Patient specific deep learning based segmentation for magnetic resonance guided prostate radiotherapy
Treatments on combined Magnetic Resonance (MR) scanners and Linear Accelerators (Linacs) for radiotherapy, called MR-Linacs, often require daily contouring. Currently, deformable image registration (DIR) algorithms propagate contours from reference scans, however large shape and size changes can be...
Saved in:
| Published in: | Physics and imaging in radiation oncology Vol. 23; pp. 38 - 42 |
|---|---|
| Main Authors: | , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Elsevier B.V
01.07.2022
Elsevier |
| Subjects: | |
| ISSN: | 2405-6316, 2405-6316 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | Treatments on combined Magnetic Resonance (MR) scanners and Linear Accelerators (Linacs) for radiotherapy, called MR-Linacs, often require daily contouring. Currently, deformable image registration (DIR) algorithms propagate contours from reference scans, however large shape and size changes can be troublesome. Artificial neural network (ANN) based contouring may alleviate this issue, however generally requires large datasets for training. Mitigating the problem of scarcity of data, we propose patient specific networks trained on a single dataset for each patient, for contouring onto the following datasets in an adaptive MR-Linacworkflow.
MR-scans from 17 prostate patients treated on an MR-Linac with contours of Clinical Target Volume (CTV), bladder and rectum were utilized. U-net shaped models were trained based on the image from the first fraction of each patient, and subsequently applied onto the following treatment images. Results were compared with manual contours in terms of the Dice coefficient and Added Path Length (APL). As benchmark, contours propagated through the clinical DIR algorithm were similarly evaluated.
In Dice coefficient the ANN output was 0.92 ± 0.03, 0.93 ± 0.07 and 0.84 ± 0.10 while for DIR 0.95 ± 0.03, 0.93 ± 0.08, 0.88 ± 0.06 for CTV, bladder and rectum respectively. Similarly, APL where 3109 ± 1642, 7250 ± 4234 and 5041 ± 2666 for ANN and 1835 ± 1621, 7236 ± 4287 and 4170 ± 2920 voxels for DIR.
Patient specific ANN models trained on images from the first fraction of a prostate MR-Linac treatment showed similar accuracy when applied to the subsequent fraction images as a clinically implemented DIR method. |
|---|---|
| AbstractList | Treatments on combined Magnetic Resonance (MR) scanners and Linear Accelerators (Linacs) for radiotherapy, called MR-Linacs, often require daily contouring. Currently, deformable image registration (DIR) algorithms propagate contours from reference scans, however large shape and size changes can be troublesome. Artificial neural network (ANN) based contouring may alleviate this issue, however generally requires large datasets for training. Mitigating the problem of scarcity of data, we propose patient specific networks trained on a single dataset for each patient, for contouring onto the following datasets in an adaptive MR-Linacworkflow.
MR-scans from 17 prostate patients treated on an MR-Linac with contours of Clinical Target Volume (CTV), bladder and rectum were utilized. U-net shaped models were trained based on the image from the first fraction of each patient, and subsequently applied onto the following treatment images. Results were compared with manual contours in terms of the Dice coefficient and Added Path Length (APL). As benchmark, contours propagated through the clinical DIR algorithm were similarly evaluated.
In Dice coefficient the ANN output was 0.92 ± 0.03, 0.93 ± 0.07 and 0.84 ± 0.10 while for DIR 0.95 ± 0.03, 0.93 ± 0.08, 0.88 ± 0.06 for CTV, bladder and rectum respectively. Similarly, APL where 3109 ± 1642, 7250 ± 4234 and 5041 ± 2666 for ANN and 1835 ± 1621, 7236 ± 4287 and 4170 ± 2920 voxels for DIR.
