{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,30]],"date-time":"2026-01-30T08:22:50Z","timestamp":1769761370437,"version":"3.49.0"},"reference-count":33,"publisher":"Springer Science and Business Media LLC","issue":"7","license":[{"start":{"date-parts":[[2022,2,7]],"date-time":"2022-02-07T00:00:00Z","timestamp":1644192000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2022,2,7]],"date-time":"2022-02-07T00:00:00Z","timestamp":1644192000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Deutsches Forschungszentrum f\u00fcr K\u00fcnstliche Intelligenz GmbH (DFKI)"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Int J CARS"],"published-print":{"date-parts":[[2022,7]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec>\n                <jats:title>Purpose<\/jats:title>\n                <jats:p>This work aims for a systematic comparison of popular shape and appearance models. Here, two statistical and four deep-learning-based shape and appearance models are compared and evaluated in terms of their expressiveness described by their generalization ability and specificity as well as further properties like input data format, interpretability and latent space distribution and dimension.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Methods<\/jats:title>\n                <jats:p>Classical shape models and their locality-based extension are considered next to autoencoders, variational autoencoders, diffeomorphic autoencoders and generative adversarial networks. The approaches are evaluated in terms of generalization ability, specificity and likeness depending on the amount of training data. Furthermore, various latent space metrics are presented in order to capture further major characteristics of the models.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Results<\/jats:title>\n                <jats:p>The experimental setup showed that locality statistical shape models yield best results in terms of generalization ability for 2D and 3D shape modeling. However, the deep learning approaches show strongly improved specificity. In the case of simultaneous shape and appearance modeling, the neural networks are able to generate more realistic and diverse appearances. A major drawback of the deep-learning models is, however, their impaired interpretability and ambiguity of the latent space.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Conclusions<\/jats:title>\n                <jats:p>It can be concluded that for applications not requiring particularly good specificity, shape modeling can be reliably established with locality-based statistical shape models, especially when it comes to 3D shapes. However, deep learning approaches are more worthwhile in terms of appearance modeling.\n<\/jats:p>\n              <\/jats:sec>","DOI":"10.1007\/s11548-022-02567-6","type":"journal-article","created":{"date-parts":[[2022,2,7]],"date-time":"2022-02-07T03:02:25Z","timestamp":1644202945000},"page":"1213-1224","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":16,"title":["A systematic comparison of generative models for medical images"],"prefix":"10.1007","volume":"17","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8854-3517","authenticated-orcid":false,"given":"Hristina","family":"Uzunova","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Matthias","family":"Wilms","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Nils D.","family":"Forkert","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Heinz","family":"Handels","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jan","family":"Ehrhardt","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2022,2,7]]},"reference":[{"key":"2567_CR1","doi-asserted-by":"crossref","unstructured":"B\u00f4ne A, Louis M, Colliot O, Durrleman S (2019) Learning Low-Dimensional Representations of Shape Data Sets with Diffeomorphic Autoencoders. In: information processing in medical imaging, pp. 195\u2013207","DOI":"10.1007\/978-3-030-20351-1_15"},{"key":"2567_CR2","doi-asserted-by":"crossref","unstructured":"Chen M, Shi X, Zhang Y, Wu D, Guizani M (2017) Deep