Automatic segmentation for magnetic resonance imaging guided individual elective lymph node irradiation in head and neck cancer patients
Background and purpose: In head and neck squamous cell carcinoma (HNSCC) patients, the radiation dose to nearby organs at risk can be reduced by restricting elective neck irradiation from lymph node levels to individual lymph nodes. However, manual delineation of every individual lymph node is time-...
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-10-01
|
| Series: | Physics and Imaging in Radiation Oncology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405631624001258 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846115989005008896 |
|---|---|
| author | Floris C.J. Reinders Mark H.F. Savenije Mischa de Ridder Matteo Maspero Patricia A.H. Doornaert Chris H.J. Terhaard Cornelis P.J. Raaijmakers Kaveh Zakeri Nancy Y. Lee Eric Aliotta Aneesh Rangnekar Harini Veeraraghavan Marielle E.P. Philippens |
| author_facet | Floris C.J. Reinders Mark H.F. Savenije Mischa de Ridder Matteo Maspero Patricia A.H. Doornaert Chris H.J. Terhaard Cornelis P.J. Raaijmakers Kaveh Zakeri Nancy Y. Lee Eric Aliotta Aneesh Rangnekar Harini Veeraraghavan Marielle E.P. Philippens |
| author_sort | Floris C.J. Reinders |
| collection | DOAJ |
| description | Background and purpose: In head and neck squamous cell carcinoma (HNSCC) patients, the radiation dose to nearby organs at risk can be reduced by restricting elective neck irradiation from lymph node levels to individual lymph nodes. However, manual delineation of every individual lymph node is time-consuming and error prone. Therefore, automatic magnetic resonance imaging (MRI) segmentation of individual lymph nodes was developed and tested using a convolutional neural network (CNN). Materials and methods: In 50 HNSCC patients (UMC-Utrecht), individual lymph nodes located in lymph node levels Ib-II-III-IV-V were manually segmented on MRI by consensus of two experts, obtaining ground truth segmentations. A 3D CNN (nnU-Net) was trained on 40 patients and tested on 10. Evaluation metrics were Dice Similarity Coefficient (DSC), recall, precision, and F1-score. The segmentations of the CNN was compared to segmentations of two observers. Transfer learning was used with 20 additional patients to re-train and test the CNN in another medical center. Results: nnU-Net produced automatic segmentations of elective lymph nodes with median DSC: 0.72, recall: 0.76, precision: 0.78, and F1-score: 0.78. The CNN had higher recall compared to both observers (p = 0.002). No difference in evaluation scores of the networks in both medical centers was found after re-training with 5 or 10 patients. Conclusion: nnU-Net was able to automatically segment individual lymph nodes on MRI. The detection rate of lymph nodes using nnU-Net was higher than manual segmentations. Re-training nnU-Net was required to successfully transfer the network to the other medical center. |
| format | Article |
| id | doaj-art-dec802e08b024a0a8001d0b685dc9a42 |
| institution | Kabale University |
| issn | 2405-6316 |
| language | English |
| publishDate | 2024-10-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Physics and Imaging in Radiation Oncology |
| spelling | doaj-art-dec802e08b024a0a8001d0b685dc9a422024-12-19T10:55:38ZengElsevierPhysics and Imaging in Radiation Oncology2405-63162024-10-0132100655Automatic segmentation for magnetic resonance imaging guided individual elective lymph node irradiation in head and neck cancer patientsFloris C.J. Reinders0Mark H.F. Savenije1Mischa de Ridder2Matteo Maspero3Patricia A.H. Doornaert4Chris H.J. Terhaard5Cornelis P.J. Raaijmakers6Kaveh Zakeri7Nancy Y. Lee8Eric Aliotta9Aneesh Rangnekar10Harini Veeraraghavan11Marielle E.P. Philippens12Department of Radiotherapy, University Medical Centre Utrecht, the Netherlands; Corresponding author.Department of Radiotherapy, University Medical Centre Utrecht, the Netherlands; Computational Imaging Group for MR Therapy and Diagnostics, Cancer and Imaging Division, University Medical Center Utrecht, Utrecht, the NetherlandsDepartment of Radiotherapy, University Medical Centre Utrecht, the NetherlandsDepartment of Radiotherapy, University Medical Centre Utrecht, the Netherlands; Computational Imaging Group for MR Therapy and Diagnostics, Cancer and Imaging Division, University Medical Center Utrecht, Utrecht, the NetherlandsDepartment of Radiotherapy, University Medical Centre Utrecht, the NetherlandsDepartment of Radiotherapy, University Medical Centre Utrecht, the NetherlandsDepartment of Radiotherapy, University Medical Centre Utrecht, the NetherlandsDepartment of Radiotherapy, Memorial Sloan Kettering Cancer Centre, New York, United StatesDepartment of Radiotherapy, Memorial Sloan Kettering Cancer Centre, New York, United StatesDepartment of Radiotherapy, Memorial Sloan Kettering Cancer Centre, New York, United StatesDepartment of Radiotherapy, Memorial Sloan Kettering Cancer Centre, New