Current Methods in Medical Image Segmentation

Pham, Dzung L.; Xu, Chenyang; Prince, Jerry L.

doi:10.1146/annurev.bioeng.2.1.315

Cited by 2,010 publications

(1,189 citation statements)

References 95 publications

Supporting

Mentioning

1,104

Contrasting

Unclassified

Order By: Relevance

“…Among many intensity‐based and atlas‐based segmentation methods, 23 , 24 five algorithms were selected. The global thresholding method was chosen as a representative thresholding method.…”

Section: Methodsmentioning

confidence: 99%

“…The fuzzy k‐means (FKM, or fuzzy c‐means) clustering method is an unsupervised method 23 , 44 that uses a soft membership function

m false(i, j false) = \frac{{‖ x_{i} - c_{j} ‖}^{- \frac{2}{r - 1}}}{true \sum_{j = 1}^{k} {‖ x_{i} - c_{j} ‖}^{- \frac{2}{r - 1}}},

where

x_{i}

is the value of each pixel,

c_{j}

is centroid of each cluster group, k is the total number of cluster groups,

‖ ■ ‖

is the Euclidean distance, and r is taken as 2. Pixels are grouped into the closest clusters by the membership function.…”

Section: Appendicesmentioning

confidence: 99%

See 1 more Smart Citation

A comparative study of automatic image segmentation algorithms for target tracking in MR‐IGRT

Feng

Kawrakow

Olsen

et al. 2016

J Applied Clin Med Phys

View full text Add to dashboard Cite

On‐board magnetic resonance (MR) image guidance during radiation therapy offers the potential for more accurate treatment delivery. To utilize the real‐time image information, a crucial prerequisite is the ability to successfully segment and track regions of interest (ROI). The purpose of this work is to evaluate the performance of different segmentation algorithms using motion images (4 frames per second) acquired using a MR image‐guided radiotherapy (MR‐IGRT) system. Manual contours of the kidney, bladder, duodenum, and a liver tumor by an experienced radiation oncologist were used as the ground truth for performance evaluation. Besides the manual segmentation, images were automatically segmented using thresholding, fuzzy k‐means (FKM), k‐harmonic means (KHM), and reaction‐diffusion level set evolution (RD‐LSE) algorithms, as well as the tissue tracking algorithm provided by the ViewRay treatment planning and delivery system (VR‐TPDS). The performance of the five algorithms was evaluated quantitatively by comparing with the manual segmentation using the Dice coefficient and target registration error (TRE) measured as the distance between the centroid of the manual ROI and the centroid of the automatically segmented ROI. All methods were able to successfully segment the bladder and the kidney, but only FKM, KHM, and VR‐TPDS were able to segment the liver tumor and the duodenum. The performance of the thresholding, FKM, KHM, and RD‐LSE algorithms degraded as the local image contrast decreased, whereas the performance of the VP‐TPDS method was nearly independent of local image contrast due to the reference registration algorithm. For segmenting high‐contrast images (i.e., kidney), the thresholding method provided the best speed (<1 ms) with a satisfying accuracy (Dice=0.95). When the image contrast was low, the VR‐TPDS method had the best automatic contour. Results suggest an image quality determination procedure before segmentation and a combination of different methods for optimal segmentation with the on‐board MR‐IGRT system.PACS number(s): 87.57.nm, 87.57.N‐, 87.61.Tg

show abstract

Section: Methodsmentioning

confidence: 99%

“…The fuzzy k‐means (FKM, or fuzzy c‐means) clustering method is an unsupervised method 23 , 44 that uses a soft membership function

m false(i, j false) = \frac{{‖ x_{i} - c_{j} ‖}^{- \frac{2}{r - 1}}}{true \sum_{j = 1}^{k} {‖ x_{i} - c_{j} ‖}^{- \frac{2}{r - 1}}},

where

x_{i}

is the value of each pixel,

c_{j}

is centroid of each cluster group, k is the total number of cluster groups,

‖ ■ ‖

is the Euclidean distance, and r is taken as 2. Pixels are grouped into the closest clusters by the membership function.…”