Features Learning for Medical Image Analysis with Convolutional Autoencoder Neural Network. IEEE Trans Big Data. pp. 1\u20131","DOI":"10.1109\/TBDATA.2017.2777862"},{"key":"2567_CR3","doi-asserted-by":"crossref","unstructured":"Cootes TF, Edwards GJ, Taylor CJ (1998) Active appearance models. In: European conference on computer vision, pp. 484\u2013498. Springer","DOI":"10.1007\/BFb0054760"},{"issue":"1","key":"2567_CR4","doi-asserted-by":"publisher","first-page":"38","DOI":"10.1006\/cviu.1995.1004","volume":"61","author":"TF Cootes","year":"1995","unstructured":"Cootes TF, Taylor CJ, Cooper DH, Graham J (1995) Active shape models-their training and application. Comput Vis Image Underst 61(1):38\u201359","journal-title":"Comput Vis Image Underst"},{"key":"2567_CR5","doi-asserted-by":"crossref","unstructured":"Davatzikos C, Tao X, Shen D (2003) Hierarchical active shape models, using the wavelet transform. IEEE Trans Med Imaging p","DOI":"10.1109\/TMI.2003.809688"},{"issue":"5","key":"2567_CR6","doi-asserted-by":"publisher","first-page":"83","DOI":"10.3390\/jimaging7050083","volume":"7","author":"M Elbattah","year":"2021","unstructured":"Elbattah M, Loughnane C, Gu\u00e9rin JL, Carette R, Cilia F, Dequen G (2021) Variational autoencoder for image-based augmentation of eye-tracking data. J Imag 7(5):83","journal-title":"J Imag"},{"key":"2567_CR7","doi-asserted-by":"publisher","first-page":"321","DOI":"10.1016\/j.neucom.2018.09.013","volume":"321","author":"M Frid-Adar","year":"2018","unstructured":"Frid-Adar M, Diamant I, Klang E, Amitai M, Goldberger J, Greenspan H (2018) GAN-based synthetic medical image augmentation for increased CNN performance in liver lesion classification. Neurocomputing 321:321\u2013331","journal-title":"Neurocomputing"},{"key":"2567_CR8","unstructured":"Ghosh P, Sajjadi MSM, Vergari A, Black M, Scholkopf B (2020) From variational to deterministic autoencoders. In: international conference on learning representations. https:\/\/openreview.net\/forum?id=S1g7tpEYDS"},{"key":"2567_CR9","unstructured":"Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y (2014) Generative Adversarial Nets. In: advances in neural information processing systems. 27, pp 2672\u20132680"},{"key":"2567_CR10","doi-asserted-by":"crossref","unstructured":"Guan S, Loew M (2020) An Internal Cluster Validity Index Using a Distance-based Separability Measure. In: 2020 IEEE 32nd international conference on tools with artificial intelligence (ICTAI), pp 827\u2013834","DOI":"10.1109\/ICTAI50040.2020.00131"},{"issue":"4","key":"2567_CR11","doi-asserted-by":"publisher","first-page":"224","DOI":"10.1002\/hbm.10123","volume":"19","author":"A Hammers","year":"2003","unstructured":"Hammers A, Allom R, Koepp MJ, Free SL, Myers R, Lemieux L, Mitchell TN, Brooks DJ, Duncan JS (2003) Three-dimensional maximum probability atlas of the human brain, with particular reference to the temporal lobe. Hum Brain Mapp 19(4):224\u2013247","journal-title":"Hum Brain Mapp"},{"issue":"4","key":"2567_CR12","doi-asserted-by":"publisher","first-page":"543","DOI":"10.1016\/j.media.2009.05.004","volume":"13","author":"T Heimann","year":"2009","unstructured":"Heimann T, Meinzer HP (2009) Statistical shape models for 3d medical image segmentation: a review. Med Image Anal 13(4):543\u2013563","journal-title":"Med Image Anal"},{"issue":"1","key":"2567_CR13","doi-asserted-by":"publisher","first-page":"332","DOI":"10.1016\/j.media.2015.10.006","volume":"26","author":"Y Hu","year":"2015","unstructured":"Hu Y, Gibson E, Ahmed HU, Moore CM, Emberton M, Barratt DC (2015) Population-based prediction of subject-specific prostate deformation for MR-to-ultrasound image registration. Med Image Anal 26(1):332\u2013344","journal-title":"Med Image Anal"},{"issue":"5","key":"2567_CR14","doi-asserted-by":"publisher","first-page":"265","DOI":"10.1007\/s11548-007-0138-9","volume":"2","author":"H Hufnagel","year":"2008","unstructured":"Hufnagel H, Pennec X, Ehrhardt J, Ayache N, Handels H (2008) Generation of a statistical shape model with probabilistic point correspondences and the