Mona G. Flores
A5031887767- Artificial Intelligence in Healthcare and Education
- COVID-19 diagnosis using AI
- Machine Learning in Healthcare
- Privacy-Preserving Technologies in Data
- AI in cancer detection
- Topic Modeling
- Radiomics and Machine Learning in Medical Imaging
- Digital Radiography and Breast Imaging
- Natural Language Processing Techniques
- Adversarial Robustness in Machine Learning
- Advanced X-ray and CT Imaging
- Lung Cancer Diagnosis and Treatment
- Reproductive System and Pregnancy
- Immune Cell Function and Interaction
- Mesenchymal stem cell research
- Polyomavirus and related diseases
- Cryptography and Data Security
- Cytokine Signaling Pathways and Interactions
- Global Cancer Incidence and Screening
- COVID-19 Clinical Research Studies
- Ethics in Clinical Research
- Atherosclerosis and Cardiovascular Diseases
- Mast cells and histamine
- Immunotherapy and Immune Responses
- Extracellular vesicles in disease
Nvidia (United States)
2020-2025
Santa Clara University
2021
University of California, Davis
2017
Stanford University
2004-2006
Chest CT is emerging as a valuable diagnostic tool for clinical management of COVID-19 associated lung disease. Artificial intelligence (AI) has the potential to aid in rapid evaluation scans differentiation findings from other entities. Here we show that series deep learning algorithms, trained diverse multinational cohort 1280 patients localize parietal pleura/lung parenchyma followed by classification pneumonia, can achieve up 90.8% accuracy, with 84% sensitivity and 93% specificity,...
Federated learning (FL) is a method used for training artificial intelligence models with data from multiple sources while maintaining anonymity, thus removing many barriers to sharing. Here we 20 institutes across the globe train FL model, called EXAM (electronic medical record (EMR) chest X-ray AI model), that predicts future oxygen requirements of symptomatic patients COVID-19 using inputs vital signs, laboratory and X-rays. achieved an average area under curve (AUC) >0.92 predicting...
Abstract There is an increasing interest in developing artificial intelligence (AI) systems to process and interpret electronic health records (EHRs). Natural language processing (NLP) powered by pretrained models the key technology for medical AI utilizing clinical narratives. However, there are few models, largest of which trained domain comparatively small at 110 million parameters (compared with billions general domain). It not clear how large can help utilize unstructured EHRs. In this...
Artificial Intelligence (AI) is having a tremendous impact across most areas of science. Applications AI in healthcare have the potential to improve our ability detect, diagnose, prognose, and intervene on human disease. For models be used clinically, they need made safe, reproducible robust, underlying software framework must aware particularities (e.g. geometry, physiology, physics) medical data being processed. This work introduces MONAI, freely available, community-supported,...
Abstract Physicians make critical time-constrained decisions every day. Clinical predictive models can help physicians and administrators by forecasting clinical operational events. Existing structured data-based have limited use in everyday practice owing to complexity data processing, as well model development deployment 1–3 . Here we show that unstructured notes from the electronic health record enable training of language models, which be used all-purpose engines with low-resistance...
Abstract Objective To demonstrate enabling multi-institutional training without centralizing or sharing the underlying physical data via federated learning (FL). Materials and Methods Deep models were trained at each participating institution using local clinical data, an additional model was FL across all of institutions. Results We found that exhibited superior performance generalizability to single institutions, with overall level significantly better than any institutional alone when...
Abstract There are enormous enthusiasm and concerns in applying large language models (LLMs) to healthcare. Yet current assumptions based on general-purpose LLMs such as ChatGPT, which not developed for medical use. This study develops a generative clinical LLM, GatorTronGPT, using 277 billion words of text including (1) 82 from 126 departments approximately 2 million patients at the University Florida Health (2) 195 diverse general English text. We train GatorTronGPT GPT-3 architecture with...
Federated learning (FL) allows the collaborative training of AI models without needing to share raw data. This capability makes it especially interesting for healthcare applications where patient and data privacy is utmost concern. However, recent works on inversion deep neural networks from model gradients raised concerns about security FL in preventing leakage In this work, we show that these attacks presented literature are impractical use-cases clients' involves updating Batch...
Federated learning (FL) enables building robust and generalizable AI models by leveraging diverse datasets from multiple collaborators without centralizing the data. We created NVIDIA FLARE as an open-source software development kit (SDK) to make it easier for data scientists use FL in their research real-world applications. The SDK includes solutions state-of-the-art algorithms federated machine approaches, which facilitate workflows distributed across enterprises enable platform developers...
