Artit Jirapatnakul
A5003735134- Lung Cancer Diagnosis and Treatment
- Radiomics and Machine Learning in Medical Imaging
- Lung Cancer Treatments and Mutations
- Medical Imaging Techniques and Applications
- Advanced X-ray and CT Imaging
- Colorectal Cancer Screening and Detection
- COVID-19 diagnosis using AI
- Medical Imaging and Pathology Studies
- Head and Neck Cancer Studies
- Radiation Dose and Imaging
- Cardiac Imaging and Diagnostics
- Liver Disease Diagnosis and Treatment
- Pleural and Pulmonary Diseases
- Injury Epidemiology and Prevention
- Infective Endocarditis Diagnosis and Management
- Chronic Obstructive Pulmonary Disease (COPD) Research
- Advanced Radiotherapy Techniques
- Occupational Health and Performance
- Air Quality and Health Impacts
- AI in cancer detection
- Artificial Intelligence in Healthcare
- Cardiovascular Health and Disease Prevention
- Biomarkers in Disease Mechanisms
- Aortic Thrombus and Embolism
- Soil Moisture and Remote Sensing
Icahn School of Medicine at Mount Sinai
2015-2024
Clinical Trial Investigators
2024
Mount Sinai Hospital
2023
Mount Sinai Health System
2021
Phoenix VA Health Care System
2021
Mount Sinai Hospital
2019
Mount Sinai Medical Center
2016
Cornell University
2007-2013
Purpose To update information regarding the usefulness of computer-aided detection (CAD) systems with a focus on most critical category, that missed cancers at earlier imaging, for manifest as solid nodule. Materials and Methods By using HIPAA-compliant institutional review board–approved protocol where informed consent was obtained, 50 lung manifested nodule computed tomographic (CT) scans in annual rounds screening (time 1) were retrospectively identified could, retrospect, be previous CT...
Background The low-dose CT (≤3 mGy) screening report of 1000 Early Lung Cancer Action Program (ELCAP) participants in 1999 led to the International ELCAP (I-ELCAP) collaboration, which enrolled 31 567 annual between 1992 and 2005. In 2006, I-ELCAP investigators reported 10-year lung cancer-specific survival 80% for 484 diagnosed with a first primary cancer through screening, high frequency clinical stage I (85%). Purpose To update cure rate by determining 20-year expanded cohort. Materials...
Background Pulmonary noncalcified nodules (NCNs) attached to the fissural or costal pleura with smooth margins and triangular lentiform, oval, semicircular (LOS) shapes at low-dose CT are recommended for annual follow-up instead of immediate workup. Purpose To determine whether management mediastinal diaphragmatic pleura-attached NCNs (M/DP-NCNs) same features as can follow recommendations. Materials Methods This retrospective study reviewed chest examinations in participants from two...
Purpose To validate the recommendation of performing annual follow-up nonsolid nodules (NSNs) identified by computed tomographic (CT) screening for lung cancer, all cases cancer manifesting as NSN in National Lung Screening Trial (NLST) were reviewed. Materials and Methods Institutional review board informed consent waived this study. The NLST database was searched to identify participants with at least one on CT scan cause death (COD) documented endpoint verification process. Among 26 722...
The prevalence and aetiology of liver fibrosis vary over time impact racial/ethnic groups unevenly. This study measured trends identified factors associated with advanced in the United States.Standardised methods were used to analyse data on 47,422 participants (≥20 years old) National Health Nutrition Examination Survey (1999-2018). Advanced was defined as Fibrosis-4 ≥2.67 and/or Forns index ≥6.9 elevated alanine aminotransferase.The estimated number people increased from 1.3 million (95%...
A three-dimensional (3-D) convolutional neural network (CNN) trained from scratch is presented for the classification of pulmonary nodule malignancy low-dose chest CT scans. Recent approval lung cancer screening in United States provides motivation determining likelihood nodules initial scan finding to minimize number follow-up actions. Classifier ensembles different combinations 3-D CNN and traditional machine learning models based on handcrafted image features are also explored. The...
Lung Cancers Manifesting as Part-Solid Nodules in the National Screening TrialRowena Yip1, Claudia I. Henschke1, Dong Ming Xu1, Kunwei Li1,2, Artit Jirapatnakul1 and David F. Yankelevitz1Audio Available | Share
We learned many unanticipated and valuable lessons since we started planning our study of low-dose computed tomography (CT) screening for lung cancer in 1991. The publication the baseline results Early Lung Cancer Action Project (ELCAP) Lancet 1999 showed that CT could identify a high proportion early, curable cancers. This stimulated large national studies to be quickly started. ELCAP design, which provided evidence about context clinical program, was able rapidly expand 12-institution New...
The Public Lung Database to address drug response (PLD) has been developed support research in computer aided diagnosis (CAD). Originally established for applications involving the characterization of pulmonary nodules, PLD augmented provide initial datasets CAD other diseases. In general, best performance a system is achieved when it trained with large amount well documented data. Such training databases are very expensive create and their lack general availability limits targets that can...
IntroductionTo maximize the benefits of computed tomographic screening for lung cancer, optimal treatment small, early cancers is needed. Limiting extent surgery spares tissue, preserves pulmonary function, and decreases operative time, complications, morbidities. It also increases likelihood resecting future new primary cancers. The goal to assess alternative treatments in a timely manner.MethodsThe focus sessions with patients physicians separately highlighted need consider their...
In addition to lung cancer, other thoracic abnormalities, such as emphysema, can be visualized within low-dose CT scans that were initially obtained in cancer screening programs, and thus, opportunistic evaluation of these diseases may highly valuable. However, manual assessment for each scan is tedious often subjective, thus we have developed an automatic, rapid computer-aided diagnosis system emphysema using attention-based multiple instance deep learning 865 LDCTs. the task determining if...
Background: Measurements are not exact, so that if a measurement is repeated, one would get different value each time. The spread of these values the uncertainty. Understanding uncertainty pulmonary nodules important for proper interpretation size and growth measurements. Larger amounts may require longer follow-up intervals to be confident any observed due actual rather than We examined influence nodule features software algorithm on small, solid nodules. Methods: Volumes 107 were measured...
To measure the width of zone transition (ZOT) between nonaerated solid tumor and surrounding nonneoplastic lung parenchyma determine extent to which ZOT influences computer-derived estimates volume based on computed tomographic (CT) images.
A calibration phantom-based method has been developed for predicting small lung nodule volume measurement bias and precision that is specific to a particular computed tomography (CT) scanner acquisition protocol.The approach involves CT scanning simple reference object with protocol, analyzing the scan estimate fundamental imaging properties of system, generating numerous simulated images target geometry using properties, measuring standard segmentation algorithm, calculating performance...
Computed tomography (CT) is a non-invasive imaging modality used to monitor human lung cancers. Typically, tumor volumes are calculated using manual or semi-automated methods that require substantial user input, and an exponential growth model predict growth. However, these measurement methodologies time-consuming can lack consistency. In addition, the availability of datasets with sequential images same needed characterize in vivo patterns for cancers limited due treatment interventions...
Estimation of nodule location and size is an important pre-processing step in some segmentation algorithms to determine the region interest. Ideally, such estimation methods will consistently find same irregardless where seed point (provided either manually or by a detection algorithm) placed relative ldquotruerdquo center nodule, should be reasonable estimate true size. We developed method that estimates using multi-scale Laplacian Gaussian (LoG) filtering. Nodule candidates near given are...