Patient specific ANN models trained on images from the first fraction of a prostate MR-Linac treatment showed similar accuracy when applied to the subsequent fraction images as a clinically implemented DIR method. Background and Purpose: Treatments on combined Magnetic Resonance (MR) scanners and Linear Accelerators (Linacs) for radiotherapy, called MR-Linacs, often require daily contouring. Currently, deformable image registration (DIR) algorithms propagate contours from reference scans, however large shape and size changes can be troublesome. Artificial neural network (ANN) based contouring may alleviate this issue, however generally requires large datasets for training. Mitigating the problem of scarcity of data, we propose patient specific networks trained on a single dataset for each patient, for contouring onto the following datasets in an adaptive MR-Linacworkflow. Materials and Methods: MR-scans from 17 prostate patients treated on an MR-Linac with contours of Clinical Target Volume (CTV), bladder and rectum were utilized. U-net shaped models were trained based on the image from the first fraction of each patient, and subsequently applied onto the following treatment images. Results were compared with manual contours in terms of the Dice coefficient and Added Path Length (APL). As benchmark, contours propagated through the clinical DIR algorithm were similarly evaluated. Results: In Dice coefficient the ANN output was 0.92 +/- 0.03, 0.93 +/- 0.07 and 0.84 +/- 0.10 while for DIR 0.95 +/- 0.03, 0.93 +/- 0.08, 0.88 +/- 0.06 for CTV, bladder and rectum respectively. Similarly, APL where 3109 +/- 1642, 7250 +/- 4234 and 5041 +/- 2666 for ANN and 1835 +/- 1621, 7236 +/- 4287 and 4170 +/- 2920 voxels for DIR. Conclusions: Patient specific ANN models trained on images from the first fraction of a prostate MR-Linac treatment showed similar accuracy when applied to the subsequent fraction images as a clinically implemented DIR method. Treatments on combined Magnetic Resonance (MR) scanners and Linear Accelerators (Linacs) for radiotherapy, called MR-Linacs, often require daily contouring. Currently, deformable image registration (DIR) algorithms propagate contours from reference scans, however large shape and size changes can be troublesome. Artificial neural network (ANN) based contouring may alleviate this issue, however generally requires large datasets for training. Mitigating the problem of scarcity of data, we propose patient specific networks trained on a single dataset for each patient, for contouring onto the following datasets in an adaptive MR-Linacworkflow.Background and PurposeTreatments on combined Magnetic Resonance (MR) scanners and Linear Accelerators (Linacs) for radiotherapy, called MR-Linacs, often require daily contouring. Currently, deformable image registration (DIR) algorithms propagate contours from reference scans, however large shape and size changes can be troublesome. Artificial neural network (ANN) based contouring may alleviate this issue, however generally requires large datasets for training. Mitigating the problem of scarcity of data, we propose patient specific networks trained on a single dataset for each patient, for contouring onto the following datasets in an adaptive MR-Linacworkflow.MR-scans from 17 prostate patients treated on an MR-Linac with contours of Clinical Target Volume (CTV), bladder and rectum were utilized. U-net shaped models were trained based on the image from the first fraction of each patient, and subsequently applied onto the following treatment images. Results were compared with manual contours in terms of the Dice coefficient and Added Path Length (APL). As benchmark, contours propagated through the clinical DIR algorithm were similarly evaluated.Materials and MethodsMR-scans from 17 prostate patients treated on an MR-Linac with contours of Clinical Target Volume (CTV), bladder and rectum were utilized. U-net shaped models were trained based on the image from the first fraction of each patient, and subsequently applied onto the following treatment images. Results were compared with manual contours in terms of the Dice coefficient and Added Path Length (APL). As benchmark, contours propagated through the clinical DIR algorithm were similarly evaluated.In Dice coefficient the ANN output was 0.92 ± 0.03, 0.93 ± 0.07 and 0.84 ± 0.10 while for DIR 0.95 ± 0.03, 0.93 ± 0.08, 0.88 ± 0.06 for CTV, bladder and rectum