There is an increasing interest in developing artificial intelligence (AI) systems to process and interpret electronic health records (EHRs). Natural language processing (NLP) powered by pretrained models the key technology for medical AI utilizing clinical narratives. However, there are few models, largest of which trained domain comparatively small at 110 million parameters (compared with billions general domain). It not clear how large can help utilize unstructured EHRs. In this study, we...
The lack of annotated datasets is a major bottleneck for training new task-specific supervised machine learning models, considering that manual annotation extremely expensive and time-consuming. To address this problem, we present MONAI Label, free open-source framework facilitates the development applications based on artificial intelligence (AI) models aim at reducing time required to annotate radiology datasets. Through researchers can develop AI focusing their domain expertise. It allows...
Abstract Objective To solve major clinical natural language processing (NLP) tasks using a unified text-to-text learning architecture based on generative large model (LLM) via prompt tuning. Methods We formulated 7 key NLP as and solved them one LLM, GatorTronGPT, developed GPT-3 trained with up to 20 billion parameters. adopted soft prompts (ie, trainable vectors) frozen where the LLM parameters were not updated frozen) only vectors of updated, known added additional prefix input layer,...
Today's challenges around global healthcare emphasize the need for large-scale collaborations between clinical and sciesntific communities. However, regulatory constraints data sharing patient privacy might hinder access to genuinely representing clinically relevant populations. We have developed an open-source federated learning framework, NVIDIA FLARE, work such restrictions while maintaining using modern cryptographic information-theoretic methods as homomorphic encryption differential...
Background. Janus kinase 3 (JAK3) mediates signal transduction from cytokine receptors using the common chain (γc). Because mutations in genes encoding γc or JAK3 result immunodeficiency, we investigated potential of a rationally designed inhibitor JAK3, CP-690,550, to prevent renal allograft rejection nonhuman primates. Methods. Life-supporting kidney transplantations were performed between mixed leukocyte reaction-mismatched, ABO blood group-matched cynomolgus monkeys. Animals treated with...
Abstract ‘Federated Learning’ (FL) is a method to train Artificial Intelligence (AI) models with data from multiple sources while maintaining anonymity of the thus removing many barriers sharing. During SARS-COV-2 pandemic, 20 institutes collaborated on healthcare FL study predict future oxygen requirements infected patients using inputs vital signs, laboratory data, and chest x-rays, constituting “EXAM” (EMR CXR AI Model) model. EXAM achieved an average Area Under Curve (AUC) over 0.92,...
Background. Janus Kinase (JAK) 3 is a tyrosine kinase essential for proper signal transduction downstream of selected cytokine receptors and robust T-cell natural killer cells activation function. JAK3 inhibition with CP-690,550 prevents acute allograft rejection. To provide further insight into the mechanisms efficacy, we investigated immunomodulatory effects in vitro vivo nonhuman primates. Methods. Pharmacodynamic assessments lymphocyte activation, function, proliferation phenotype were...
Immunosuppression via Janus kinase (JAK) 3 inhibition affords significant prolongation of allograft survival. We investigated the effects an immunosuppressive regimen combining JAK3 inhibitor CP-690,550 with mycophenolate mofetil (MMF) in nonhuman primates (NHPs).Life-supporting kidney transplantations were performed between ABO-compatible, MLR-mismatched NHPs. Animals treated orally twice a day and MMF (n=8) or alone (n=2) euthanized at 90 earlier due to rejection.Mean survival time...
<title>Abstract</title> Artificial intelligence (AI) methods for the automatic detection and quantification of COVID-19 lesions in chest computed tomography (CT) might play an important role monitoring management disease. We organized international challenge competition development comparison AI algorithms this task, which we supported with public data state-of-the-art benchmark methods. Board Certified Radiologists annotated 295 images from two sources (A B) training (n=199, source A),...
Summary: We are building the world's first Virtual Child–a computer model of normal and cancerous human development at level each individual cell. The Child will “develop cancer” that we subject to unlimited virtual clinical trials pinpoint, predict, prioritize potential new treatments, bringing forward day when no child dies cancer, giving one opportunity lead a full healthy life.
ABSTRACT Objective To develop a large pretrained clinical language model from scratch using transformer architecture; systematically examine how models of different sizes could help 5 natural processing (NLP) tasks at linguistic levels. Methods We created corpus with >90 billion words narratives (>82 words), scientific literature (6 and general English text (2.5 words). developed GatorTron the BERT architecture including 345 million, 3.9 billion, 8.9 parameters, compared three existing...
Abstract In recent years, the role of Artificial Intelligence (AI) in medical imaging has become increasingly prominent, with majority AI applications approved by FDA being and radiology 2023. The surge model development to tackle clinical challenges underscores necessity for preparing high-quality data. Proper data preparation is crucial as it fosters creation standardized reproducible models while minimizing biases. Data curation transforms raw into a valuable, organized, dependable...