Automated thresholded region classification using a robust feature selection method for PET-CT

Bi, Lei; Kim, Jin Man; Wen, Lingfeng; Feng, Dagan; Fulham, Michael

doi:10.1109/isbi.2015.7164146

Cited by 3 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Over the years many different types of segmentation techniques have been developed, for example, [7][8][9]. Some of the previous techniques include thresholding [10], k-means clustering [11], watersheds [12], followed by more advanced algorithms such as active contour methods [8,13], graph cuts [14], random walks [15], conditional and Markov random fields [16] to name a few. In recent years, particularly the last decade, the field of Machine Learning (ML) and Deep Learning (DL) has seen exponential growth and has produced models that have shown remarkable performance across many benchmark datasets and many different problem domains [17,18].…”

Section: Introductionmentioning

confidence: 99%

Deep Learning-Based Methods for Prostate Segmentation in Magnetic Resonance Imaging

et al. 2021

View full text Add to dashboard Cite

Magnetic Resonance Imaging-based prostate segmentation is an essential task for adaptive radiotherapy and for radiomics studies whose purpose is to identify associations between imaging features and patient outcomes. Because manual delineation is a time-consuming task, we present three deep-learning (DL) approaches, namely UNet, efficient neural network (ENet), and efficient residual factorized convNet (ERFNet), whose aim is to tackle the fully-automated, real-time, and 3D delineation process of the prostate gland on T2-weighted MRI. While UNet is used in many biomedical image delineation applications, ENet and ERFNet are mainly applied in self-driving cars to compensate for limited hardware availability while still achieving accurate segmentation. We apply these models to a limited set of 85 manual prostate segmentations using the k-fold validation strategy and the Tversky loss function and we compare their results. We find that ENet and UNet are more accurate than ERFNet, with ENet much faster than UNet. Specifically, ENet obtains a dice similarity coefficient of 90.89% and a segmentation time of about 6 s using central processing unit (CPU) hardware to simulate real clinical conditions where graphics processing unit (GPU) is not always available. In conclusion, ENet could be efficiently applied for prostate delineation even in small image training datasets with potential benefit for patient management personalization.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep Learning-Based Methods for Prostate Segmentation in Magnetic Resonance Imaging

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Unfortunately, a full comparison of such algorithms is not possible as the relative studies often concern different body district and specific types of abnormality, e.g., lung tumours [40][41][42], oesophageal tumours [43], and nasopharyngeal tumours [44]. Studies on discrimination of pathological structures in whole-body PET have been conducted as well, and some preliminary results have shown that different anatomical areas pose different challenges [45,46]. In addition, numerous PET-based radiomics studies have been proposed and the results of the relative analysis are highly dependent on the method used to derive the BTV [47,48].…”

Section: Introductionmentioning

confidence: 99%

A smart and operator independent system to delineate tumours in Positron Emission Tomography scans

Comelli

Stefano

Russo

et al. 2018

Computers in Biology and Medicine

View full text Add to dashboard Cite

Positron Emission Tomography (PET) imaging has an enormous potential to improve radiation therapy treatment planning offering complementary functional information with respect to other anatomical imaging approaches. The aim of this study is to develop an operator independent, reliable, and clinically feasible system for biological tumour volume delineation from PET images. Under this design hypothesis, we combine several known approaches in an original way to deploy a system with a high level of automation. The proposed system automatically identifies the optimal region of interest around the tumour and performs a slice-by-slice marching local active contour segmentation. It automatically stops when a "cancer-free" slice is identified. User intervention is limited at drawing an initial rough contour around the cancer region. By design, the algorithm performs the segmentation minimizing any dependence from the initial input, so that the final result is extremely repeatable. To assess the performances under different conditions, our system is evaluated on a dataset comprising five synthetic experiments and fifty oncological lesions located in different anatomical regions (i.e. lung, head and neck, and brain) using PET studies with 18F-fluoro-2-deoxy-d-glucose and 11C-labeled Methionine radio-tracers. Results on synthetic lesions demonstrate enhanced performances when compared against the most common PET segmentation methods. In clinical cases, the proposed system produces accurate segmentations (average dice similarity coefficient: 85.36 ± 2.94%, 85.98 ± 3.40%, 88.02 ± 2.75% in the lung, head and neck, and brain district, respectively) with high agreement with the gold standard (determination coefficient R = 0.98). We believe that the proposed system could be efficiently used in the everyday clinical routine as a medical decision tool, and to provide the clinicians with additional information, derived from PET, which can be of use in radiation therapy, treatment, and planning.

