A5036703455- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
- COVID-19 epidemiological studies
- Artificial Intelligence in Healthcare and Education
- Prostate Cancer Diagnosis and Treatment
- Genetics, Bioinformatics, and Biomedical Research
- SARS-CoV-2 and COVID-19 Research
- COVID-19 and Mental Health
- Gene expression and cancer classification
- COVID-19 and healthcare impacts
- Infection Control and Ventilation
- Advanced Image and Video Retrieval Techniques
- Dermatological and Skeletal Disorders
- Disaster Response and Management
- Molecular Biology Techniques and Applications
- Fungal Infections and Studies
- Colorectal Cancer Screening and Detection
- SARS-CoV-2 detection and testing
- Digital Imaging in Medicine
- Neutropenia and Cancer Infections
- Biomedical Text Mining and Ontologies
- Clinical Laboratory Practices and Quality Control
- Oral Health Pathology and Treatment
- Digital Imaging for Blood Diseases
- Surgical Simulation and Training
United States Department of the Navy
2020-2023
Uniformed Services University of the Health Sciences
2020-2023
Defense Systems (United States)
2021-2022
United States Navy
2021
University of Guam
2021
Unifor
2021
Naval Medical Center San Diego
2018-2020
Naval Medical Research Command
2020
Navy and Marine Corps Force Health Protection Command
2020
Bellingham Technical College
2018
For prostate cancer patients, the Gleason score is one of most important prognostic factors, potentially determining treatment independent stage. However, scoring based on subjective microscopic examination tumor morphology and suffers from poor reproducibility. Here we present a deep learning system (DLS) for whole-slide images prostatectomies. Our was developed using 112 million pathologist-annotated image patches 1226 slides, evaluated an validation dataset 331 slides. Compared to...
Context.— Nodal metastasis of a primary tumor influences therapy decisions for variety cancers. Histologic identification cells in lymph nodes can be laborious and error-prone, especially small foci. Objective.— To evaluate the application clinical implementation state-of-the-art deep learning–based artificial intelligence algorithm (LYmph Node Assistant or LYNA) detection metastatic breast cancer sentinel node biopsies. Design.— Whole slide images were obtained from...
<h3>Importance</h3> For prostate cancer, Gleason grading of the biopsy specimen plays a pivotal role in determining case management. However, is associated with substantial interobserver variability, resulting need for decision support tools to improve reproducibility routine clinical practice. <h3>Objective</h3> To evaluate ability deep learning system (DLS) grade diagnostic specimens. <h3>Design, Setting, and Participants</h3> The DLS was evaluated using 752 deidentified digitized images...
An outbreak of coronavirus disease 2019 (Covid-19) occurred on the U.S.S. Theodore Roosevelt, a nuclear-powered aircraft carrier with crew 4779 personnel.We obtained clinical and demographic data for all members, including results testing by real-time reverse-transcriptase polymerase chain reaction (rRT-PCR). All members were followed up minimum 10 weeks, regardless test or absence symptoms.The was predominantly young (mean age, 27 years) in general good health, meeting U.S. Navy standards...
Breast cancer management depends on biomarkers including estrogen receptor, progesterone and human epidermal growth factor receptor 2 (ER/PR/HER2). Though existing scoring systems are widely used well-validated, they can involve costly preparation variable interpretation. Additionally, discordances between histology expected biomarker findings prompt repeat testing to address biological, interpretative, or technical reasons for unexpected results.We developed three independent deep learning...
Recent developments in machine learning have stimulated intense interest software that may augment or replace human experts. Machine impact pathology practice by offering new capabilities analysis, interpretation, and outcomes prediction using images other data. The principles of operation management systems are unfamiliar to pathologists, who anticipate a need for additional education be effective as expert users managers the tools.To provide background on practicing including an overview...
Abstract Infectious threats, like the COVID-19 pandemic, hinder maintenance of a productive and healthy workforce. If subtle physiological changes precede overt illness, then proactive isolation testing can reduce labor force impacts. This study hypothesized that an early infection warning service based on wearable monitoring predictive models created with machine learning could be developed deployed. We prototype tool, first deployed June 23, 2020, delivered continuously updated scores risk...