respectively. Similarly, APL where 3109 ± 1642, 7250 ± 4234 and 5041 ± 2666 for ANN and 1835 ± 1621, 7236 ± 4287 and 4170 ± 2920 voxels for DIR.ResultsIn Dice coefficient the ANN output was 0.92 ± 0.03, 0.93 ± 0.07 and 0.84 ± 0.10 while for DIR 0.95 ± 0.03, 0.93 ± 0.08, 0.88 ± 0.06 for CTV, bladder and rectum respectively. Similarly, APL where 3109 ± 1642, 7250 ± 4234 and 5041 ± 2666 for ANN and 1835 ± 1621, 7236 ± 4287 and 4170 ± 2920 voxels for DIR.Patient specific ANN models trained on images from the first fraction of a prostate MR-Linac treatment showed similar accuracy when applied to the subsequent fraction images as a clinically implemented DIR method.ConclusionsPatient specific ANN models trained on images from the first fraction of a prostate MR-Linac treatment showed similar accuracy when applied to the subsequent fraction images as a clinically implemented DIR method. Background and Purpose: Treatments on combined Magnetic Resonance (MR) scanners and Linear Accelerators (Linacs) for radiotherapy, called MR-Linacs, often require daily contouring. Currently, deformable image registration (DIR) algorithms propagate contours from reference scans, however large shape and size changes can be troublesome. Artificial neural network (ANN) based contouring may alleviate this issue, however generally requires large datasets for training. Mitigating the problem of scarcity of data, we propose patient specific networks trained on a single dataset for each patient, for contouring onto the following datasets in an adaptive MR-Linacworkflow. Materials and Methods: MR-scans from 17 prostate patients treated on an MR-Linac with contours of Clinical Target Volume (CTV), bladder and rectum were utilized. U-net shaped models were trained based on the image from the first fraction of each patient, and subsequently applied onto the following treatment images. Results were compared with manual contours in terms of the Dice coefficient and Added Path Length (APL). As benchmark, contours propagated through the clinical DIR algorithm were similarly evaluated. Results: In Dice coefficient the ANN output was 0.92 ± 0.03, 0.93 ± 0.07 and 0.84 ± 0.10 while for DIR 0.95 ± 0.03, 0.93 ± 0.08, 0.88 ± 0.06 for CTV, bladder and rectum respectively. Similarly, APL where 3109 ± 1642, 7250 ± 4234 and 5041 ± 2666 for ANN and 1835 ± 1621, 7236 ± 4287 and 4170 ± 2920 voxels for DIR. Conclusions: Patient specific ANN models trained on images from the first fraction of a prostate MR-Linac treatment showed similar accuracy when applied to the subsequent fraction images as a clinically implemented DIR method. |
| Author | Tilly, David Fransson, Samuel Strand, Robin |
| Author_xml | – sequence: 1 givenname: Samuel surname: Fransson fullname: Fransson, Samuel email: samuel.fransson@akademiska.se organization: Department of Medical Physics, Uppsala University Hospital, Uppsala, Sweden – sequence: 2 givenname: David surname: Tilly fullname: Tilly, David organization: Department of Medical Physics, Uppsala University Hospital, Uppsala, Sweden – sequence: 3 givenname: Robin surname: Strand fullname: Strand, Robin organization: Department of Surgical Sciences, Uppsala University, Uppsala, Sweden |
| BackLink | https://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-482662$$DView record from Swedish Publication Index (Uppsala universitet) |
| BookMark | eNqFks1u1DAUhSNUREvpC7DKkgUTbMd2EoSQqvJXqRIsgK11x77OeMjYwU5azdvjNkViuigrW_Z3jq7OPc-LIx88FsVLSipKqHyzrcZNDBUjjFVEVoTQJ8UJ40SsZE3l0T_34-IspS0hhDVdLWryrDiuRSM7SslJgd9gcuinMo2onXW6NIhjOSBE73xfriGhKRP2uwxlNPjShljuoPc4ZTpiCh68xrKfncnoGEPKIJYRjAvTBiOM-xfFUwtDwrP787T48enj94svq6uvny8vzq9WWrRyWlmUzDDooDW85tTUzAhtWy2YFRIFtaThDWMNp0IYsIyj7GTXNGDXgJQ39WlxufiaAFs1RreDuFcBnLp7CLFXEPPYAyro1hRbAthZwzXlQGSOB7VujCWG1dnr9eKVbnCc1wduH9zP8zu3eVa8ZVKyjL9f8Mzu0OgcV4ThQHX4491G9eFadazmjMls8OreIIbfM6ZJ7VzSOAzgMcxJMdmyphWsJhltF1TnsFNEq7RblpOd3aAoUbcdUTmC3BF12xFFpModyVL2QPp3wkdF7xYR5t1dO4wq6VwajcZF1FMO1z0uf_tArgfnnYbhF-7_J_4DdLLszg |
| CitedBy_id | crossref_primary_10_1002_mp_17580 crossref_primary_10_1088_1361_6560_accaca crossref_primary_10_3390_a18040233 crossref_primary_10_1016_j_zemedi_2023_05_001 crossref_primary_10_1177_09287329241306201 crossref_primary_10_1016_j_phro_2024_100648 crossref_primary_10_1002_mp_17732 crossref_primary_10_1016_j_phro_2023_100500 crossref_primary_10_1016_j_phro_2025_100794 crossref_primary_10_1016_j_phro_2023_100498 crossref_primary_10_1259_bjro_20230030 crossref_primary_10_1016_j_phro_2024_100551 crossref_primary_10_1016_j_phro_2024_100573 crossref_primary_10_1007_s00066_024_02277_9 crossref_primary_10_1016_j_phro_2025_100766 crossref_primary_10_1007_s13246_023_01249_0 crossref_primary_10_1140_epjp_s13360_024_05660_8 crossref_primary_10_1016_j_radonc_2024_110345 crossref_primary_10_1016_j_radonc_2024_110667 crossref_primary_10_3390_cancers15235629 crossref_primary_10_1186_s13014_023_02330_4 |
