Publicação
Realistic 3D infant head surfaces augmentation to improve AI-based diagnosis of cranial deformities
| Resumo: | Evaluation of the head shape of newborns is needed to detect cranial deformities, disturbances in head growth, and consequently, to predict short- and long-term neurodevelopment. Currently, there is a lack of automatic tools to provide a detailed evaluation of the head shape. Artificial intelligence (AI) methods, namely deep learning (DL), can be explored to develop fast and automatic approaches for shape evaluation. However, due to the clinical variability of patients’ head anatomy, generalization of AI networks to the clinical needs is paramount and extremely challenging. In this work, a new framework is proposed to augment the 3D data used for training DL networks for shape evaluation. The proposed augmentation strategy deforms head surfaces towards different deformities. For that, a point-based 3D morphable model (p3DMM) is developed to generate a statistical model representative of head shapes of different cranial deformities. Afterward, a constrained transformation approach (3DHT) is applied to warp a head surface towards a target deformity by estimating a dense motion field from a sparse one resulted from the p3DMM. Qualitative evaluation showed that the proposed method generates realistic head shapes indistinguishable from the real ones. Moreover, quantitative experiments demonstrated that DL networks training with the proposed augmented surfaces improves their performance in terms of head shape analysis. Overall, the introduced augmentation allows to effectively transform a given head surface towards different deformity shapes, potentiating the development of DL approaches for head shape analysis. |
|---|---|
| Autores principais: | Torres, Helena R. |
| Outros Autores: | Oliveira, Bruno; Morais, Pedro André Gonçalves; Fritze, Anne; Rüdiger, Mario; Fonseca, Jaime C.; Vilaça, João L. |
| Assunto: | 3D data augmentation Deep learning Head deformities Morphable models Motion transformation |
| Ano: | 2022 |
| País: | Portugal |
| Tipo de documento: | artigo |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade do Minho |
| Idioma: | inglês |
| Origem: | RepositóriUM - Universidade do Minho |
| _version_ | 1866876998453821440 |
|---|---|
| author | Torres, Helena R. |
| author2 | Oliveira, Bruno Morais, Pedro André Gonçalves Fritze, Anne Rüdiger, Mario Fonseca, Jaime C. Vilaça, João L. |
| author2_role | author author author author author author |
| author_facet | Torres, Helena R. Oliveira, Bruno Morais, Pedro André Gonçalves Fritze, Anne Rüdiger, Mario Fonseca, Jaime C. Vilaça, João L. |
| author_role | author |
| contributor_name_str_mv | Universidade do Minho |
| country_str | PT |
| creators_json_txt | [{\"Person.name\":\"Torres, Helena R.\"},{\"Person.name\":\"Oliveira, Bruno\"},{\"Person.name\":\"Morais, Pedro André Gonçalves\"},{\"Person.name\":\"Fritze, Anne\"},{\"Person.name\":\"Rüdiger, Mario\"},{\"Person.name\":\"Fonseca, Jaime C.\"},{\"Person.name\":\"Vilaça, João L.\"}] |
| datacite.contributors.contributor.contributorName.fl_str_mv | Universidade do Minho |
| datacite.creators.creator.creatorName.fl_str_mv | Torres, Helena R. Oliveira, Bruno Morais, Pedro André Gonçalves Fritze, Anne Rüdiger, Mario Fonseca, Jaime C. Vilaça, João L. |
| datacite.date.Accepted.fl_str_mv | 2022-01-01T00:00:00Z |
| datacite.date.available.fl_str_mv | 2024-04-03T13:00:55Z |
| datacite.date.embargoed.fl_str_mv | 2024-04-03T13:00:55Z |
| datacite.rights.fl_str_mv | http://purl.org/coar/access_right/c_abf2 |
| datacite.subjects.subject.fl_str_mv | 3D data augmentation Deep learning Head deformities Morphable models Motion transformation |
| datacite.titles.title.fl_str_mv | Realistic 3D infant head surfaces augmentation to improve AI-based diagnosis of cranial deformities |
| dc.contributor.none.fl_str_mv | Universidade do Minho |
| dc.creator.none.fl_str_mv | Torres, Helena R. Oliveira, Bruno Morais, Pedro André Gonçalves Fritze, Anne Rüdiger, Mario Fonseca, Jaime C. Vilaça, João L. |
| dc.date.Accepted.fl_str_mv | 2022-01-01T00:00:00Z |
| dc.date.available.fl_str_mv | 2024-04-03T13:00:55Z |
| dc.date.embargoed.fl_str_mv | 2024-04-03T13:00:55Z |
| dc.format.none.fl_str_mv | application/pdf |
| dc.identifier.none.fl_str_mv | https://hdl.handle.net/1822/90519 |
| dc.language.none.fl_str_mv | eng |
| dc.publisher.none.fl_str_mv | Elsevier |
| dc.rights.none.fl_str_mv | http://purl.org/coar/access_right/c_abf2 |
| dc.subject.none.fl_str_mv | 3D data augmentation Deep learning Head deformities Morphable models Motion transformation |
| dc.title.fl_str_mv | Realistic 3D infant head surfaces augmentation to improve AI-based diagnosis of cranial deformities |
| dc.type.none.fl_str_mv | http://purl.org/coar/resource_type/c_6501 |
| description | Evaluation of the head shape of newborns is needed to detect cranial deformities, disturbances in head growth, and consequently, to predict short- and long-term neurodevelopment. Currently, there is a lack of automatic tools to provide a detailed evaluation of the head shape. Artificial intelligence (AI) methods, namely deep learning (DL), can be explored to develop fast and automatic approaches for shape evaluation. However, due to the clinical variability of patients’ head anatomy, generalization of AI networks to the clinical needs is paramount and extremely challenging. In this work, a new framework is proposed to augment the 3D data used for training DL networks for shape evaluation. The proposed augmentation strategy deforms head surfaces towards different deformities. For that, a point-based 3D morphable model (p3DMM) is developed to generate a statistical model representative of head shapes of different cranial deformities. Afterward, a constrained transformation approach (3DHT) is applied to warp a head surface towards a target deformity by estimating a dense motion field from a sparse one resulted from the p3DMM. Qualitative evaluation showed that the proposed method generates realistic head shapes indistinguishable from the real ones. Moreover, quantitative experiments demonstrated that DL networks training with the proposed augmented surfaces improves their performance in terms of head shape analysis. Overall, the introduced augmentation allows to effectively transform a given head surface towards different deformity shapes, potentiating the development of DL approaches for head shape analysis. |
