A5043024040- Radiomics and Machine Learning in Medical Imaging
- Medical Imaging Techniques and Applications
- Medical Image Segmentation Techniques
- 3D Shape Modeling and Analysis
- Advanced X-ray and CT Imaging
- Computer Graphics and Visualization Techniques
- Augmented Reality Applications
- Surgical Simulation and Training
- Lung Cancer Diagnosis and Treatment
- AI in cancer detection
- MRI in cancer diagnosis
- Advanced Radiotherapy Techniques
- Image and Object Detection Techniques
- Inhalation and Respiratory Drug Delivery
- Advanced Neural Network Applications
- Image Retrieval and Classification Techniques
- Advanced Vision and Imaging
- Digital Image Processing Techniques
- Respiratory Support and Mechanisms
- Image and Signal Denoising Methods
- Chronic Obstructive Pulmonary Disease (COPD) Research
- Robotics and Sensor-Based Localization
- Advanced Image and Video Retrieval Techniques
- Digital Imaging for Blood Diseases
- Advanced MRI Techniques and Applications
University of Iowa
2015-2024
Engineering Arts (United States)
2019
Carver Bible College
2014
University of Iowa Hospitals and Clinics
2012
Institute of Computer Vision and Applied Computer Sciences
2002-2009
Graz University of Technology
2000-2007
University of Graz
2004
Joanneum Research
2000
<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> This paper presents a comparison study between 10 automatic and six interactive methods for liver segmentation from contrast-enhanced CT images. It is based on results the "MICCAI 2007 Grand Challenge" workshop, where 16 teams evaluated their algorithms common database. A collection of 20 clinical images with reference segmentations was provided to train tune in advance. Participants were also...
This paper describes a framework for establishing reference airway tree segmentation, which was used to quantitatively evaluate fifteen different extraction algorithms in standardized manner. Because of the sheer difficulty involved manually constructing complete standard from scratch, we propose construct using results all that are be evaluated. We start by subdividing each segmented into its individual branch segments. Each segment is then visually scored trained observers determine...
Segmentation of lungs with (large) lung cancer regions is a nontrivial problem. We present new fully automated approach for segmentation such high-density pathologies. Our method consists two main processing steps. First, novel robust active shape model (RASM) matching utilized to roughly segment the outline lungs. The initial position RASM found by means rib cage detection method. Second, an optimal surface finding further adapt result lung. Left and right are segmented individually. An...
In liver surgery planning, 2D and desktop-based 3D systems offer surgeons limited assistance. By using VR technology to liberate from input devices such as the mouse keyboard, this planning system better supports surgeons. User studies show that is both effective easy use.
The authors evaluated the apparent diffusion coefficient (ADC) in assessment of vertebral metastases and acute compression fractures 22 patients with known or suspected metastases. On basis significantly (P <.03) different ADCs, (0.69 x 10(-3) mm2/sec) pathologic (0.65 can be safely distinguished from bodies (1.66 benign (1.62 mm2/sec). Thus, use ADCs may increase specificity magnetic resonance imaging these patients.
Background. Imaging biomarkers hold tremendous promise for precision medicine clinical applications. Development of such relies heavily on image post-processing tools automated quantitation. Their deployment in the context research necessitates interoperability with systems. Comparison established outcomes and evaluation tasks motivate integration imaging data, use standardized approaches to support annotation sharing analysis results semantics. We developed methodology these Positron...
The impact of PET on radiation therapy is held back by poor methods defining functional volumes interest. Many new software tools are being proposed for contouring target but the different approaches not adequately compared and their accuracy poorly evaluated due to illdefinition ground truth. This paper compares largest cohort date established, emerging methods, in terms variability. We emphasise spatial present a metric that addresses lack unique 30 used at 13 institutions contour VOIs...
A new graph-based approach for segmentation of luminal and external elastic lamina (EEL) surface coronary vessels in gated 20 MHz intravascular ultrasound (IVUS) image sequences (volumes) is presented. The consists a fully automated stage ("new automated" or NA) user-guided computer-aided refinement refinement" NR) stage. Both approaches are based on the LOGISMOS simultaneous dual-surface segmentation. This combination allows user to efficiently combine general information about IVUS...
Active appearance models (AAMs) have been successfully used for a variety of segmentation tasks in medical image analysis. However, gross disturbances objects can occur routine clinical setting caused by pathological changes or interventions. This poses problem AAM-based segmentation, since the method is inherently not robust. In this paper, novel robust AAM (RAAM) matching algorithm presented. Compared to previous approaches, no assumptions are made regarding kind gray-value disturbance...