| Cites_doi | 10.1002/mp.12737 10.1186/1748-717X-7-160 10.1007/s10278-019-00227-x 10.1016/j.adro.2020.04.027 10.1016/j.phro.2022.02.008 10.1002/mp.13765 10.1002/mp.13994 10.3109/0284186X.2010.503662 10.1016/j.phro.2019.12.001 10.1109/icABCD51485.2021.9519338 10.1080/0284186X.2019.1620331 10.1109/TPAMI.2018.2858826 10.1117/1.JMI.7.6.064003 10.1109/TUFFC.2020.3015081 10.3390/diagnostics10110959 10.1002/mp.13200 10.1002/mp.13620 10.1016/j.radonc.2018.12.011 |
| ContentType | Journal Article |
| Copyright | 2022 The Authors 2022 The Authors. 2022 The Authors 2022 |
| Copyright_xml | – notice: 2022 The Authors – notice: 2022 The Authors. – notice: 2022 The Authors 2022 |
| DBID | 6I. AAFTH AAYXX CITATION 7X8 5PM ACNBI ADTPV AOWAS D8T DF2 ZZAVC DOA |
| DOI | 10.1016/j.phro.2022.06.001 |
| DatabaseName | ScienceDirect Open Access Titles Elsevier:ScienceDirect:Open Access CrossRef MEDLINE - Academic PubMed Central (Full Participant titles) SWEPUB Uppsala universitet full text SwePub SwePub Articles SWEPUB Freely available online SWEPUB Uppsala universitet SwePub Articles full text DOAJ Directory of Open Access Journals |
| DatabaseTitle | CrossRef MEDLINE - Academic |
| DatabaseTitleList | MEDLINE - Academic |
| Database_xml | – sequence: 1 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website – sequence: 2 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Medicine |
| EISSN | 2405-6316 |
| EndPage | 42 |
| ExternalDocumentID | oai_doaj_org_article_a9b1e80ae9fd4c14a06793ecc7df0d23 oai_DiVA_org_uu_482662 PMC9234226 10_1016_j_phro_2022_06_001 S2405631622000495 |
| GroupedDBID | .1- .FO 0R~ AAEDW AALRI AAXUO AAYWO ABMAC ACGFS ACVFH ADBBV ADCNI ADVLN AEUPX AFJKZ AFPUW AFRHN AFTJW AIGII AITUG AJUYK AKBMS AKRWK AKYEP ALMA_UNASSIGNED_HOLDINGS AMRAJ APXCP BCNDV EBS EJD FDB GROUPED_DOAJ M41 M~E O9- OK1 ROL RPM SSZ Z5R 0SF 6I. AACTN AAFTH NCXOZ RIG AAYXX CITATION 7X8 5PM ACNBI ADTPV AOWAS D8T DF2 ZZAVC |
| ID | FETCH-LOGICAL-c586t-fe62d2a9a8d4341d32d5cf8c52f56e51f07472274155daf24e696977afbae1473 |
| IEDL.DBID | DOA |
| ISICitedReferencesCount | 26 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000836498300006&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 2405-6316 |
| IngestDate | Fri Oct 03 12:32:15 EDT 2025 Tue Nov 04 17:01:48 EST 2025 Thu Aug 21 14:06:50 EDT 2025 Sun Nov 09 14:12:13 EST 2025 Tue Nov 18 21:59:23 EST 2025 Thu Nov 13 04:18:58 EST 2025 Thu Jul 20 20:10:51 EDT 2023 Tue Aug 26 17:19:20 EDT 2025 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Language | English |
| License | This is an open access article under the CC BY license. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c586t-fe62d2a9a8d4341d32d5cf8c52f56e51f07472274155daf24e696977afbae1473 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
| OpenAccessLink | https://doaj.org/article/a9b1e80ae9fd4c14a06793ecc7df0d23 |
| PMID | 35769110 |
| PQID | 2682785230 |
| PQPubID | 23479 |
| PageCount | 5 |
| ParticipantIDs | doaj_primary_oai_doaj_org_article_a9b1e80ae9fd4c14a06793ecc7df0d23 swepub_primary_oai_DiVA_org_uu_482662 pubmedcentral_primary_oai_pubmedcentral_nih_gov_9234226 proquest_miscellaneous_2682785230 crossref_citationtrail_10_1016_j_phro_2022_06_001 crossref_primary_10_1016_j_phro_2022_06_001 elsevier_sciencedirect_doi_10_1016_j_phro_2022_06_001 elsevier_clinicalkey_doi_10_1016_j_phro_2022_06_001 |
| PublicationCentury | 2000 |
| PublicationDate | 2022-07-01 |
| PublicationDateYYYYMMDD | 2022-07-01 |
| PublicationDate_xml | – month: 07 year: 2022 text: 2022-07-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationTitle | Physics and imaging in radiation oncology |
| PublicationYear | 2022 |
| Publisher | Elsevier B.V Elsevier |
| Publisher_xml | – name: Elsevier B.V – name: Elsevier |
