts the domain question already have experience making progress towards goals. Based on our experiences, we detail different aspects of this type broken down into four high-level categories: communication, data, modeling,...
Abstract Prior efforts have manifested that functional connectivity (FC) network disruptions are concerned with cognitive disorder in presbycusis. The present research was designed to investigate the topological reorganization and classification performance of low-order (LOFC) high-order (HOFC) networks patients Resting-state magnetic resonance imaging (Rs-fMRI) data were obtained 60 presbycusis 50 matched healthy control subjects (HCs). LOFC HOFC then constructed, metrics from constructed...
Generative machine learning models are being increasingly viewed as a way to share sensitive data between institutions. While there has been work on developing differentially private generative modeling approaches, these approaches generally lead sub-par sample quality, limiting their use in real world applications. Another line of focused which higher quality samples but currently lack any formal privacy guarantees. In this work, we propose the first framework for membership estimation...