show abstract

“…In previous work, we conducted preliminary studies to address the simultaneous classification of abnormalities and multiple normal structures on whole-body PET-CT studies [32][33][34]. Our approach was to detect all the potential abnormalities e.g., thresholding and then iteratively filtering out normal structures rather than model lesions that can have inconsistent shapes and localization information.…”

Section: Introduction [ 18 F]fluorodeoxyglucose Positron Emission Tommentioning

confidence: 99%

“…We extended this work to cluster the thresholded fragments thereby increasing the discriminative power of the features derived from clustered fragments when compared to using individual fragments [33]. We also investigated the optimal feature representation to individual structures using a structure based feature selection strategy together with a SVM for classification [34]. These previous approaches relied on using individual thresholded fragments which lack discriminative power especially for small fragments (as shown in Figure 1b) [32,34].…”

Section: Introduction [ 18 F]fluorodeoxyglucose Positron Emission Tommentioning

confidence: 99%

“…We also investigated the optimal feature representation to individual structures using a structure based feature selection strategy together with a SVM for classification [34]. These previous approaches relied on using individual thresholded fragments which lack discriminative power especially for small fragments (as shown in Figure 1b) [32,34]. The clustering based method assembled the fragments of the same structure (see In this study, we propose a novel algorithm that uses a multi-scale superpixel-based encoding method (MSE) and a class-driven feature selection and classification model (CFSC) for sFEPU classification in whole-body PET-CT lymphoma studies.…”

Section: Introduction [ 18 F]fluorodeoxyglucose Positron Emission Tommentioning

confidence: 99%

See 1 more Smart Citation

Automatic detection and classification of regions of FDG uptake in whole-body PET-CT lymphoma studies

Kim

Kumar

et al. 2017

Computerized Medical Imaging and Graphics

Self Cite

View full text Add to dashboard Cite

[F]-Fluorodeoxyglucose (FDG) positron emission tomography-computed tomography (PET-CT) scans of lymphoma patients usually show disease involvement as foci of increased radiotracer uptake. Existing methods for detecting abnormalities, model the characteristics of these foci; this is challenging due to the inconsistent shape and localization information about the lesions. Thresholding the degree of FDG uptake is the standard method to separate different sites of involvement. But may fragment sites into smaller regions, and may also incorrectly identify sites of normal physiological FDG uptake and normal FDG excretion (sFEPU) such as the kidneys, bladder, brain and heart. These sFEPU can obscure sites of abnormal uptake, which can make image interpretation problematic. Identifying sFEPU is therefore important for improving the sensitivity of lesion detection and image interpretation. Existing methods to identify sFEPU are inaccurate because they fail to account for the low inter-class differences between sFEPU fragments and their inconsistent localization information. In this study, we address this issue by using a multi-scale superpixel-based encoding (MSE) to group the individual sFEPU fragments into larger regions, thereby, enabling the extraction of highly discriminative image features via domain transferred convolutional neural networks. We then classify there regions into one of the sFEPU classes using a class-driven feature selection and classification model (CFSC) method that avoids overfitting to the most frequently occurring classes. Our experiments on 40 whole-body lymphoma PET-CT studies show that our method achieved better accuracy (an average F-score of 91.73%) compared to existing methods in the classification of sFEPU.

show abstract

Automated thresholded region classification using a robust feature selection method for PET-CT

Cited by 3 publications

References 15 publications

Deep Learning-Based Methods for Prostate Segmentation in Magnetic Resonance Imaging

Deep Learning-Based Methods for Prostate Segmentation in Magnetic Resonance Imaging

A smart and operator independent system to delineate tumours in Positron Emission Tomography scans

Automatic detection and classification of regions of FDG uptake in whole-body PET-CT lymphoma studies

Contact Info

Product

Resources

About