1887
banner image
No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Registration of lung nodules using a semi-rigid model: Method and preliminary results
Rent:
Rent this article for
USD
10.1118/1.2432073
/content/aapm/journal/medphys/34/2/10.1118/1.2432073
http://aip.metastore.ingenta.com/content/aapm/journal/medphys/34/2/10.1118/1.2432073

Figures

Image of FIG. 1.
FIG. 1.

Illustration of the semi-rigid model with three retained structures, where and are defined as translation and rotation respectively, for structure .

Image of FIG. 2.
FIG. 2.

Diagram of the registration system.

Image of FIG. 3.
FIG. 3.

Cross-sectional images exemplifying registration results for the physiologically-based simulated transformation for a simulated nodule. (a) Nodule (half major axis of the largest ellipsoid: ) in source volume. The two white rings surround cross-sections of the two principle structures extracted for this example. (b)–(f) Target volumes. (b) Nodule maintains size and shape. Distance error was . (c) Nodule disappears. Distance error was . (d) Nodule grows uniformly by a factor of 2. Distance error was . (e) Nodule shrinks uniformly by a factor of 2. Distance error was . (f) Nodule grows nonuniformly. Distance error was .

Image of FIG. 4.
FIG. 4.

Experimental initial registration errors: (a) Ten simulated nodules. (b) Histogram of initial registration errors for the 97 nodules in the patient experiments.

Image of FIG. 5.
FIG. 5.

Mean and its 95% confidence interval of the distance errors between our method and a rigid registration method for (a) simulated nodules and for the (b) real patient experiments as a function of cube side .

Image of FIG. 6.
FIG. 6.

Mean distance error and its 95% confidence interval as a function of parameter settings. (a) Error versus minimum retained value, , with threshold , as parameter. (b) Error versus threshold, , with minimum retained value, , as parameter. Error as function of (c) number of structures , (d) nodule weight , (e) elastic energy weight , and (f) search range , respectively.

Image of FIG. 7.
FIG. 7.

The outlier when . Left: Nodule (encircled) in segmented source volume. Right: Nodule (encircled) in the target volume that grew uniformly by a factor of 2 in the segmented target volume, which became contiguous with the chest wall.

Image of FIG. 8.
FIG. 8.

An example of nodule in contact with blood vessel in one of the pairs of actual patient scans. (a) Source image. (b) Target image. Registration distance error for this nodule was .

Image of FIG. 9.
FIG. 9.

Segmentation steps (from left to right): Portion of original image, showing juxtapleural nodule in center. Left lung contour excludes nodule. Edge closing using line segment test. Portion of segmented lung with nodule included.

Tables

Generic image for table
TABLE I.

Results for the simulated nodules.

Loading

Article metrics loading...

/content/aapm/journal/medphys/34/2/10.1118/1.2432073
2007-01-24
2014-04-17
Loading

Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Registration of lung nodules using a semi-rigid model: Method and preliminary results
http://aip.metastore.ingenta.com/content/aapm/journal/medphys/34/2/10.1118/1.2432073
10.1118/1.2432073
SEARCH_EXPAND_ITEM