Technical report on semiautomatic segmentation using the Adobe Photoshop - PubMed Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2005 Dec;18(4):333-43.
doi: 10.1007/s10278-005-6704-1.

Technical report on semiautomatic segmentation using the Adobe Photoshop

Jin Seo Park  1 , Min Suk Chung, Sung Bae Hwang, Yong Sook Lee, Dong-Hwan Har
Affiliations

Technical report on semiautomatic segmentation using the Adobe Photoshop

Jin Seo Park et al. J Digit Imaging. 2005 Dec.

Abstract

The purpose of this research is to enable users to semiautomatically segment the anatomical structures in magnetic resonance images (MRIs), computerized tomographs (CTs), and other medical images on a personal computer. The segmented images are used for making 3D images, which are helpful to medical education and research. To achieve this purpose, the following trials were performed. The entire body of a volunteer was scanned to make 557 MRIs. On Adobe Photoshop, contours of 19 anatomical structures in the MRIs were semiautomatically drawn using MAGNETIC LASSO TOOL and manually corrected using either LASSO TOOL or DIRECT SELECTION TOOL to make 557 segmented images. In a similar manner, 13 anatomical structures in 8,590 anatomical images were segmented. Proper segmentation was verified by making 3D images from the segmented images. Semiautomatic segmentation using Adobe Photoshop is expected to be widely used for segmentation of anatomical structures in various medical images.

PubMed Disclaimer

Figures

Fig 1
Fig 1
(a) MRIs of the entire body. (b) One form of segmented images (TIFF files) where contours of the anatomical structures are filled with colors. (c) Another form of segmented images (AI files) where only the contours are colored.
Fig 2
Fig 2
(a) Temporary segmented image showing the skin's selection and cerebrum's selection on the MRI. (b) LAYERS WINDOW showing the MRI layer, skin's selection layer, and cerebrum's selection layer.
Fig 3
Fig 3
Temporary segmented image showing a zoomed-in work path, which consists of anchor points (□), segments between the anchor points, and control points (◊).
Fig 4
Fig 4
(a) Temporary segmented images showing the pink-filled skin's selection and green-filled cerebrum's selection. (b) LAYERS WINDOW showing the MRI layer, skin's selection layer, cerebrum's selection layer, skin's color layer, and cerebrum's color layer.
Fig 5
Fig 5
Simplified procedures for making the 160th coronal image. (a) The 160th row images in five triplicated images (t111.tif, t384.tif, t585.tif, t1050.tif) are cut. (b) They are pasted to the first row of the 160th temporary coronal image (tc160.psd). (c) They are downward distributed to make the 160th coronal image (c160.tif).
Fig 6
Fig 6
(a) Coronal images and (b) sagittal images made of the segmented images from the MRIs.
Fig 7
Fig 7
Procedures for making 3D image of the skin. (a) Contours of the skin are stacked. (b) Quadrilateral and triangular surfaces are filled between the neighboring contours. (c) The surfaces are filled with a color.
Fig 8
Fig 8
Three-dimensional image of several anatomical structures made of the segmented images from the MRIs.
Fig 9
Fig 9
(a) Anatomical images of the entire body. (b) Segmented images, where contours of the anatomical structures are filled with colors.
Fig 10
Fig 10
(a) Coronal images and (b) sagittal images made of the segmented images from the anatomical images.
Fig 11
Fig 11
(a) Three-dimensional image of skin and (b) 3D images of several anatomical structures made of the segmented images from the anatomical images.
Fig 12
Fig 12
Temporary segmented images showing (a) the selection, part of which is drawn along the contour of another anatomical structure (arrow) because of the high PIXEL WIDTH, and (b) the selection, part of which is not drawn along the uneven contour of the anatomical structure (arrow) because of the low FREQUENCY.
Fig 13
Fig 13
Various forms of the segmented images. (a) Contours of the anatomical structures are filled with colors. (b) Contours are filled with colors except for the skin's contours. (c) Only contours are colored. (d, e) Original images remain in the background.
Fig 14
Fig 14
Segmented images showing (a) the obscure contour after using ANTI-ALIASED, (b) the obscure contour after using FEATHER with high pixels, and (c) the distinct contour.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Kerr JP, Knapp D, Frake B, Sellberg M. True color surface anatomy. Mapping the Visible Human to patient-specific CT data. Comput Med Imaging Graph. 2000;24:153–164. doi: 10.1016/S0895-6111(00)00015-X. - DOI - PubMed
    1. Sachse FB, Werner CD, Meyer-Waarden K, Dössel O. Development of a human body model for numerical calculation of electrical fields. Comput Med Imaging Graph. 2000;24:165–171. doi: 10.1016/S0895-6111(00)00016-1. - DOI - PubMed
    1. Hosono M, Nakano Y, Urayama S, Konishi J, Uokawa K, Tanaka Y. Three-dimensional display of cardiac structures using reconstructed magnetic resonance imaging. J Digit Imaging. 1995;8:105–115. - PubMed
    1. Qin X, Ou Z, Zhang Y, Hou J. Data structure representation of 3D reconstruction system of medical images and surface model construction. Sheng Wu Yi Xue Gong Cheng Xue Za Zhi. 2002;19:239–243. - PubMed
    1. Uokawa K, Nakano Y, Urayama S, Uyama C, Kurokawa K, Ikeda K, Koito H, Tanaka Y. An improved computer methods to prepare 3D magnetic resonance images of thoracic structures. J Digit Imaging. 1997;10:85–95. - PMC - PubMed

Publication types

MeSH terms