| dirty | 0 |
| eu_rights_str_mv | openAccess |
| format | article |
| fulltext.url.fl_str_mv | https://prod-dspace.uminho.pt/bitstreams/35d6c8bf-f9cf-47d1-933f-8de759e2b983/download |
| id | rum_fdf84cd549703b6528811c08bac70ef6 |
| identifier.url.fl_str_mv | https://hdl.handle.net/1822/90519 |
| instacron_str | repositorium |
| institution | Universidade do Minho |
| instname_str | Universidade do Minho |
| language | eng |
| network_acronym_str | rum |
| network_name_str | RepositóriUM - Universidade do Minho |
| oai_identifier_str | oai:repositorium.uminho.pt:1822/90519 |
| organization_str_mv | urn:organizationAcronym:repositorium |
| person_str_mv | Torres, Helena R. Oliveira, Bruno Morais, Pedro André Gonçalves Fritze, Anne Rüdiger, Mario Fonseca, Jaime C. Vilaça, João L. |
| publishDate | 2022 |
| publisher.none.fl_str_mv | Elsevier |
| reponame_str | RepositóriUM - Universidade do Minho |
| repository_id_str | urn:repositoryAcronym:rum |
| service_str_mv | urn:repositoryAcronym:rum |
| spelling | engElsevierporEvaluation of the head shape of newborns is needed to detect cranial deformities, disturbances in head growth, and consequently, to predict short- and long-term neurodevelopment. Currently, there is a lack of automatic tools to provide a detailed evaluation of the head shape. Artificial intelligence (AI) methods, namely deep learning (DL), can be explored to develop fast and automatic approaches for shape evaluation. However, due to the clinical variability of patients’ head anatomy, generalization of AI networks to the clinical needs is paramount and extremely challenging. In this work, a new framework is proposed to augment the 3D data used for training DL networks for shape evaluation. The proposed augmentation strategy deforms head surfaces towards different deformities. For that, a point-based 3D morphable model (p3DMM) is developed to generate a statistical model representative of head shapes of different cranial deformities. Afterward, a constrained transformation approach (3DHT) is applied to warp a head surface towards a target deformity by estimating a dense motion field from a sparse one resulted from the p3DMM. Qualitative evaluation showed that the proposed method generates realistic head shapes indistinguishable from the real ones. Moreover, quantitative experiments demonstrated that DL networks training with the proposed augmented surfaces improves their performance in terms of head shape analysis. Overall, the introduced augmentation allows to effectively transform a given head surface towards different deformity shapes, potentiating the development of DL approaches for head shape analysis.application/pdfporRealistic 3D infant head surfaces augmentation to improve AI-based diagnosis of cranial deformitiesTorres, Helena R.Oliveira, BrunoMorais, Pedro André GonçalvesFritze, AnneRüdiger, MarioFonseca, Jaime C.Vilaça, João L.HostingInstitutionOrganizationalUniversidade do Minhoe-mailmailto:repositorium@usdb.uminho.ptrepositorium@usdb.uminho.ptISSNIsPartOf1532-0464DOIIsPartOf10.1016/j.jbi.2022.1041212024-04-03T13:00:55Z20222024-04-03T11:58:31Z2022-01-01T00:00:00ZHandlehttps://hdl.handle.net/1822/90519http://purl.org/coar/access_right/c_abf2open access3D data augmentationDeep learningHead deformitiesMorphable modelsMotion transformation6705731 bytesliteraturehttp://purl.org/coar/resource_type/c_6501journal articlehttp://purl.org/coar/access_right/c_abf2application/pdffulltexthttps://prod-dspace.uminho.pt/bitstreams/35d6c8bf-f9cf-47d1-933f-8de759e2b983/download |
| spellingShingle | Realistic 3D infant head surfaces augmentation to improve AI-based diagnosis of cranial deformities Torres, Helena R. 3D data augmentation Deep learning Head deformities Morphable models Motion transformation |
| status | SINGLETON |
| subject.fl_str_mv | 3D data augmentation Deep learning Head deformities Morphable models Motion transformation |
| title | Realistic 3D infant head surfaces augmentation to improve AI-based diagnosis of cranial deformities |
| title_full | Realistic 3D infant head surfaces augmentation to improve AI-based diagnosis of cranial deformities |
| title_fullStr | Realistic 3D infant head surfaces augmentation to improve AI-based diagnosis of cranial deformities |
| title_full_unstemmed | Realistic 3D infant head surfaces augmentation to improve AI-based diagnosis of cranial deformities |
| title_short | Realistic 3D infant head surfaces augmentation to improve AI-based diagnosis of cranial deformities |
| title_sort | Realistic 3D infant head surfaces augmentation to improve AI-based diagnosis of cranial deformities |
| topic | 3D data augmentation Deep learning Head deformities Morphable models Motion transformation |
| topic_facet | 3D data augmentation Deep learning Head deformities Morphable models Motion transformation |
| url | https://hdl.handle.net/1822/90519 |
| visible | 1 |