| References | Lamb, Cao, Kishan, Agazaryan, Thomas, Shaverdian (b0010) 2017; 9 Jansen, Kuijf, Dhara, Weaver, Jan Biessels, Strand (b0080) 2020; 7 Woo, Lee (b0115) 2021 Deng, Dong, Socher, Li, Li, Fei-Fei (b0060) 2009 Zhang, Paulson, Lim, Hall, Ahunbay, Mickevicius (b0040) 2020; 5 Vaassen, Hazelaar, Vaniqui, Gooding, van der Heyden, Canters (b0105) 2020; 13 Milletari, Navab, Ahmadi (b0125) 2016 Wang, Rimner, Hu, Tyagi, Jiang, Yorke (b0130) 2019; 46 Karimi D, Warfield SK, Gholipour A. Critical Assessment of Transfer Learning for Medical Image Segmentation with Fully Convolutional Neural Networks. arXiv:2006.00356 [cs.CV] 2020;1–11. Klambauer, Unterthiner, Mayr, Hochreiter (b0085) 2017 Willigenburg, Zachiu, Lagendijk, van der Voort van Zyp, de Boer, Raaymakers (b0050) 2022; 21 Eppenhof, Maspero, Savenije, de Boer, van der Voort van Zyp, Raaymakers (b0045) 2020; 47 Kiljunen, Akram, Niemelä, Löyttyniemi, Seppälä, Heikkila (b0120) 2020; 10 Hesamian, Jia, He, Kennedy (b0055) 2019; 32 Shivdeo A, Lokwani R, Kulkarni V, Kharat A, Pant A. Comparative Evaluation of 3D and 2D Deep Learning Techniques for Semantic Segmentation in CT Scans. arXiv:2101.07612 [eess.IV] 2021. Rigaud, Simon, Castelli, Lafond, Acosta, Haigron (b0015) 2019; 58 Elmahdy, Ahuja, Van Der Heide, Staring (b0075) 2020 Elmahdy, Jagt, Zinkstok, Qiao, Shahzad, Sokooti (b0035) 2019; 46 Lin, Goyal, Girshick, He, Dollár (b0090) 2020; 42 Xie J, Xu B, Chuang Z. Horizontal and Vertical Ensemble with Deep Representation for Classification. arXiv:1306.2759 [cs.LG] 2013. Tijssen, Philippens, Paulson, Glitzner, Chugh, Wetscherek (b0005) 2019; 132 Thörnqvist, Petersen, Høyer, Bentzen, Muren (b0030) 2010; 49 Loi, Fusella, Lanzi, Cagni, Garibaldi, Iacoviello (b0020) 2018; 45 La Macchia, Fellin, Amichetti, Cianchetti, Gianolini, Paola (b0025) 2012; 7 Amiri, Brooks, Rivaz (b0065) 2020; 67 Gooding, Smith, Tariq, Aljabar, Peressutti, van der Stoep (b0100) 2018; 45 Woo (10.1016/j.phro.2022.06.001_b0115) 2021 Elmahdy (10.1016/j.phro.2022.06.001_b0035) 2019; 46 Hesamian (10.1016/j.phro.2022.06.001_b0055) 2019; 32 La Macchia (10.1016/j.phro.2022.06.001_b0025) 2012; 7 Tijssen (10.1016/j.phro.2022.06.001_b0005) 2019; 132 Rigaud (10.1016/j.phro.2022.06.001_b0015) 2019; 58 Milletari (10.1016/j.phro.2022.06.001_b0125) 2016 Gooding (10.1016/j.phro.2022.06.001_b0100) 2018; 45 10.1016/j.phro.2022.06.001_b0110 Kiljunen (10.1016/j.phro.2022.06.001_b0120) 2020; 10 Eppenhof (10.1016/j.phro.2022.06.001_b0045) 2020; 47 Willigenburg (10.1016/j.phro.2022.06.001_b0050) 2022; 21 Jansen (10.1016/j.phro.2022.06.001_b0080) 2020; 7 Elmahdy (10.1016/j.phro.2022.06.001_b0075) 2020 Vaassen (10.1016/j.phro.2022.06.001_b0105) 2020; 13 Lamb (10.1016/j.phro.2022.06.001_b0010) 2017; 9 Thörnqvist (10.1016/j.phro.2022.06.001_b0030) 2010; 49 Wang (10.1016/j.phro.2022.06.001_b0130) 2019; 46 Deng (10.1016/j.phro.2022.06.001_b0060) 2009 Lin (10.1016/j.phro.2022.06.001_b0090) 2020; 42 10.1016/j.phro.2022.06.001_b0095 10.1016/j.phro.2022.06.001_b0070 Amiri (10.1016/j.phro.2022.06.001_b0065) 2020; 67 Klambauer (10.1016/j.phro.2022.06.001_b0085) 2017 Zhang (10.1016/j.phro.2022.06.001_b0040) 2020; 5 Loi (10.1016/j.phro.2022.06.001_b0020) 2018; 45 |
| References_xml | – start-page: 972 year: 2017 end-page: 981 ident: b0085 article-title: Self-Normalizing Neural Networks publication-title: NIPS’17 Proc 31st Int Conf Neural Inf Process Syst – reference: Xie J, Xu B, Chuang Z. Horizontal and Vertical Ensemble with Deep Representation for Classification. arXiv:1306.2759 [cs.LG] 2013. – volume: 58 start-page: 1225 year: 2019 end-page: 1237 ident: b0015 article-title: Deformable image registration for radiation therapy: principle, methods, applications and evaluation publication-title: Acta Oncol – reference: Shivdeo A, Lokwani R, Kulkarni V, Kharat A, Pant A. Comparative Evaluation of 3D and 2D Deep Learning Techniques for Semantic Segmentation in CT Scans. arXiv:2101.07612 [eess.IV] 2021. – volume: 49 start-page: 1023 year: 2010 end-page: 1032 ident: b0030 article-title: Propagation of target and organ at risk contours in radiotherapy of prostate cancer using deformable image registration publication-title: Acta Oncol – start-page: 565 year: 2016 end-page: 571 ident: b0125 article-title: V-Net: Fully convolutional neural networks for volumetric medical image segmentation publication-title: 4th Int Conf 3D Vis – volume: 47 start-page: 1238 year: 2020 end-page: 1248 ident: b0045 article-title: Fast contour propagation for MR-guided prostate radiotherapy using convolutional neural networks publication-title: Med Phys – volume: 132 start-page: 114 year: 2019 end-page: 120 ident: b0005 article-title: MRI commissioning of 1.5T MR-linac systems – a