expectation maximization-iterative closest point algorithm. Int J Comput Assist Radiol Surg 2(5):265\u2013273","journal-title":"Int J Comput Assist Radiol Surg"},{"issue":"8","key":"2567_CR15","doi-asserted-by":"publisher","first-page":"1211","DOI":"10.1007\/s11548-018-1785-8","volume":"13","author":"D Karimi","year":"2018","unstructured":"Karimi D, Samei G, Kesch C, Nir G, Salcudean SE (2018) Prostate segmentation in mri using a convolutional neural network architecture and training strategy based on statistical shape models. Int J Comput Assist Radiol Surg 13(8):1211\u20131219","journal-title":"Int J Comput Assist Radiol Surg"},{"key":"2567_CR16","unstructured":"Kingma D, Welling M (2014) Auto-Encoding Variational Bayes. In: international conference on learning representations"},{"key":"2567_CR17","doi-asserted-by":"crossref","unstructured":"Kirschner M, Becker M, Wesarg S (2011) 3D Active Shape Model Segmentation with Nonlinear Shape Priors. In: medical image computing and computer-assisted intervention \u2013 MICCAI 2011, pp. 492\u2013499","DOI":"10.1007\/978-3-642-23629-7_60"},{"key":"2567_CR18","doi-asserted-by":"publisher","first-page":"146","DOI":"10.1016\/j.media.2017.02.004","volume":"37","author":"J Kr\u00fcger","year":"2017","unstructured":"Kr\u00fcger J, Ehrhardt J, Handels H (2017) Statistical appearance models based on probabilistic correspondences. Med Image Anal 37:146\u2013159","journal-title":"Med Image Anal"},{"key":"2567_CR19","unstructured":"Larsen ABL, S\u00f8nderby SK, Larochelle H, Winther O (2016) Autoencoding beyond pixels using a learned similarity metric. In: international conference on machine learning, pp 1558\u20131566"},{"key":"2567_CR20","doi-asserted-by":"publisher","first-page":"161","DOI":"10.1007\/978-3-319-66182-7_19","volume-title":"Medical Image Computing and Computer Assisted Intervention - MICCAI 2017, LNCS","author":"F Milletari","year":"2017","unstructured":"Milletari F, Rothberg A, Jia J, Sofka M (2017) Integrating statistical prior knowledge into convolutional neural networks. In: Descoteaux M, Maier-Hein L, Franz A, Jannin P, Collins DL, Duchesne S (eds) Medical Image Computing and Computer Assisted Intervention - MICCAI 2017, LNCS, vol 10433. Springer, Cham, pp 161\u2013168"},{"key":"2567_CR21","unstructured":"Oord AV, Kalchbrenner N, Kavukcuoglu K (2016) Pixel Recurrent Neural Networks. In: proceedings of The 33rd international conference on machine learning, pp 1747\u20131756. PMLR"},{"key":"2567_CR22","unstructured":"Rezende D, Mohamed S (2015) Variational Inference with Normalizing Flows. In: proceedings of the 32nd international conference on machine learning, pp 1530\u20131538. PMLR"},{"issue":"2","key":"2567_CR23","first-page":"115","volume":"31","author":"JP Royston","year":"1982","unstructured":"Royston JP (1982) An extension of shapiro and Wilk\u2019s W test for normality to large samples. J R Stat Soc Ser C Appl Stat. 31(2):115\u2013124","journal-title":"J R Stat Soc Ser C Appl Stat."},{"key":"2567_CR24","doi-asserted-by":"publisher","first-page":"30","DOI":"10.1016\/j.media.2019.01.010","volume":"54","author":"T Schlegl","year":"2019","unstructured":"Schlegl T, Seeb\u00f6ck P, Waldstein SM, Langs G, Schmidt-Erfurth U (2019) F-AnoGAN: fast unsupervised anomaly detection with generative adversarial networks. Med Image Anal 54:30\u201344","journal-title":"Med Image Anal"},{"key":"2567_CR25","doi-asserted-by":"crossref","unstructured":"Sudre CH, Li W, Vercauteren T, Ourselin S, Jorge\u00a0Cardoso M (2017) Generalised Dice Overlap as a Deep Learning Loss Function for Highly Unbalanced Segmentations. In: deep learning in medical image analysis and multimodal learning for clinical decision support, pp 240\u2013248","DOI":"10.1007\/978-3-319-67558-9_28"},{"key":"2567_CR26","doi-asserted-by":"crossref","unstructured":"Uzunova H, Handels H, Ehrhardt J (2021) Guided Filter Regularization for Improved Disentanglement of Shape and Appearance in Diffeomorphic Autoencoders. In: medical imaging with deep learning \u2013 