Development of a multiple-template matching technique for removal of false positives in a computer-aided diagnostic scheme

2001 ◽  
Author(s):  
Qiang Li ◽  
Shigehiko Katsuragawa ◽  
Roger M. Engelmann ◽  
Samuel G. Armato III ◽  
Heber MacMahon ◽  
...  
Keyword(s):  
Template Matching ◽  
False Positives ◽  
Computer Aided ◽  
Matching Technique ◽  
Multiple Template

An improved species based genetic algorithm and its application in multiple template matching for embroidered pattern inspection

2011 ◽  
Vol 38 (12) ◽  
pp. 15172-15182 ◽  
Author(s):  
Na Dong ◽  
Chun-Ho Wu ◽  
Wai-Hung Ip ◽  
Zeng-Qiang Chen ◽  
Ching-Yuen Chan ◽  
...  
Keyword(s):  
Genetic Algorithm ◽  
Template Matching ◽  
Multiple Template

scholarly journals Hierarchical Region Based Template Matching Technique for Global Motion Reduction of Coronary Cineangiograms

2015 ◽  
Vol 7 (2) ◽  
pp. 156-161 ◽  
Author(s):  
Kulathilake K. A. S. H. ◽  
Ranathunga L. ◽  
Constantine G. R. ◽  
Abdullah N. A.
Keyword(s):  
Template Matching ◽  
Global Motion ◽  
Matching Technique

Spectral yield estimation of NTS explosions

1991 ◽  
Vol 81 (4) ◽  
pp. 1292-1308
Author(s):  
Steven R. Taylor ◽  
Farid U. Dowla
Keyword(s):  
Shock Wave ◽  
Template Matching ◽  
Spectral Shape ◽  
Source Material ◽  
Matching Function ◽  
Spectral Matching ◽  
Matching Technique ◽  
Regression Techniques ◽  
The Difference ◽  
Unknown Event

Abstract The yields of 299 NTS explosions have been estimated from Pn, Pg and Lg spectra (between 0.1 and 10 Hz) at four regional seismic stations. A spectral template matching technique is used where the spectra from an explosion of unknown yield are compared with the spectra of explosions of known yield. A matching function is defined that is a scaled inverse of the difference between the spectra from the known and unknown explosions. The yields from the seven closest matching explosions are then averaged to estimate the yield of the unknown event. The spectral matching technique appears to perform as well as standard regression techniques utilizing mb(Pn) and mb(Lg) measurements except that no geologic information (such as gas-filled porosity) is required. However, the spectral matching technique is only applicable to very well-calibrated test sites. The key to spectral matching is that the spectral shape is sensitive to the near-source geology. In addition to affecting the absolute spectral levels (i.e., coupling), the dynamic response of the near source material to the radiated shock wave is a major factor controlling the shape of the radiated spectra. The spectral shape can therefore be used as an indicator for predicting the coupling of an explosion, which can be subsequently used to predict its yield.

Abstract T P32: Computer Aided Detection of Asymptomatic Cerebral Microbleeds on Magnetic Resonance Images

Stroke ◽  
2014 ◽  
Vol 45 (suppl_1) ◽  
Author(s):  
Yan Li ◽  
Yining Huang ◽  
Jue Zhang ◽  
Jing Fang
Keyword(s):  
Magnetic Resonance ◽  
Rating Scale ◽  
Characteristic Curve ◽  
Magnetic Resonance Images ◽  
False Positives ◽  
Cerebral Microbleeds ◽  
Cad System ◽  
Automated Method ◽  
Skull Stripping ◽  
Computer Aided

Purpose: Manual rating of Cerebral microbleeds (CMBs) is time-consuming and inconsistent. Since the presence and number of CMBs have become a potential diagnostic and prognostic biomarker of stroke, an automatic identification method is required. We proposed a computer aided diagnosis (CAD) system for the detection of the CMBs on the magnetic resonance (MR) images automatically. Methods: Eighty-one patients were recruited in this study. CMBs on the MR T2* weighted images were manually rated according to the Microbleed Anatomic Rating Scale (MARS) criteria. Our automated method consisted of two steps: i) Pre-processing: After skull stripping, isolated islands of points were removed while holes were restored to avoid over segmentation. Local threshold segmentation was applied for the initial candidate selection. ii) Identification model: Seven features were extracted from each candidate: area, roundness, intensity, average of the boundary, contrast, shape-intensity and location-mark (according to the probability density templates calculated from the location information of the CMBs). For further identification of each candidate, Random Forest (RF) model was used to distinguish CMBs from the mimics. Results: A total of 337 CMBs in the 81 patients were studied. Comparing with the counting from the experienced doctors, high sensitivity of 92% (310/337) was achieved after pre-processing. The RF model eliminated most of the false-positives while maintaining a reliable sensitivity of 94% (291/310) and specificity of 96% (4272/4450). The area under the Receiver operating characteristic curve was 0.98 ± 0.02 for the detection model. In summary, this CAD system had an overall sensitivity of 86% (291/337) and specificity of 96% (4272/4450), producing only 2.2 false-positives per subject. Conclusion: This presented strategy is technically effective. The results indicate that it has the potential to be used for clinical detection of CMBs.

Sign in / Sign up

Export Citation Format

Share Document