multi-institutional study publication-title: Radiother Oncol – volume: 67 start-page: 2510 year: 2020 end-page: 2518 ident: b0065 article-title: Fine-tuning U-net for ultrasound image segmentation: different layers, different outcomes publication-title: IEEE Trans Ultrason Ferroelectr Freq Control – volume: 9 year: 2017 ident: b0010 article-title: Online adaptive radiation therapy: implementation of a new process of care publication-title: Cureus – volume: 45 start-page: 748 year: 2018 end-page: 757 ident: b0020 article-title: Performance of commercially available deformable image registration platforms for contour propagation using patient-based computational phantoms: A multi-institutional study publication-title: Med Phys – volume: 7 start-page: 1 year: 2012 end-page: 16 ident: b0025 article-title: Systematic evaluation of three different commercial software solutions for automatic segmentation for adaptive therapy in head-and-neck, prostate and pleural cancer publication-title: Radiat Oncol – year: 2021 ident: b0115 article-title: Comparison of tissue segmentation performance between 2D U-Net and 3D U-Net on brain MR Images publication-title: Int Conf Electron Information, Commun ICEIC 2021 – volume: 32 start-page: 582 year: 2019 end-page: 596 ident: b0055 article-title: Deep learning techniques for medical image segmentation: achievements and challenges publication-title: J Digit Imaging – volume: 42 start-page: 318 year: 2020 end-page: 327 ident: b0090 article-title: Focal loss for dense object detection publication-title: IEEE Trans Pattern Anal Mach Intell – volume: 7 start-page: 1 year: 2020 end-page: 15 ident: b0080 article-title: Patient-specific fine-tuning of convolutional neural networks for follow-up lesion quantification publication-title: J Med Imaging – volume: 21 start-page: 62 year: 2022 end-page: 65 ident: b0050 article-title: Fast and accurate deformable contour propagation for intra-fraction adaptive magnetic resonance-guided prostate radiotherapy publication-title: Phys Imaging Radiat Oncol – start-page: 577 year: 2020 end-page: 580 ident: b0075 article-title: Patient-specific finetuning of deep learning models for adaptive radiotherapy in prostate CT publication-title: Proc - Int Symp Biomed Imaging – volume: 10 year: 2020 ident: b0120 article-title: A deep learning-based automated CT segmentation of prostate cancer anatomy for radiation therapy planning-a retrospective multicenter study publication-title: Diagnostics – volume: 5 start-page: 1350 year: 2020 end-page: 1358 ident: b0040 article-title: A patient-specific autosegmentation strategy using multi-input deformable image registration for magnetic resonance imaging-guided online adaptive radiation therapy: a feasibility study publication-title: Adv Radiat Oncol – reference: Karimi D, Warfield SK, Gholipour A. Critical Assessment of Transfer Learning for Medical Image Segmentation with Fully Convolutional Neural Networks. arXiv:2006.00356 [cs.CV] 2020;1–11. – volume: 46 start-page: 4699 year: 2019 end-page: 4707 ident: b0130 article-title: Toward predicting the evolution of lung tumors during radiotherapy observed on a longitudinal MR imaging study via a deep learning algorithm publication-title: Med Phys – volume: 46 start-page: 3329 year: 2019 end-page: 3343 ident: b0035 article-title: Robust contour propagation using deep learning and image registration for online adaptive proton therapy of prostate cancer publication-title: Med Phys – volume: 13 start-page: 1 year: 2020 end-page: 6 ident: b0105 article-title: Evaluation of measures for assessing time-saving of automatic organ-at-risk segmentation in radiotherapy publication-title: Phys Imaging Radiat Oncol – start-page: 248 year: 2009 end-page: 255 ident: b0060 article-title: ImageNet: A large-scale hierarchical image database publication-title: IEEE Conf Comput Vis Pattern Recognit – volume: 45 start-page: 5105 year: 2018 end-page: 5115 ident: b0100 article-title: Comparative evaluation of autocontouring in clinical practice: a practical method using the Turing test publication-title: Med Phys – volume: 45 start-page: 748 year: 2018 ident: 10.1016/j.phro.2022.06.001_b0020 article-title: Performance of commercially available deformable image registration platforms for contour propagation using patient-based computational phantoms: A multi-institutional study publication-title: Med Phys doi: 10.1002/mp.12737 – volume: 7 start-page: 1 year: 2012 