Section: Appendicesmentioning

confidence: 99%

A comparative study of automatic image segmentation algorithms for target tracking in MR‐IGRT

Feng

Kawrakow

Olsen

et al. 2016

J Applied Clin Med Phys

View full text Add to dashboard Cite

show abstract

“…MR imaging of the brain provides useful information about its anatomical structure, enabling and facilitating quantitative pathological or clinical investigation. Brain segmentation is a useful image processing method [1]. It assigns unique labels to two or more classes, e.g.…”

Section: Image Segmentationmentioning

confidence: 99%

“…Particularly, in the image segmentation problem, it is the classification truth of each pixel. For simplicity, let us assume a two-class truth by labelling the non-tumour class as C 0 and tumour class as C 1 . Medical images typically present a challenge with a large number of pixel data available, for example, N = 256 × 256 = 65 536 in a grey-scale image.…”

Section: Gold Standardmentioning

confidence: 99%

Three validation metrics for automated probabilistic image segmentation of brain tumours

Zou

Wells

Kikinis

et al. 2004

Statistics in Medicine

View full text Add to dashboard Cite

SUMMARYThe validity of brain tumour segmentation is an important issue in image processing because it has a direct impact on surgical planning. We examined the segmentation accuracy based on three twosample validation metrics against the estimated composite latent gold standard, which was derived from several experts' manual segmentations by an EM algorithm. The distribution functions of the tumour and control pixel data were parametrically assumed to be a mixture of two beta distributions with different shape parameters. We estimated the corresponding receiver operating characteristic curve, Dice similarity coefficient, and mutual information, over all possible decision thresholds. Based on each validation metric, an optimal threshold was then computed via maximization. We illustrated these methods on MR imaging data from nine brain tumour cases of three different tumour types, each consisting of a large number of pixels. The automated segmentation yielded satisfactory accuracy with varied optimal thresholds. The performances of these validation metrics were also investigated via Monte Carlo simulation. Extensions of incorporating spatial correlation structures using a Markov random field model were considered.

show abstract

“…It can be roughly classified into three categories, namely, manual, semiautomatic (interactive), and automatic segmentation (6) . Manual segmentation is usually time‐consuming and experience dependent.…”

Section: Introductionmentioning

confidence: 99%

SPARSE: Seed Point Auto‐Generation for Random Walks Segmentation Enhancement in medical inhomogeneous targets delineation of morphological MR and CT images

Chen

Zhen

et al. 2015

J Applied Clin Med Phys

View full text Add to dashboard Cite

In medical image processing, robust segmentation of inhomogeneous targets is a challenging problem. Because of the complexity and diversity in medical images, the commonly used semiautomatic segmentation algorithms usually fail in the segmentation of inhomogeneous objects. In this study, we propose a novel algorithm imbedded with a seed point autogeneration for random walks segmentation enhancement, namely SPARSE, for better segmentation of inhomogeneous objects. With a few user‐labeled points, SPARSE is able to generate extended seed points by estimating the probability of each voxel with respect to the labels. The random walks algorithm is then applied upon the extended seed points to achieve improved segmentation result. SPARSE is implemented under the compute unified device architecture (CUDA) programming environment on graphic processing unit (GPU) hardware platform. Quantitative evaluations are performed using clinical homogeneous and inhomogeneous cases. It is found that the SPARSE can greatly decrease the sensitiveness to initial seed points in terms of location and quantity, as well as the freedom of selecting parameters in edge weighting function. The evaluation results of SPARSE also demonstrate substantial improvements in accuracy and robustness to inhomogeneous target segmentation over the original random walks algorithm.PACS number: 87.57.nm

show abstract

Current Methods in Medical Image Segmentation

Cited by 2,010 publications

References 95 publications

A comparative study of automatic image segmentation algorithms for target tracking in MR‐IGRT

A comparative study of automatic image segmentation algorithms for target tracking in MR‐IGRT

Three validation metrics for automated probabilistic image segmentation of brain tumours

SPARSE: Seed Point Auto‐Generation for Random Walks Segmentation Enhancement in medical inhomogeneous targets delineation of morphological MR and CT images

Contact Info

Product

Resources

About