A Fuzzy Locally Adaptive Bayesian Segmentation Approach for Volume Determination in PET

Hatt, Mathieu; Rest, Catherine Cheze Le; Turzo, A.; Roux, Christian; Visvikis, Dimitris

doi:10.1109/tmi.2008.2012036

Cited by 297 publications

(310 citation statements)

References 19 publications

Supporting

Mentioning

298

Contrasting

Unclassified

Order By: Relevance

“…There are early reports that textural analysis, an additional tool quantifying intratumoural heterogeneity of 18 FDG-PET tracer uptake, may improve prediction of response and prognosis and it is hypothesised that image heterogeneity may be related to underlying biology and reflect the behaviour of malignant tumours [3][4][5][6][7].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The precision of textural analysis in 18F-FDG-PET scans of oesophageal cancer

et al. 2015

View full text Add to dashboard Cite

Abstract!Objectives: Measuring tumour heterogeneity by textural analysis in 18 F-fluorodeoxyglucose positron emission tomography ( 18 F-FDG PET) provides predictive and prognostic information but technical aspects of image processing can influence parameter measurements. We therefore tested effects of image smoothing, segmentation and quantisation on the precision of heterogeneity measurements.Methods: Sixty-four 18 F-FDG PET/CT scans of oesophageal cancer were processed using different Gaussian smoothing levels (2.0, 2.5, 3.0, 3.5, 4.0mm), maximum standardised uptake values (SUV max ) segmentation thresholds (45%, 50%, 55%, 60%) and quantisation (8,16, 32, 64, 128 2) Quantisation shows large effects on precision of heterogeneity parameters in 18 F-FDG PET/CT.

show abstract

Section: Introductionmentioning

confidence: 99%

“…The measurement of tumour heterogeneity in 18 F-FDG PET images can be achieved by using statistical or model-based methods. Statistical-based textural analysis can be further categorised into first-, second-and higher-order statistical methods of increasing complexity, respectively [8][9][10][11][12][13][14][15].…”

Section: Introductionmentioning

confidence: 99%

The precision of textural analysis in 18F-FDG-PET scans of oesophageal cancer

et al. 2015

View full text Add to dashboard Cite

show abstract

“…Positron emission tomography (PET) successfully images the lesion metabolic activity. Recently, PET images were modeled as a fuzzy Gaussian mixture to delineate tumor lesions accurately [1]. In this work, we propose a statistical lesion activity computation (SLAC) approach to robustly estimate TLA directly from the modeled Gaussian partial volume mixtures.…”

mentioning

confidence: 99%

“…In this work, we propose a statistical lesion activity computation (SLAC) approach to robustly estimate TLA directly from the modeled Gaussian partial volume mixtures. TLA was estimated from 3 state-ofthe-art PET delineation schemes, namely a stochastic (FLAB [1]), a gradient based (GDM [2]) and an adaptive threshold based (ATM [3]) method, for comparison. A threshold based region growing method (T40 -40% threshold) was also evaluated.…”

mentioning

confidence: 99%

Fuzzy Statistical Unsupervised Learning Based Total Lesion Metabolic Activity Estimation in Positron Emission Tomography Images

George

Vunckx

Tejpar

et al. 2011

Machine Learning in Medical Imaging

View full text Add to dashboard Cite

Accurate total tumor lesion metabolic activity (TLA) estimation is critical for tumor staging and follow up studies. Positron emission tomography (PET) successfully images the lesion metabolic activity. Recently, PET images were modeled as a fuzzy Gaussian mixture to delineate tumor lesions accurately [1]. In this work, we propose a statistical lesion activity computation (SLAC) approach to robustly estimate TLA directly from the modeled Gaussian partial volume mixtures. TLA was estimated from 3 state-ofthe-art PET delineation schemes, namely a stochastic (FLAB [1]), a gradient based (GDM [2]) and an adaptive threshold based (ATM [3]) method, for comparison. A threshold based region growing method (T40 -40% threshold) was also evaluated. Synthetic lesions were simulated and reconstructed using maximum likelihood expectation maximization (MLEM) algorithm with 4 iterations over 16 subsets as in clinical routine (post-smoothed with 5mm Gaussian full width half maximum (FWHM)). Methods: Statistical TLALet the observed PET image and the hidden segmentation map be realizations y = {y s } s∈S and x = {x s } s∈S of the random fields Y = {Y s } s∈S and X = {X s } s∈S respectively, where S = {1, . . . , N } is the set of voxels. Integrating the activity y s over the whole S, the total PET activity (TPA), constituting contributions from the lesion (TLA) and the background activities, can be mathematically defined asThe observed histogram h(ζ) containing the frequency of occurrence for each ζ can be related to the density f (ξ) defining the distribution of Y s = ξ. E(Y s ) is the associated expectation. Let c = {c k } k∈K , where K = {1, . . . , Q}, be the Q class labels associated with the segmentation map. Then Y s can be modeled as a finite mixture of conditional densities such that f (ξ) = k γ k f (ξ | c k ), where γ k stands for the mixing probabilities P (X s = c k ) and f (ξ | c k ) denotes the density defining the distribution ofTo model Y s with the PVE, let the noise associated with the tumor and the background be modeled by 2 Gaussian random variables Y 1 and Y 0 with the density N (µ 1 , σ 2 1 ) (f (ξ | X s = 1)) and N (µ 0 , σ 2 0 ) (f (ξ | X s = 0)) defining the distributions of Y s conditional to X s = 1 and X s = 0 respectively. Then the partial volume activities can be modeled by linear combinations of these independent random variables as. . , M . Therefore, for the tumor and the background, µ M +1 = µ 1 and µ 0 = µ 0 respectively, with 0 = 0 and M +1 = 1. Here c = {0, 1, k }, Q = M +2. In eq. 1, out of the total activity in a fuzzy mixture classoriginates from the tumor lesion. Adding up the contributions from 1 hard class and M fuzzy classes representing tumor and partial volume voxels respectively, we propose that, the statistical TLA can be estimated from TPA as A line profile through the PET image (37, 28, 22, 17, 13 and 10mm lesions) and the ground truth is shown in red and black lines respectively. Area in light gray shade denotes the TLA computed from the delineation profile, where as the area in d...