ident: 10.1016/j.phro.2022.06.001_b0025 article-title: Systematic evaluation of three different commercial software solutions for automatic segmentation for adaptive therapy in head-and-neck, prostate and pleural cancer publication-title: Radiat Oncol doi: 10.1186/1748-717X-7-160 – start-page: 248 year: 2009 ident: 10.1016/j.phro.2022.06.001_b0060 article-title: ImageNet: A large-scale hierarchical image database publication-title: IEEE Conf Comput Vis Pattern Recognit – volume: 32 start-page: 582 year: 2019 ident: 10.1016/j.phro.2022.06.001_b0055 article-title: Deep learning techniques for medical image segmentation: achievements and challenges publication-title: J Digit Imaging doi: 10.1007/s10278-019-00227-x – volume: 5 start-page: 1350 year: 2020 ident: 10.1016/j.phro.2022.06.001_b0040 article-title: A patient-specific autosegmentation strategy using multi-input deformable image registration for magnetic resonance imaging-guided online adaptive radiation therapy: a feasibility study publication-title: Adv Radiat Oncol doi: 10.1016/j.adro.2020.04.027 – ident: 10.1016/j.phro.2022.06.001_b0095 – volume: 21 start-page: 62 year: 2022 ident: 10.1016/j.phro.2022.06.001_b0050 article-title: Fast and accurate deformable contour propagation for intra-fraction adaptive magnetic resonance-guided prostate radiotherapy publication-title: Phys Imaging Radiat Oncol doi: 10.1016/j.phro.2022.02.008 – ident: 10.1016/j.phro.2022.06.001_b0070 – volume: 46 start-page: 4699 year: 2019 ident: 10.1016/j.phro.2022.06.001_b0130 article-title: Toward predicting the evolution of lung tumors during radiotherapy observed on a longitudinal MR imaging study via a deep learning algorithm publication-title: Med Phys doi: 10.1002/mp.13765 – volume: 47 start-page: 1238 year: 2020 ident: 10.1016/j.phro.2022.06.001_b0045 article-title: Fast contour propagation for MR-guided prostate radiotherapy using convolutional neural networks publication-title: Med Phys doi: 10.1002/mp.13994 – volume: 49 start-page: 1023 year: 2010 ident: 10.1016/j.phro.2022.06.001_b0030 article-title: Propagation of target and organ at risk contours in radiotherapy of prostate cancer using deformable image registration publication-title: Acta Oncol doi: 10.3109/0284186X.2010.503662 – year: 2021 ident: 10.1016/j.phro.2022.06.001_b0115 article-title: Comparison of tissue segmentation performance between 2D U-Net and 3D U-Net on brain MR Images – volume: 9 year: 2017 ident: 10.1016/j.phro.2022.06.001_b0010 article-title: Online adaptive radiation therapy: implementation of a new process of care publication-title: Cureus – volume: 13 start-page: 1 year: 2020 ident: 10.1016/j.phro.2022.06.001_b0105 article-title: Evaluation of measures for assessing time-saving of automatic organ-at-risk segmentation in radiotherapy publication-title: Phys Imaging Radiat Oncol doi: 10.1016/j.phro.2019.12.001 – ident: 10.1016/j.phro.2022.06.001_b0110 doi: 10.1109/icABCD51485.2021.9519338 – start-page: 565 year: 2016 ident: 10.1016/j.phro.2022.06.001_b0125 article-title: V-Net: Fully convolutional neural networks for volumetric medical image segmentation – volume: 58 start-page: 1225 year: 2019 ident: 10.1016/j.phro.2022.06.001_b0015 article-title: Deformable image registration for radiation therapy: principle, methods, applications and evaluation publication-title: Acta Oncol doi: 10.1080/0284186X.2019.1620331 – volume: 42 start-page: 318 year: 2020 ident: 10.1016/j.phro.2022.06.001_b0090 article-title: Focal loss for dense object detection publication-title: IEEE Trans Pattern Anal Mach Intell doi: 10.1109/TPAMI.2018.2858826 – volume: 7 start-page: 1 year: 2020 ident: 10.1016/j.phro.2022.06.001_b0080 article-title: Patient-specific fine-tuning of convolutional neural networks for follow-up lesion quantification publication-title: J Med Imaging doi: 10.1117/1.JMI.7.6.064003 – volume: 67 start-page: 2510 year: 2020 ident: 10.1016/j.phro.2022.06.001_b0065 article-title: Fine-tuning U-net for ultrasound image segmentation: different layers, different outcomes publication-title: IEEE Trans Ultrason Ferroelectr Freq Control doi: 10.1109/TUFFC.2020.3015081 – volume: 10 year: 2020 ident: 10.1016/j.phro.2022.06.001_b0120 article-title: A deep learning-based automated CT segmentation of prostate cancer anatomy for radiation therapy planning-a retrospective multicenter study publication-title: Diagnostics doi: 10.3390/diagnostics10110959 – volume: 45 start-page: 5105 