MIDL","DOI":"10.1007\/978-3-658-36932-3_16"},{"key":"2567_CR27","doi-asserted-by":"crossref","unstructured":"Uzunova H, Kruse J, Kaftan P, Wilms M, Forkert ND, Handels H, Ehrhardt J (2021) Analysis of Generative Shape Modeling Approaches: Latent Space Properties and Interpretability. In: Bildverarbeitung F\u00fcr Die Medizin 2021: proceedings, German workshop on medical image computing, Regensburg, march 7-9, 2021, pp. 344\u2013349","DOI":"10.1007\/978-3-658-33198-6_84"},{"issue":"3","key":"2567_CR28","doi-asserted-by":"publisher","first-page":"451","DOI":"10.1007\/s11548-018-1898-0","volume":"14","author":"H Uzunova","year":"2019","unstructured":"Uzunova H, Schultz S, Handels H, Ehrhardt J (2019) Unsupervised pathology detection in medical images using conditional variational autoencoders. Int J Comput Assist Radiol Surg 14(3):451\u2013461","journal-title":"Int J Comput Assist Radiol Surg"},{"key":"2567_CR29","doi-asserted-by":"crossref","unstructured":"Uzunova H, Wilms M, Handels H, Ehrhardt J (2017) Training CNNs for Image Registration from Few Samples with Model-based Data Augmentation. In: medical image computing and computer assisted intervention - MICCAI 2017, pp. 223\u2013231","DOI":"10.1007\/978-3-319-66182-7_26"},{"issue":"1","key":"2567_CR30","doi-asserted-by":"publisher","first-page":"19","DOI":"10.1016\/j.media.2005.02.002","volume":"10","author":"B van Ginneken","year":"2006","unstructured":"van Ginneken B, Stegmann MB, Loog M (2006) Segmentation of anatomical structures in chest radiographs using supervised methods: a comparative study on a public database. Med Image Anal 10(1):19\u201340","journal-title":"Med Image Anal"},{"key":"2567_CR31","doi-asserted-by":"crossref","unstructured":"Wilms M, Ehrhardt J, Forkert ND (2020) A Kernelized Multi-level Localization Method for Flexible Shape Modeling with Few Training Data. In: medical image computing and computer assisted intervention \u2013 MICCAI 2020, pp. 765\u2013775","DOI":"10.1007\/978-3-030-59719-1_74"},{"key":"2567_CR32","doi-asserted-by":"publisher","first-page":"17","DOI":"10.1016\/j.media.2017.02.003","volume":"38","author":"M Wilms","year":"2017","unstructured":"Wilms M, Handels H, Ehrhardt J (2017) Multi-resolution multi-object statistical shape models based on the locality assumption. Med Image Anal 38:17\u201329","journal-title":"Med Image Anal"},{"key":"2567_CR33","doi-asserted-by":"crossref","unstructured":"Yu X, Zhou F, Chandraker M (2016) Deep deformation network for object landmark localization. In: B.\u00a0Leibe, J.\u00a0Matas, N.\u00a0Sebe, M.\u00a0Welling (eds.) European conference on computer vision \u2013 ECCV 2016, LNCS, pp. 52\u201370. Springer","DOI":"10.1007\/978-3-319-46454-1_4"}],"container-title":["International Journal of Computer Assisted Radiology and Surgery"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11548-022-02567-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s11548-022-02567-6\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11548-022-02567-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,6,18]],"date-time":"2022-06-18T07:16:40Z","timestamp":1655536600000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s11548-022-02567-6"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,2,7]]},"references-count":33,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2022,7]]}},"alternative-id":["2567"],"URL":"https:\/\/doi.org\/10.1007\/s11548-022-02567-6","relation":{},"ISSN":["1861-6429"],"issn-type":[{"value":"1861-6429","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,2,7]]},"assertion":[{"value":"23 April 2021","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"14 January 2022","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"7 February 2022","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Hristina Uzunova, Matthias Wilms, Nils D. Forkert, Heinz Handels and Jan Ehrhardt declare that they have no conflict of interest. This article does not contain any studies with human participants or animals performed by any of the authors. For this type of study, formal consent is not required.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}]}}