Probabilistic Learning Coherent Point Drift for 3D Ultrasound Fetal Head Registration
Quantification of brain growth is crucial for the assessment of fetal well being, for which ultrasound (US) images are the chosen clinical modality. However, they present artefacts, such as acoustic occlusion, especially after the 18th gestational week, when cranial calcification appears. Fetal US v...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Hindawi Limited
2021
|
| Online Access: | https://hdl.handle.net/1721.1/132941 |
| _version_ | 1826197949957275648 |
|---|---|
| author | Perez–Gonzalez, Jorge Arámbula Cosío, Fernando Huegel, Joel C. Medina-Bañuelos, Verónica |
| author2 | Massachusetts Institute of Technology. Center for Extreme Bionics |
| author_facet | Massachusetts Institute of Technology. Center for Extreme Bionics Perez–Gonzalez, Jorge Arámbula Cosío, Fernando Huegel, Joel C. Medina-Bañuelos, Verónica |
| author_sort | Perez–Gonzalez, Jorge |
| collection | MIT |
| description | Quantification of brain growth is crucial for the assessment of fetal well being, for which ultrasound (US) images are the chosen clinical modality. However, they present artefacts, such as acoustic occlusion, especially after the 18th gestational week, when cranial calcification appears. Fetal US volume registration is useful in one or all of the following cases: to monitor the evolution of fetometry indicators, to segment different structures using a fetal brain atlas, and to align and combine multiple fetal brain acquisitions. This paper presents a new approach for automatic registration of real 3D US fetal brain volumes, volumes that contain a considerable degree of occlusion artefacts, noise, and missing data. To achieve this, a novel variant of the coherent point drift method is proposed. This work employs supervised learning to segment and conform a point cloud automatically and to estimate their subsequent weight factors. These factors are obtained by a random forest-based classification and are used to appropriately assign nonuniform membership probability values of a Gaussian mixture model. These characteristics allow for the automatic registration of 3D US fetal brain volumes with occlusions and multiplicative noise, without needing an initial point cloud. Compared to other intensity and geometry-based algorithms, the proposed method achieves an error reduction of 7.4% to 60.7%, with a target registration error of only 6.38 ± 3.24 mm. This makes the herein proposed approach highly suitable for 3D automatic registration of fetal head US volumes, an approach which can be useful to monitor fetal growth, segment several brain structures, or even compound multiple acquisitions taken from different projections. |
| first_indexed | 2024-09-23T10:56:15Z |
| format | Article |
| id | mit-1721.1/132941 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T10:56:15Z |
| publishDate | 2021 |
| publisher | Hindawi Limited |
| record_format | dspace |
| spelling | mit-1721.1/1329412022-09-27T16:02:33Z Probabilistic Learning Coherent Point Drift for 3D Ultrasound Fetal Head Registration Perez–Gonzalez, Jorge Arámbula Cosío, Fernando Huegel, Joel C. Medina-Bañuelos, Verónica Massachusetts Institute of Technology. Center for Extreme Bionics Quantification of brain growth is crucial for the assessment of fetal well being, for which ultrasound (US) images are the chosen clinical modality. However, they present artefacts, such as acoustic occlusion, especially after the 18th gestational week, when cranial calcification appears. Fetal US volume registration is useful in one or all of the following cases: to monitor the evolution of fetometry indicators, to segment different structures using a fetal brain atlas, and to align and combine multiple fetal brain acquisitions. This paper presents a new approach for automatic registration of real 3D US fetal brain volumes, volumes that contain a considerable degree of occlusion artefacts, noise, and missing data. To achieve this, a novel variant of the coherent point drift method is proposed. This work employs supervised learning to segment and conform a point cloud automatically and to estimate their subsequent weight factors. These factors are obtained by a random forest-based classification and are used to appropriately assign nonuniform membership probability values of a Gaussian mixture model. These characteristics allow for the automatic registration of 3D US fetal brain volumes with occlusions and multiplicative noise, without needing an initial point cloud. Compared to other intensity and geometry-based algorithms, the proposed method achieves an error reduction of 7.4% to 60.7%, with a target registration error of only 6.38 ± 3.24 mm. This makes the herein proposed approach highly suitable for 3D automatic registration of fetal head US volumes, an approach which can be useful to monitor fetal growth, segment several brain structures, or even compound multiple acquisitions taken from different projections. 2021-10-12T19:52:16Z 2021-10-12T19:52:16Z 2020-01 2019-08 2020-06-07T08:00:21Z Article http://purl.org/eprint/type/JournalArticle 1748-670X 1748-6718 https://hdl.handle.net/1721.1/132941 Jorge Perez–Gonzalez et al. “Probabilistic Learning Coherent Point Drift for 3D Ultrasound Fetal Head Registration,” Computational and Mathematical Methods in Medicine (January 2020): 4271519. en http://dx.doi.org/10.1155/2020/4271519 Computational and Mathematical Methods in Medicine Attribution 4.0 International http://creativecommons.org/licenses/by/4.0/ Copyright © 2020 Jorge Perez–Gonzalez et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. text/xml application/pdf Hindawi Limited Hindawi |
| spellingShingle | Perez–Gonzalez, Jorge Arámbula Cosío, Fernando Huegel, Joel C. Medina-Bañuelos, Verónica Probabilistic Learning Coherent Point Drift for 3D Ultrasound Fetal Head Registration |
| title | Probabilistic Learning Coherent Point Drift for 3D Ultrasound Fetal Head Registration |
| title_full | Probabilistic Learning Coherent Point Drift for 3D Ultrasound Fetal Head Registration |
| title_fullStr | Probabilistic Learning Coherent Point Drift for 3D Ultrasound Fetal Head Registration |
| title_full_unstemmed | Probabilistic Learning Coherent Point Drift for 3D Ultrasound Fetal Head Registration |
| title_short | Probabilistic Learning Coherent Point Drift for 3D Ultrasound Fetal Head Registration |
| title_sort | probabilistic learning coherent point drift for 3d ultrasound fetal head registration |
| url | https://hdl.handle.net/1721.1/132941 |
| work_keys_str_mv | AT perezgonzalezjorge probabilisticlearningcoherentpointdriftfor3dultrasoundfetalheadregistration AT arambulacosiofernando probabilisticlearningcoherentpointdriftfor3dultrasoundfetalheadregistration AT huegeljoelc probabilisticlearningcoherentpointdriftfor3dultrasoundfetalheadregistration AT medinabanuelosveronica probabilisticlearningcoherentpointdriftfor3dultrasoundfetalheadregistration |