Performance measure characterization for evaluating neuroimage segmentation algorithms

Chang, Herng Hua; Zhuang, Audrey H.; Valentino, Daniel J.; Chu, Woei Chyn

doi:10.1016/j.neuroimage.2009.03.068

Cited by 237 publications

(122 citation statements)

References 38 publications

Supporting

Mentioning

118

Contrasting

Unclassified

Order By: Relevance

“…This provided additional evidence on the validity and applicability of our image processing protocol and selection criterions for segmenting TC. Our findings based on the Dice and Jaccard analysis were in agrement with those of the previous studies who sized other body organs or anatomical structures (Chang et al, 2009;Duda et al, 2001;Lee et al, 2003;La Macchia et al, 2012).…”

Section: Discussionsupporting

confidence: 91%

Characterization of Tympanic Cavity Volume in Newborns using Computerized Tomography Scanning

et al. 2016

View full text Add to dashboard Cite

Morphol., 34(1):189-196, 2016. SUMMARY:This study reports tympanic cavity (TC) volume in newborns, which was missing in the literature. Ex vivo histology and computerized tomography (CT) scans were performed on temporal bone and data were analyzed in part using software developed in house. CT images with a slice thickness of 0.5 mm were obtained from 5 newborn cadavers and analyzed independently by two expert researchers. The border of the TC was delineated manually and measurement of area of interest was calculated on masked images. Then, the area measurements from all sections were added to estimate the total volume. The agreements between the histological and CT findings were then compared for accuracy, repeatability and reliability. The Dice and Jaccard similarity coefficient measures were used as a statistical validation metric to evaluate the assessor's performance in manual volume segmentation. Good assessor agreement was observed with average Dice values above 0.8 indicating that consistent and reliable volume measurements were feasible. The proposed protocol was shown to be accurate in calculating the TC volume, and thus can be used for computer-assisted presurgical planning or for diagnosing structural alterations in TC.

show abstract

Section: Discussionsupporting

confidence: 91%

Characterization of Tympanic Cavity Volume in Newborns using Computerized Tomography Scanning

et al. 2016

View full text Add to dashboard Cite

show abstract

“…For the regional evaluation, the following metrics were calculated: sensitivity, specificity, positive predictive value (PPV), Jaccard index [23], Dice index [15], accuracy, conformity and sensibility [11]. These parameters measure set agreement in terms of false positive, false negative, true negative and true positive counts.…”

Section: Regional Evaluationmentioning

confidence: 99%

“…Accuracy is defined as Conformity and sensibility are new metrics better suited for evaluating segmentation results with small objects. Conformity is defined by measuring the ratio of the number of mis-segmented voxels to the number of correctly segmented voxels using the following equation [11]:…”

Section: Regional Evaluationmentioning

confidence: 99%

Three-dimensional segmentation of retroperitoneal masses using continuous convex relaxation and accumulated gradient distance for radiotherapy planning

Suárez-Mejías

Pérez-Carrasco

Serrano

et al. 2016

Med Biol Eng Comput

View full text Add to dashboard Cite

An innovative algorithm has been developed for the segmentation of retroperitoneal tumors in 3D radiological images. This algorithm makes it possible for radiation oncologists and surgeons semiautomatically to select tumors for possible future radiation treatment and surgery. It is based on continuous convex relaxation methodology, the main novelty being the introduction of accumulated gradient distance, with intensity and gradient information being incorporated into the segmentation process. The algorithm was used to segment 26 CT image volumes. The results were compared with manual contouring of the same tumors. The proposed algorithm achieved 90 % sensitivity, 100 % specificity and 84 % positive predictive value, obtaining a mean distance to the closest point of 3.20 pixels. The algorithm's dependence on the initial manual contour was also analyzed, with results showing that the algorithm substantially reduced the variability of the manual segmentation carried out by different specialists. The algorithm was also compared with four benchmark algorithms (thresholding, edge-based level-set, region-based level-set and continuous max-flow with two labels). To the best of our knowledge, this is the first time the segmentation of retroperitoneal tumors for radiotherapy planning has been addressed.

show abstract

“…The speed term in (15) is multiplied by the level set function updated when its movem urvature , which : (14) l set function that which would be control function, given as: …”

Section: E Implementationmentioning

confidence: 99%

A level set based deformable model for segmentation of human brain MR images

Chang

2014

2014 7th International Conference on Biomedical Engineering and Informatics

Self Cite

View full text Add to dashboard Cite

Segmentation of brain tissue from non-brain tissue, also known as skull stripping, has been challenging due to the complexity of anatomical brain structures and variable parameters of MR imaging modalities. It has been one of the most important preprocessing steps in medical image analysis. We propose a new brain segmentation algorithm that is based on a level set based deformable model. Two different sources of forces are proposed to evolve the level set-based contour. First, the brain surface attraction force is calculated based on the gray level intensity distribution of the brain, which is designed to automatically adjust the intensive parameters in response to different slices. The other force is a morphological smoothing force based on the mean curvature, which is further weighted by the differences between the mean intensity values inside and outside the contour to the zero level set intensity values. Experimental results indicated that the proposed algorithm is effectively accurate and robust, which is a promising tool in many skull stripping applications.

show abstract

Performance measure characterization for evaluating neuroimage segmentation algorithms

Cited by 237 publications

References 38 publications

Characterization of Tympanic Cavity Volume in Newborns using Computerized Tomography Scanning

Characterization of Tympanic Cavity Volume in Newborns using Computerized Tomography Scanning

Three-dimensional segmentation of retroperitoneal masses using continuous convex relaxation and accumulated gradient distance for radiotherapy planning

A level set based deformable model for segmentation of human brain MR images

Contact Info

Product

Resources

About