Journal article
FAME - A Flexible Appearance Modelling Environment
Combined modelling of pixel intensities and shape has proven to be a very robust and widely applicable approach to interpret images. As such the Active Appearance Model (AAM) framework has been applied to a wide variety of problems within medical image analysis. This paper summarises AAM applications within medicine and describes a public domain implementation, namely the Flexible Appearance Modelling Environment (FAME).
We give guidelines for the use of this research platform, and show that the optimisation techniques used renders it applicable to interactive medical applications. To increase performance and make models generalise better, we apply parallel analysis to obtain automatic and objective model truncation.
Further, two different AAM training methods are compared along with a reference case study carried out on cross-sectional short-axis cardiac magnetic resonance images and face images. Source code and annotated data sets needed to reproduce the results are put in the public domain for further investigation.
Language: | English |
---|---|
Publisher: | IEEE |
Year: | 2003 |
Pages: | 1319-1331 |
ISSN: | 1558254x and 02780062 |
Types: | Journal article |
DOI: | 10.1109/TMI.2003.817780 |
ORCIDs: | Ersbøll, Bjarne Kjær and Larsen, Rasmus |
Active Appearance Models face segmentation left ventricular segmentation public domain training data and software
Active appearance model Algorithms Biomedical equipment Biomedical imaging Computer Simulation FAME Face Female Guidelines Heart Ventricles Humans Image Interpretation, Computer-Assisted Image analysis Magnetic Resonance Imaging Male Medical services Models, Biological Models, Statistical Neural Networks, Computer Pattern Recognition, Automated Pixel Rendering (computer graphics) Reproducibility of Results Robustness Sensitivity and Specificity Shape Video Recording biomedical MRI cardiology cross-sectional short-axis cardiac magnetic resonance images face images flexible appearance modeling environment image segmentation medical diagnostic imaging medical image analysis medical image processing modelling objective model truncation pixel intensities reference case study shape