year: 2018 ident: 10.1016/j.phro.2022.06.001_b0100 article-title: Comparative evaluation of autocontouring in clinical practice: a practical method using the Turing test publication-title: Med Phys doi: 10.1002/mp.13200 – volume: 46 start-page: 3329 year: 2019 ident: 10.1016/j.phro.2022.06.001_b0035 article-title: Robust contour propagation using deep learning and image registration for online adaptive proton therapy of prostate cancer publication-title: Med Phys doi: 10.1002/mp.13620 – start-page: 972 year: 2017 ident: 10.1016/j.phro.2022.06.001_b0085 article-title: Self-Normalizing Neural Networks – start-page: 577 year: 2020 ident: 10.1016/j.phro.2022.06.001_b0075 article-title: Patient-specific finetuning of deep learning models for adaptive radiotherapy in prostate CT publication-title: Proc - Int Symp Biomed Imaging – volume: 132 start-page: 114 year: 2019 ident: 10.1016/j.phro.2022.06.001_b0005 article-title: MRI commissioning of 1.5T MR-linac systems – a multi-institutional study publication-title: Radiother Oncol doi: 10.1016/j.radonc.2018.12.011 |
| SSID | ssj0002793530 |
| Score | 2.3347297 |
| Snippet | Treatments on combined Magnetic Resonance (MR) scanners and Linear Accelerators (Linacs) for radiotherapy, called MR-Linacs, often require daily contouring.... Background and Purpose: Treatments on combined Magnetic Resonance (MR) scanners and Linear Accelerators (Linacs) for radiotherapy, called MR-Linacs, often... |
| SourceID | doaj swepub pubmedcentral proquest crossref elsevier |
| SourceType | Open Website Open Access Repository Aggregation Database Enrichment Source Index Database Publisher |
| StartPage | 38 |
| SubjectTerms | Computerized Image Processing Datoriserad bildbehandling Original |
| Title | Patient specific deep learning based segmentation for magnetic resonance guided prostate radiotherapy |
| URI | https://www.clinicalkey.com/#!/content/1-s2.0-S2405631622000495 https://dx.doi.org/10.1016/j.phro.2022.06.001 https://www.proquest.com/docview/2682785230 https://pubmed.ncbi.nlm.nih.gov/PMC9234226 https://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-482662 https://doaj.org/article/a9b1e80ae9fd4c14a06793ecc7df0d23 |
| Volume | 23 |
| WOSCitedRecordID | wos000836498300006&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVAON databaseName: DOAJ Directory of Open Access Journals customDbUrl: eissn: 2405-6316 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002793530 issn: 2405-6316 databaseCode: DOA dateStart: 20170101 isFulltext: true titleUrlDefault: https://www.doaj.org/ providerName: Directory of Open Access Journals – providerCode: PRVHPJ databaseName: ROAD: Directory of Open Access Scholarly Resources customDbUrl: eissn: 2405-6316 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002793530 issn: 2405-6316 databaseCode: M~E dateStart: 20170101 isFulltext: true titleUrlDefault: https://road.issn.org providerName: ISSN International Centre |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1Nb9QwELWgQogLKl8ilFZGghMKrJ3EcY5tacWBVj0A2pvl-mNJ1WZXu5tKXPjtzNjOasOhvXDZw669kWcmmed43htC3ntWF0JLnQvPUVSb8bwRJThEiNowCxDaB8n8b_X5uZxOm4utVl9YExblgaPhPuvmkjk50a7xtjSs1Pjmo4AL19ZPLA86n5O62dpMXYXjtKaoQqMRyFhVLgomEmMmFneBoZD5x_mneBgxykpBvH-UnLbA57-lkyOB0ZCUTnfJ04Qm6WFcxTPywHXPyeOzdF7-griLKJtKkVCJRUHUOregqVXEjGIOs3TlZjeJgtRRALH0Rs86JDdS2IvPUZHD0VnfWhi6QJIIwFO61LZN5K3fL8mP05Pvx1_z1FghN5UU69w7wS3XjZa2hCxmC24r46WpuK-Eq5hHVX3OA9iw2vPSiUYAUNT-UjtW1sUrstPNO_eaUK0LwEh2YhrANiVsN4rSaWngEcu8B69nhA2GVSapjmPzi2s1lJddKXSGQmeoWGOXkY-bOYuouXHn6CP012Yk6mWHLyCKVIoidV8UZaQYvK0GSio8ROGP2jsvXW1mJcASgci9894NAaXgbsYjGt25eb9SXEheS3xTn5F6FGmj9Y1_6dpfQRccsDryojPyIcbkaMqX9udhMEnfqxL2lIK_-R-G2yNPcGGxgPkt2Vkve7dPHpnbdbtaHpCH9VQehFsTPs_-nPwFpBhACg |
| linkProvider | Directory of Open Access Journals |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Patient+specific+deep+learning+based+segmentation+for+magnetic+resonance+guided+prostate+radiotherapy&rft.jtitle=Physics+and+imaging+in+radiation+oncology&rft.au=Fransson%2C+Samuel&rft.au=Tilly%2C+David&rft.au=Strand%2C+Robin&rft.date=2022-07-01&rft.pub=Elsevier+B.V&rft.issn=2405-6316&rft.eissn=2405-6316&rft.volume=23&rft.spage=38&rft.epage=42&rft_id=info:doi/10.1016%2Fj.phro.2022.06.001&rft.externalDocID=S2405631622000495 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2405-6316&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2405-6316&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2405-6316&client=summon |