Histologic grading of breast cancer involves review and scoring three well-established morphologic features: mitotic count, nuclear pleomorphism, tubule formation. Taken together, these features form the basis Nottingham Grading System which is used to inform characterization prognosis. In this study, we develop deep learning models perform histologic all components using digitized hematoxylin eosin-stained slides containing invasive carcinoma. We first evaluate model performance...
ABSTRACT We describe a patient with subclinical coccidioidomycosis who experienced rapid disease dissemination shortly after SARS-CoV-2 infection, suggesting host immune response dysregulation to by SARS-CoV-2. hypothesize that disrupted cell-mediated signaling may result infection leading functional exhaustion and CD8+ T-cell senescence impairment in cellular Coccidioides infection.
Microscopic interpretation of histopathology images underlies many important diagnostic and treatment decisions. While advances in vision-language modeling raise new opportunities for analysis such images, the gigapixel-scale size whole slide (WSIs) introduces unique challenges. Additionally, pathology reports simultaneously highlight key findings from small regions while also aggregating across multiple slides, often making it difficult to create robust image-text pairs. As such, remain a...
Abstract Introduction Extreme environmental conditions (e.g., natural light conditions, time zone transitions), and specific operational ship motion due to ice breaking) combined with a variety of other occupational stressors (including shift work, psychological stress, performing physically strenuous activities), poses unique challenges when conducting polar operations. The overarching aim this project is assess the sailors experience operating in regions order identify performance-limiting...
Artificial Intelligence (AI) for decision support and diagnosis in pathology could provide immense value to society, improving patient outcomes alleviating workload demands on pathologists. However, this potential cannot be realized until sufficient methods testing evaluation of such AI systems are developed adopted. We present a novel metric multi-class classification algorithms pathology, Error Severity Index (ESI), address the needs pathologists lab managers evaluating systems.
Abstract Gene expression profiling (GEP) provides valuable information for the care of breast cancer patients. However, test itself is expensive and can take a long time to process. In contrast, microscopic examination hematoxylin eosin (H&E) stained tissue inexpensive, fast, integrated into standard care. This work explores possibility predicting ESR1 gene from H&E images, its use in clinical variables patient outcomes. We utilized weakly supervised method train deep learning model...
Several machine learning algorithms have demonstrated high predictive capability in the identification of cancer within digitized pathology slides. The Augmented Reality Microscope (ARM) has allowed these to be seamlessly integrated workflow by overlaying their inferences onto its microscopic field view real time. We present an independent assessment LYmph Node Assistant (LYNA) models, state-of-the-art for breast metastases lymph node biopsies, optimized usage on ARM. assessed models 40...
Abstract Determining when individuals should be released from quarantine is critical for successfully managing a COVID-19 outbreak and local protocols frequently call testing during the period, generally after reasonable incubation which raises question about interpretation of test results period. We report negative predictive value SARS-CoV-2 qPCR tests based on retrospective longitudinal analysis 5349 collected 1227 US service members infected with aboard USS Theodore Roosevelt (CVN-71)...
Machine learning methods are being widely used in medicine to aid cancer diagnosis and detection. In the area of digital pathology, prediction heat maps produced by convolutional neural networks (CNN) have already exceeded performance a trained pathologist with no time constraints. To train deep networks, large datasets accurately labeled ground truth data required; however, whole slide images often on scale 10+ gigapixels when digitized at 40X magnification, contain multiple magnification...
Abstract Infectious threats, like the COVID-19 pandemic, hinder maintenance of a productive and healthy workforce. If subtle physiological changes precede overt illness, then proactive isolation testing can reduce labor force impacts. This study hypothesized that an early infection warning service based on wearable monitoring real-time machine learning could be developed deployed. We prototype tool, first deployed June 23, 2020, delivered continuously updated scores risk for SARS-CoV-2. Data...
Incidental findings of adrenal masses are increasingly found with the use thoracic and abdominal imaging studies. It is standard practice to determine if these hormonally active or nonfunctional, malignant benign.
Abstract Gene expression profiling (GEP) represents an important approach to inform breast cancer treatment. However, access GEP involves challenges associated with cost, tissue transportation, and turn around time. In this work, we explore the prediction of estrogen receptor gene (ESR1) directly from images hematoxylin eosin (H&E) stained, formalin-fixed paraffin-embedded (FFPE) tissue. Since H&E staining is a fast inexpensive component standard preparation in pathology, preserving...