A5081098585- AI in cancer detection
- Digital Radiography and Breast Imaging
- Global Cancer Incidence and Screening
- Cancer Risks and Factors
- Image and Signal Denoising Methods
- Gene expression and cancer classification
- Geochemistry and Geologic Mapping
- Medical Imaging Techniques and Applications
- Breast Cancer Treatment Studies
- Water Quality and Resources Studies
- Radiomics and Machine Learning in Medical Imaging
- Advanced Data Compression Techniques
- Advanced Image Fusion Techniques
- Soil Geostatistics and Mapping
- Optical Imaging and Spectroscopy Techniques
- Colorectal Cancer Screening and Detection
- Spectroscopy and Chemometric Analyses
- Estrogen and related hormone effects
- Lung Cancer Treatments and Mutations
- Image Retrieval and Classification Techniques
- Breast Lesions and Carcinomas
- Statistical Methods and Bayesian Inference
- Data Analysis with R
- Electrical and Bioimpedance Tomography
- Ultrasound Imaging and Elastography
Moffitt Cancer Center
2016-2025
University of South Florida
2000-2020
Mayo Clinic
2012
Katholisches Klinikum Lünen/Werne , St.-Marien-Hospital Lünen
2011
Purpose To compare the classification of breast density with two automated methods, Volpara (version 1.5.0; Matakina Technology, Wellington, New Zealand) and Quantra 2.0; Hologic, Bedford, Mass), clinical Breast Imaging Reporting Data System (BI-RADS) classifications to examine associations these measures cancer risk. Materials Methods In this study, 1911 patients 4170 control subjects matched for age, race, examination date, mammography machine were evaluated. Participants underwent at Mayo...
Abstract Mammographic percent density (PD) is a strong risk factor for breast cancer, but there has been relatively little systematic evaluation of other features in mammographic images that might additionally predict cancer risk. We evaluated the association large number image texture with using clinic-based case-control study digitized film mammograms, all screening mammograms before diagnosis. The sample was split into training (123 cases and 258 controls) validation 264 data sets....
Purpose To extract radiologic features from small pulmonary nodules (SPNs) that did not meet the original criteria for a positive screening test and identify associated with lung cancer risk by using data images National Lung Screening Trial (NLST). Materials Methods Radiologic in SPNs baseline low-dose computed tomography (CT) studies NLST to be considered examination were extracted. identified 73 incident case patients who given diagnosis of at either first or second follow-up study 157...
Abstract Breast density is a strong risk factor for breast cancer; however, no standard assessment method exists. An automated was modified and compared with semi-automated, user-assisted thresholding (Cumulus method) the Imaging Reporting Data System four-category tissue composition measure their ability to predict future cancer risk. The three estimation methods were evaluated in matched case-control (n = 372 n 713, respectively) study at Mayo Clinic using digitized film mammograms....
Abstract Introduction Mammographic density is a strong risk factor for breast cancer. Image acquisition technique varies across mammograms to limit radiation and produce clinically useful image. We examined whether parameters at the time of mammography were associated with mammographic confounded cancer association. Methods this question within Mayo Mammography Health Study (MMHS) cohort, comprised 19,924 women (51.2% eligible) seen in Clinic screening practice from 2003 2006. A case-cohort...
Abstract Background: Reductions in breast density with tamoxifen and aromatase inhibitors may be an intermediate marker of treatment response. We compare changes volumetric among cancer cases using or (AI) to untreated women without cancer. Methods: Breast a digital mammogram prior diagnosis after initiation (n = 366) AI 403) sample controls 2170) were identified from the Mayo Clinic Mammography Practice San Francisco Registry. Volumetric percent (VPD) dense volume (DV) measured Volpara...
In this study we developed 25 computed tomography descriptors among 117 patients with lung adenocarcinoma to semiquantitatively assess their association overall survival. Pleural attachment was significantly associated an increased risk of death and texture most important for distinguishing histological subtypes. This approach has the potential support automated analyses develop decision-support clinical tools.Computed (CT) characteristics derived from noninvasive images that represent...
