A method for quantitative evaluation of statistical shape models using morphometry

Gollmer, Sebastian T.; Buzug, Thorsten M.

doi:10.1109/isbi.2010.5490312

Cited by 5 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, geometric depictions need to be efficient and compact, so that they can describe the shape of a structure with a minimal number of parameters (model compactness). Finally, such a model needs to be able to accurately describe members of the population outside of the training sample (model generalization) (Styner et al, 2003; Gollmer and Buzug, 2010). More specifically, and following Styner et al, we implemented the following “goodness” measures: rooted mean squared distance to in-training-set landmarks (accuracy) for different amounts of explained model variance, accumulated variance (compactness), approximation error (RMSE) to each training sample in a leave-one-out setting (generalization) and the average (RMSE) distance of uniformly distributed, randomly generated objects in the model shape space to their nearest member in the training set (specificity) (Styner et al, 2003).…”

Section: Methodsmentioning

confidence: 99%

Statistical Shape Modeling of Skeletal Anatomy for Sex Discrimination: Their Training Size, Sexual Dimorphism, and Asymmetry

Audenaert

Pattyn

Steenackers

et al. 2019

Front. Bioeng. Biotechnol.

View full text Add to dashboard Cite

Purpose: Statistical shape modeling provides a powerful tool for describing and analyzing human anatomy. By linearly combining the variance of the shape of a population of a given anatomical entity, statistical shape models (SSMs) identify its main modes of variation and may approximate the total variance of that population to a selected threshold, while reducing its dimensionality. Even though SSMs have been used for over two decades, they lack in characterization of their goodness of prediction, in particular when defining whether these models are actually representative for a given population.Methods: The current paper presents, to the authors' knowledge, the most extent lower limb anatomy shape model considering the pelvis, femur, patella, tibia, fibula, talus, and calcaneum to date. The present study includes the segmented training shapes (n = 542) obtained from 271 lower limb CT scans. The different models were evaluated in terms of accuracy, compactness, generalizability as well as specificity.Results: The size of training samples needed in each model so that it can be considered population covering was estimated to approximate around 200 samples, based on the generalizability properties of the different models. Simultaneously differences in gender and patterns in left-right asymmetry were identified and characterized. Size was found to be the most pronounced sexual discriminator whereas intra-individual variations in asymmetry were most pronounced at the insertion site of muscles.Conclusion: For models aimed at population covering descriptive studies, the number of training samples required should amount a sizeable 200 samples. The geometric morphometric method for sex discrimination scored excellent, however, it did not largely outperformed traditional methods based on discrete measures.

show abstract

Section: Methodsmentioning

confidence: 99%

Statistical Shape Modeling of Skeletal Anatomy for Sex Discrimination: Their Training Size, Sexual Dimorphism, and Asymmetry

Audenaert

Pattyn

Steenackers

et al. 2019

Front. Bioeng. Biotechnol.

View full text Add to dashboard Cite

show abstract

“…To test if we had included enough mandibles to describe the entire orthognathic population, a leave-on-out-cross-validation experiment was performed with 3 mandibles. The results were plotted as a generalization graph to determine if the current study’s results could be generalized to the entire population [ 9 , 10 ].…”

Section: Methodsmentioning

confidence: 99%

Morphological Variation of the Mandible in the Orthognathic Population—A Morphological Study Using Statistical Shape Modelling

Wel

Qiu

Spijkervet

et al. 2023

JPM

View full text Add to dashboard Cite

The aim of this study was to investigate the value of 3D Statistical Shape Modelling for orthognathic surgery planning. The goal was to objectify shape variations in the orthognathic population and differences between male and female patients by means of a statistical shape modelling method. Pre-operative CBCT scans of patients for whom 3D Virtual Surgical Plans (3D VSP) were developed at the University Medical Center Groningen between 2019 and 2020 were included. Automatic segmentation algorithms were used to create 3D models of the mandibles, and the statistical shape model was built through principal component analysis. Unpaired t-tests were performed to compare the principal components of the male and female models. A total of 194 patients (130 females and 64 males) were included. The mandibular shape could be visually described by the first five principal components: (1) The height of the mandibular ramus and condyles, (2) the variation in the gonial angle of the mandible, (3) the width of the ramus and the anterior/posterior projection of the chin, (4) the lateral projection of the mandible’s angle, and (5) the lateral slope of the ramus and the inter-condylar distance. The statistical test showed significant differences between male and female mandibular shapes in 10 principal components. This study demonstrates the feasibility of using statistical shape modelling to inform physicians about mandible shape variations and relevant differences between male and female mandibles. The information obtained from this study could be used to quantify masculine and feminine mandibular shape aspects and to improve surgical planning for mandibular shape manipulations.

show abstract

“…D represents the metric used to compute the distance between the shapes: by varying D, the significance of the difference changes. According to the approach presented in [42] , the Symmetric Mean (SM) distance calculated between the nearest points is used in this paper as the D metric. Using the SM calculated between the nearest points, instead of exploiting the same pairwise correspondences already defined during the TS definition, ensures to exclude from the evaluation any aspect regarding the registration process, thus providing only an assessment of the model's ability to match a given shape.…”

Section: Statistical Shape Model Evaluationmentioning

confidence: 99%

Hym3d: A Hybrid Method for the Automatic 3d Reconstruction of a Defective Cranial Vault

Marzola

Mussa

McGreevy

et al. 2023

Preprint

View full text Add to dashboard Cite

A method for quantitative evaluation of statistical shape models using morphometry

Cited by 5 publications

References 11 publications

Statistical Shape Modeling of Skeletal Anatomy for Sex Discrimination: Their Training Size, Sexual Dimorphism, and Asymmetry

Statistical Shape Modeling of Skeletal Anatomy for Sex Discrimination: Their Training Size, Sexual Dimorphism, and Asymmetry

Morphological Variation of the Mandible in the Orthognathic Population—A Morphological Study Using Statistical Shape Modelling

Hym3d: A Hybrid Method for the Automatic 3d Reconstruction of a Defective Cranial Vault

Contact Info

Product

Resources

About