show abstract

“…For a given data set composed of a set of items, a statistical classification framework attempts to label each item to with some level of certainty, like in [267]. For examples, More complex segmentation methodologies have been proposed to solve the lung tumor delineation problem [268,[271][272][273][274][275][276][277][278][279]. For example, Li et al [278] used an adaptive region growing method that extracts the tumor boundaries using deformable models in PET.…”

Section: Pet Segmentation Techniquesmentioning

confidence: 99%

Computational methods for the analysis of functional 4D-CT chest images.

Soliman¹

View full text Add to dashboard Cite

DEDICATIONThis dissertation is dedicated to my father's soul.iii ACKNOWLEDGMENTS In the name of Allah the most merciful, the most compassionate. All deepest thanks are due to Almighty Allah for the uncountable gifts given to me. Medical imaging is an important emerging technology that has been intensively used in the last few decades for disease diagnosis and monitoring as well as for the assessment of treatment effectiveness. Medical images provide a very large amount of valuable information that is too huge to be exploited by radiologists and physicians. Therefore, the design of computer-aided diagnostic (CAD) system, which can be used as an assistive tool for the medical community, is of a great importance. This dissertation deals with the development of a complete CAD system for lung cancer patients, which remains the leading cause of cancer-related death in the USA. In 2014, there were approximately 224,210 new cases of lung cancer and 159,260 related deaths. The process begins with the detection of lung cancer which is detected through the diagnosis of lung nodules (a manifestation of lung cancer). These nodules are approximately spherical regions of primarily high density tissue that are visible in computed tomography (CT) images of the lung. The treatment of these lung cancer nodules is complex, nearly 70% of lung cancer patients require radiation therapy as part of their treatment. Radiation-induced lung injury is a limiting toxicity that may decrease cure rates and increase morbidity and mortality treatment. By finding ways to accurately detect, at early stage, and hence prevent lung injury, it will have significant positive consequences for lung cancer patients.v The ultimate goal of this dissertation is to develop a clinically usable CAD system that can improve the sensitivity and specificity of early detection of radiation-induced lung injury based on the hypotheses that radiated lung tissues may get affected and suffer decrease of their functionality as a side effect of radiation therapy treatment. These hypotheses have been validated by demonstrating that automatic segmentation of the lung regions and registration of consecutive respiratory phases to estimate their elasticity, ventilation, and texture features to provide discriminatory descriptors that can be used for early detection of radiation-induced lung injury. The proposed methodologies will lead to novel indexes for distinguishing normal/healthy and injured lung tissues in clinical decisionmaking. To achieve this goal, a CAD system for accurate detection of radiation-induced lung injury that requires three basic components has been developed. These components are the lung fields segmentation, lung registration, and features extraction and tissue classification.This dissertation starts with an exploration of the available medical imaging modalities to present the importance of medical imaging in today's clinical applications. Secondly, the methodologies, challenges, and limitations of recent CAD systems for lung cancer detection are covered...

show abstract

A Fuzzy Locally Adaptive Bayesian Segmentation Approach for Volume Determination in PET

Cited by 297 publications

References 19 publications

The precision of textural analysis in 18F-FDG-PET scans of oesophageal cancer

The precision of textural analysis in 18F-FDG-PET scans of oesophageal cancer

Fuzzy Statistical Unsupervised Learning Based Total Lesion Metabolic Activity Estimation in Positron Emission Tomography Images

Computational methods for the analysis of functional 4D-CT chest images.

Contact Info

Product

Resources

About