A5103250520- Interstitial Lung Diseases and Idiopathic Pulmonary Fibrosis
- Lung Cancer Diagnosis and Treatment
- COVID-19 diagnosis using AI
- Statistical Methods in Clinical Trials
- Cancer Genomics and Diagnostics
- AI in cancer detection
- Inhalation and Respiratory Drug Delivery
- Gene Regulatory Network Analysis
- Pneumonia and Respiratory Infections
- Advanced Causal Inference Techniques
- Emergency and Acute Care Studies
- Artificial Intelligence in Healthcare and Education
- Health Systems, Economic Evaluations, Quality of Life
- Machine Learning in Healthcare
- Dysphagia Assessment and Management
- Systemic Sclerosis and Related Diseases
- Computational Drug Discovery Methods
- Occupational and environmental lung diseases
- Explainable Artificial Intelligence (XAI)
- Biosimilars and Bioanalytical Methods
University of California, Los Angeles
2020-2025
UCLA Health
2020
To translate artificial intelligence (AI) algorithms into clinical practice requires generalizability of models to real-world data. One the main obstacles is data shift, a distribution mismatch between model training and real environments. Explainable AI techniques offer tools detect mitigate shift problem develop reliable for practice. Most medical trained with datasets gathered from limited environments, such as restricted disease populations center-dependent acquisition conditions. The...
This study determined hospitalization rates of elderly Americans for pneumonia from 1991 through 1998.Epidemiologic data were described 273,143 hospitalizations.Annual hospitalizations aspiration increased by 93.5%. Pneumonia steeply with age, especially among men. Black men at highest risk aspiration, unspecified, Klebsiella, "other gram-negative," and staphylococcal pneumonia; White had the Haemophilus pneumococcal rates. Among women, Blacks predominated in Klebsiella Whites...
Deep learning (DL)-based systems have not yet been broadly implemented in clinical practice, part due to unknown robustness across multiple imaging protocols. To this end, we aim evaluate the performance of several previously developed DL-based models, which were trained distinguish idiopathic pulmonary fibrosis (IPF) from non-IPF among interstitial lung disease (ILD) patients, under standardized reference CT In study, utilized scans ILD subjects, acquired using various protocols, assess...
Idiopathic pulmonary fibrosis (IPF) is a progressive, irreversible, and usually fatal lung disease of unknown reasons, generally affecting the elderly population. Early diagnosis IPF crucial for triaging patients' treatment planning into anti-fibrotic or treatments other causes fibrosis. However, current workflow complicated time-consuming, which involves collaborative efforts from radiologists, pathologists, clinicians it largely subject to inter-observer variability.
Domain knowledge (DK) acquired from prior studies is important for medical diagnosis. This paper leverages the population-level DK using an optimality design criterion to train a deep learning model in end-to-end manner. In this study, problem of interest at patient level diagnose subject with idiopathic pulmonary fibrosis (IPF) among subjects interstitial lung disease (ILD) computed tomography (CT). IPF diagnosis complicated process multidisciplinary discussion experts and interobserver...
We propose a Multi-scale, domain knowledge-Guided Attention model (MGA-Net) for weakly supervised problem-disease diagnosis with only coarse scan-level labels. The use of guided attention models encourages the deep learning-based to focus on area interests (in our case, lung parenchyma), at different resolutions, in an end-to-end manner. research interest is diagnose subjects idiopathic pulmonary fibrosis (IPF) among interstitial disease (ILD) using axial chest high resolution computed...
Two major challenges hinder the deployment of deep learning-based systems into clinical practice: need for numerous high-quality well-labeled data and lack explainability. Attention models, originated from natural language processing, have been popular to address label scarcity problem encourage model In this work, we developed a domain knowledge-guided attention disease diagnosis with only coarse scan-level labels population-level knowledge. The use guided models encourages focus on area...
In immuno-oncology, developing combination therapies to overcome resistance single agent or induce synergistic effects has become a new focus. To accelerate the screening process identify promising combinations based on objective response rates, we propose Bayesian adaptive Umbrella Trial design simultaneously evaluate of an investigational compound with different backbones, where information borrowing across is allowed increase trial efficiency. A robust approach developed strike balance...
Domain knowledge acquired from pilot studies is important for medical diagnosis. This paper leverages the population-level domain based on D-optimal design criterion to judiciously select CT slices that are meaningful disease diagnosis task. As an illustrative example, of idiopathic pulmonary fibrosis (IPF) among interstitial lung (ILD) patients used this work. IPF complicated and subject inter-observer variability. We aim construct a time/memory-efficient model using high resolution...
RATIONALE Distinguishing idiopathic pulmonary fibrosis (IPF) among non-IPF interstitial lung disease (ILD) subjects is an important, but challenging task.Making a correct from single HRCT study will be helpful to identify patients for early anti-fibrotic treatment.Our aim develop automated deep learning-based tool based on chest scans using both 2D and 3D convolutional neural networks (CNN) models.METHODS From image database 1089 were collected, including 389 IPF, 81 myositis, 170 systemic...