Rationale: Air trapping and airflow obstruction are being increasingly identified in infants with cystic fibrosis. These findings commonly attributed to airway infection, inflammation, mucus buildup.Objectives: To learn if air present before the onset of infection inflammation fibrosis.Methods: On day they born, piglets fibrosis lack inflammation. Therefore, we used newborn wild-type assess trapping, size, lung volume inspiratory expiratory X-ray computed tomography scans. Micro–computed...
The purpose of this work was to develop, validate, and compare a highly computer-aided method for the segmentation hot lesions in head neck 18F-FDG PET scans.A semiautomated developed, which transforms problem into graph-based optimization problem. For purpose, graph structure around user-provided approximate lesion centerpoint is constructed suitable cost function derived based on local image statistics. To handle frequently occurring situations that are ambiguous (e.g., adjacent each other...
Liver segmentation is an important prerequisite for the assessment of liver cancer treatment options like tumor resection, image-guided radiation therapy (IGRT), radiofrequency ablation, etc. The purpose this work was to evaluate a new approach segmentation.A graph cuts method combined with three-dimensional virtual reality based refinement approach. developed interactive system allowed user manipulate volume chunks and∕or surfaces instead 2D contours in cross-sectional images (i.e,...
Radiation treatment of cancers like prostate or cervix cancer requires considering nearby bone structures vertebrae. In this work, we present and validate a novel automated method for the 3D segmentation individual lumbar thoracic vertebra in computed tomography (CT) scans. It is based on single, low-complexity convolutional neural network (CNN) architecture which works well even if little application-specific training data are available. volume patch-based processing, enabling handling...
Radiomics is an image analysis approach for extracting large amounts of quantitative information from medical images using a variety computational methods. Our goal was to evaluate the utility radiomic feature 18F-fluorothymidine positron emission tomography (FLT PET) obtained at baseline in prediction treatment response patients with head and neck cancer. Thirty advanced-stage oropharyngeal or laryngeal cancer, treated definitive chemoradiation therapy, underwent FLT PET imaging before...
In the era of precision oncology and publicly available datasets, amount information for each patient case has dramatically increased. From clinical variables PET-CT radiomics measures to DNA-variant RNA expression profiles, such a wide variety data presents multitude challenges. Large datasets are subject sparsely and/or inconsistently populated fields. Corresponding sequencing profiles can suffer from problem high-dimensionality, where making useful inferences be difficult without...
Model-based segmentation methods have the advantage of incorporating a priori shape information into process but suffer from drawback that model must be initialized sufficiently close to target. We propose novel approach for initializing an active (ASM) and apply it 3D lung in CT scans. Our method constructs atlas consisting set representative features average shape. The ASM pose parameters are found by transforming based on affine transform computed matching between new image features....
For multicenter clinical studies, characterizing the robustness of image-derived radiomics features is essential. Features calculated on PET images have been shown to be very sensitive image noise. The purpose this work was investigate efficacy a relatively simple harmonization strategy feature and agreement. A purpose-built texture pattern phantom scanned 10 different scanners in 7 institutions with various acquisition reconstruction protocols. An technique based equalizing...
Surgical planning of liver tumor resections requires detailed three-dimensional (3D) understanding the complex arrangement vasculature, segments and tumors. Knowledge about location sizes is important for choosing an optimal surgical resection approach predicting postoperative residual capacity. The aim this work to facilitate such process by developing a robust method portal vein tree segmentation. also investigates impact vessel segmentation on approximation segment volumes. For...
This paper presents a novel system for interactive visualization and manipulation of medical datasets surgery planning based on hybrid VR / Tablet PC user interface. The goal the is to facilitate efficient visual inspection correction surface models generated by automated segmentation algorithms x-ray computed tomography scans, needed surgical interventions. Factors like quality visualization, nature dataset interaction efficiency strongly influence design decisions, in particular interface,...
RATIONALE AND OBJECTIVES: This article describes issues and methods that are specific to the measurement of change in tumor volume as measured from computed tomographic (CT) images how these would relate establishment CT volumetrics a biomarker patient response therapy. The primary focus is on lung tumors, but approach should be generalizable other anatomic regions. MATERIALS METHODS: first addressed various sources bias variance volumes, which discussed context variation its impact early...