A5028670837- Lung Cancer Diagnosis and Treatment
- Radiomics and Machine Learning in Medical Imaging
- Medical Imaging Techniques and Applications
- Advanced X-ray and CT Imaging
- Radiation Dose and Imaging
- Chronic Obstructive Pulmonary Disease (COPD) Research
- Lung Cancer Treatments and Mutations
- Radiology practices and education
- AI in cancer detection
- MRI in cancer diagnosis
- Pleural and Pulmonary Diseases
- Respiratory Support and Mechanisms
- Cardiac Imaging and Diagnostics
- Medical Image Segmentation Techniques
- Advanced Radiotherapy Techniques
- Atomic and Subatomic Physics Research
- Medical Imaging and Pathology Studies
- Advanced Image and Video Retrieval Techniques
- Computer Graphics and Visualization Techniques
- Esophageal Cancer Research and Treatment
- Scientific Measurement and Uncertainty Evaluation
- COVID-19 diagnosis using AI
- Dental Radiography and Imaging
Philips (United States)
2008-2021
IPS Research (United States)
2016-2021
University of Maryland, Baltimore
2016-2021
Montefiore Medical Center
2019
Albert Einstein College of Medicine
2019
Highland Community College - Illinois
2007-2018
Philips (Finland)
2006-2011
Philips (Israel)
2010
Purpose: The development of computer‐aided diagnostic (CAD) methods for lung nodule detection, classification, and quantitative assessment can be facilitated through a well‐characterized repository computed tomography (CT) scans. Lung Image Database Consortium (LIDC) Resource Initiative (IDRI) completed such database, establishing publicly available reference the medical imaging research community. Initiated by National Cancer Institute (NCI), further advanced Foundation Institutes Health...
Background The American College of Radiology updated Lung Imaging Reporting and Data System (Lung-RADS) version 1.0 to 1.1 in May 2019, with the two key changes involving perifissural nodules (PFNs) ground-glass (GGNs) now designated as a negative screening result. This study examines effects these using National Screening Trial (NLST) data. Purpose To determine frequency PFNs GGNs reclassified from category 3 or 4A more benign 2 Lung-RADS 1.1, compared 1.0, CT scans NLST. Materials Methods...
Purpose To compare the Vancouver risk calculator (VRC) with American College of Radiology (ACR) Lung Imaging Reporting and Data System (Lung-RADS) in predicting malignancy National Screening Trial (NLST). Materials Methods A total 2813 patients 4408 nodules (4078 solid, 330 subsolid) were available from NLST for evaluation. Nodules scored by using VRC nine parameters (output was percentage likelihood malignancy; threshold set as greater than 5%) Lung-RADS category 2-4B; defined 4A or 4B;...
Purpose To assess the likelihood of malignancy among a subset nodules in National Lung Screening Trial (NLST) by using risk calculator based on nodule and patient characteristics. Materials Methods All authors received approval for use NLST data. An institutional review board exemption waiver informed consent were granted to author with an academic appointment. Nodule characteristics attributes regard benign malignant applied from group Vancouver, Canada. Patient populations their compared...
We have analyzed 3000 nodule delineations and malignancy ratings of pulmonary nodules made by expert observers in the IDRI CT lung image database. The agreement between volume from automatic segmentation is almost as high inter-observer agreement. For experts' rating quite modest. Linear support vector regression models been tested to emulate a small number automatically computed numerical features. Machine-observer evaluated using linear correlation weighted kappa coefficient. results...
The purpose of this study was to evaluate the use a computed tomographic lung nodule computer-aided detection (CAD) software as second reader for radiology residents.The involved 110 cases from 4 sites. Three expert radiologists identified nodules that were 30 mm in maximum diameter form ground truth. These then interpreted by 6 board-certified and residents. residents read each case without with CAD (Lung Nodule Assesment, Extended Brilliance Workspace; Philips Healthcare, Highlands...
Computer aided characterization aims to support the differential diagnosis of indeterminate pulmonary nodules. A number published studies have correlated automatically computed features from image processing with clinical diagnoses malignancy vs. benignity. Often, however, a high was trained on relatively small diagnosed nodules, raising certain skepticism as how salient and numerically robust various really are. On way towards computer which is trusted in practice, credibility individual...
A real-time matching algorithm for follow-up chest CT scans can significantly reduce the workload on radiologists by automatically finding corresponding location in first or second scan, respectively. The objective of this study was to assess accuracy a fast and versatile single-point registration thoracic scans. The is based automatic lung segmentations both scans, individually left right lung. Whenever user clicks an arbitrary structure lung, coarse position point other scan...
A volume-rendering (VR) technique known as Hesse rendering applies image-enhancement filters to three-dimensional imaging volumes and depicts the filter responses in a color-coded fashion. Unlike direct VR, which makes use of intensities, operates on basis shape properties, such that nodular structures resulting renderings have different colors than do tubular thus are easily visualized. The mouse-click sensitive can be used navigate locations possible anomalies original images. is meant...
Computer aided quantification of emphysema in high resolution CT data is based on identifying low attenuation areas below clinically determined Hounsfield thresholds. However, the prone to error since a gravity effect can influence mean healthy lung parenchyma up ± 50 HU between ventral and dorsal areas. Comparing ultra-low-dose (7 mAs) standard-dose (70 scans each patient we show that measurement ventrodorsal specific but reproducible. It be measured corrected an unsupervised way using...
Presence of emphysema is recognized to be one the single most significant risk factors in models for prediction lung cancer. Therefore, an automatically computed score would a prime candidate as additional numerical feature computer aided diagnosis (CADx) indeterminate pulmonary nodules. We have applied several histogram-based scores 460 thoracic CT scans from IDRI image database, and analyzed conjunction with 3000 nodule malignancy ratings 1232 nodules made by expert observers. Despite...
We present a simple rendering scheme for thoracic CT datasets which yields color coding based on local differential geometry features rather than Hounsfield densities. The curvatures are computed several resolution scales and mapped onto different colors, thereby enhancing nodular tubular structures. can be used as navigation device to quickly access points of possible chest anomalies, in particular lung nodules lymph nodes. underlying principle is use the nodule overview alternative...
Automated segmentation of lung lobes in thoracic CT images has relevance for various diagnostic purposes like localization tumors within the or quantification emphysema. Since emphysema is a known risk factor cancer, both are even related to each other. The main steps pipeline described this paper detector and based on watershed algorithm, lobe mesh model adaptation. procedure was applied data sets base Image Database Resource Initiative (IDRI) that currently contains over 500 scans with...
To facilitate systematic calibration and validation of quantitative airway measurements on CT for COPD diagnosis, an acrylic plastic phantom has been designed with array cylindrical tubes varying lumen diameter wall thickness in a way, which can be manufactured by inexpensive 3D-printing. Accuracy reproducibility the 3Dprinting have confirmed measurements. The multipliable, unobtrusive scanned simultaneously patient each exam, allows scan-specific calibration, thus improve multicenter study...
Based on prior studies spectral CT has shown a higher sensitivity for malignant lesions than conventional at the cost of lower specificity. For radiologists, it also offers degree certainty in diagnosis benign lesions. The objective this study was to evaluate economic impact patients suspected occult cancer medical center Denmark.This secondary analysis using de-identified data from prospective receiving contrast-enhanced scan. suggested follow-up examinations both and CT, costs payer's...