Abstract Percent mammographic density (PMD) is a strong breast cancer risk factor, however, other features, such as V, the standard deviation (SD) of pixel intensity, may be associated with risk. We assessed whether PMD, automated PMD (APD), and yielded independent associations included 1900 cases 3921 matched controls from Nurses’ Health Study (NHS) NHSII. Using digitized film mammograms, we estimated using computer-assisted thresholding technique. APD V were determined an computer...
We show that digitized mammograms can be considered as evolving from a simple process. A given image results passing random input field through linear filtering operation, where the filter transfer function has self-similar characteristic. By estimating functional form of and solving corresponding equation, analysis shows gray value distribution spectral content approximated with parametric methods. The work gives explanation for variegated appearance multimodal character common to...
A multiresolution statistical method for identifying clinically normal tissue in digitized mammograms is used to construct an algorithm separating regions from potentially abnormal regions; that is, small may contain isolated calcifications. This the initial phase of development a general automatic recognition mammograms. The first step decompose image with wavelet expansion yields sum independent images, each containing different levels detail. When calcifications are present, there strong...
The spectral content of mammograms acquired from using a full field digital mammography (FFDM) system are analyzed. Fourier methods used to show that the FFDM image power spectra obey an inverse law; in average sense, images may be considered as fields. Two data representations analyzed and compared (1) raw data, (2) logarithm data. employed analyze technique based on integrating plane with octave ring sectioning developed previously, approach rings constant width for this work. Both allow...
Mammographic density has been established as a strong risk factor for breast cancer, primarily using digitized film mammograms. Full-field digital mammography (FFDM) is replacing mammography, different properties than film, and provides both raw processed clinical display representation images. We evaluated compared FFDM measures their associations with cancer. A case-control study of 180 cases controls matched by age, postmenopausal hormone use, screening history was conducted. Mammograms...
Mammographic breast density declines during menopause. We assessed changes in volumetric across the menopausal transition and factors that influence these changes.Women without a history of cancer, who had full field digital mammograms both pre- postmenopausal periods, at least 2 years apart, were sampled from four facilities within San Francisco Mammography Registry 2007 to 2013. Dense volume (DV) was using Volpara on time period. Annualized change DV postmenopause estimated linear mixed...
Abstract Introduction: Experimental studies have shown anti-carcinogenic properties of vitamin D, but meta-analyses randomized trials D supplementation show reduction in cancer mortality not incidence. This study evaluated the effects moderate-dose D3 on mammographic features associated with breast cancer, and gene expression pathways tumors. Methods: As part an ancillary to VITamin OmegA-3 TriaL (VITAL) trial, we collected digital mammograms from women ≥ 55 years old, a 2X2 factorial design...
A statistical methodology is presented based on a chi‐square probability analysis that allows the automated discrimination of radiolucent tissue (fat) from radiographic densities (fibroglandular tissue) in digitized mammograms. The method earlier work developed at this facility shows mammograms may be considered as evolving linear filtering operation where random input field passed through process. process reversible which solution with knowledge obtained raw image (the output). analogous to...
Calibrating mammograms to produce a standardized breast density measurement for cancer risk analysis requires an accurate spatial measure of the compressed thickness. Thickness inaccuracies due nominal system readout value and compression paddle orientation induce unacceptable errors in calibration.A thickness correction was developed evaluated using fully specified two-component surrogate model. A previously calibration approach based on effective radiation attenuation coefficient...
PLS initially creates uncorrelated latent variables which are linear combinations of the original input vectors Xi, where weights used to determine combinations, proportional covariance. Secondly, a least squares regression is then performed on subset extracted that lead lower and biased variance transformed data. This process, leads estimate coefficients when compared Ordinary Least Squares approach. Classical Principal Component Analysis (PCA), kernel ridge (KRR) techniques well known...