York, United StatesDepartment of Radiotherapy, Memorial Sloan Kettering Cancer Centre, New York, United StatesDepartment of Radiotherapy, University Medical Centre Utrecht, the NetherlandsBackground and purpose: In head and neck squamous cell carcinoma (HNSCC) patients, the radiation dose to nearby organs at risk can be reduced by restricting elective neck irradiation from lymph node levels to individual lymph nodes. However, manual delineation of every individual lymph node is time-consuming and error prone. Therefore, automatic magnetic resonance imaging (MRI) segmentation of individual lymph nodes was developed and tested using a convolutional neural network (CNN). Materials and methods: In 50 HNSCC patients (UMC-Utrecht), individual lymph nodes located in lymph node levels Ib-II-III-IV-V were manually segmented on MRI by consensus of two experts, obtaining ground truth segmentations. A 3D CNN (nnU-Net) was trained on 40 patients and tested on 10. Evaluation metrics were Dice Similarity Coefficient (DSC), recall, precision, and F1-score. The segmentations of the CNN was compared to segmentations of two observers. Transfer learning was used with 20 additional patients to re-train and test the CNN in another medical center. Results: nnU-Net produced automatic segmentations of elective lymph nodes with median DSC: 0.72, recall: 0.76, precision: 0.78, and F1-score: 0.78. The CNN had higher recall compared to both observers (p = 0.002). No difference in evaluation scores of the networks in both medical centers was found after re-training with 5 or 10 patients. Conclusion: nnU-Net was able to automatically segment individual lymph nodes on MRI. The detection rate of lymph nodes using nnU-Net was higher than manual segmentations. Re-training nnU-Net was required to successfully transfer the network to the other medical center.http://www.sciencedirect.com/science/article/pii/S2405631624001258Deep learningArtificial intelligenceLymph nodesMagnetic resonance imagingRadiotherapySquamous cell carcinoma of head and neck |
| spellingShingle | Floris C.J. Reinders Mark H.F. Savenije Mischa de Ridder Matteo Maspero Patricia A.H. Doornaert Chris H.J. Terhaard Cornelis P.J. Raaijmakers Kaveh Zakeri Nancy Y. Lee Eric Aliotta Aneesh Rangnekar Harini Veeraraghavan Marielle E.P. Philippens Automatic segmentation for magnetic resonance imaging guided individual elective lymph node irradiation in head and neck cancer patients Physics and Imaging in Radiation Oncology Deep learning Artificial intelligence Lymph nodes Magnetic resonance imaging Radiotherapy Squamous cell carcinoma of head and neck |
| title | Automatic segmentation for magnetic resonance imaging guided individual elective lymph node irradiation in head and neck cancer patients |
| title_full | Automatic segmentation for magnetic resonance imaging guided individual elective lymph node irradiation in head and neck cancer patients |
| title_fullStr | Automatic segmentation for magnetic resonance imaging guided individual elective lymph node irradiation in head and neck cancer patients |
| title_full_unstemmed | Automatic segmentation for magnetic resonance imaging guided individual elective lymph node irradiation in head and neck cancer patients |
| title_short | Automatic segmentation for magnetic resonance imaging guided individual elective lymph node irradiation in head and neck cancer patients |
| title_sort | automatic segmentation for magnetic resonance imaging guided individual elective lymph node irradiation in head and neck cancer patients |
| topic | Deep learning Artificial intelligence Lymph nodes Magnetic resonance imaging Radiotherapy Squamous cell carcinoma of head and neck |
| url | http://www.sciencedirect.com/science/article/pii/S2405631624001258 |
| work_keys_str_mv | AT floriscjreinders automaticsegmentationformagneticresonanceimagingguidedindividualelectivelymphnodeirradiationinheadandneckcancerpatients AT markhfsavenije automaticsegmentationformagneticresonanceimagingguidedindividualelectivelymphnodeirradiationinheadandneckcancerpatients AT mischaderidder automaticsegmentationformagneticresonanceimagingguidedindividualelectivelymphnodeirradiationinheadandneckcancerpatients AT matteomaspero automaticsegmentationformagneticresonanceimagingguidedindividualelectivelymphnodeirradiationinheadandneckcancerpatients AT patriciaahdoornaert automaticsegmentationformagneticresonanceimagingguidedindividualelectivelymphnodeirradiationinheadandneckcancerpatients AT chrishjterhaard automaticsegmentationformagneticresonanceimagingguidedindividualelectivelymphnodeirradiationinheadandneckcancerpatients AT cornelispjraaijmakers automaticsegmentationformagneticresonanceimagingguidedindividualelectivelymphnodeirradiationinheadandneckcancerpatients AT kavehzakeri automaticsegmentationformagneticresonanceimagingguidedindividualelectivelymphnodeirradiationinheadandneckcancerpatients AT nancyylee automaticsegmentationformagneticresonanceimagingguidedindividualelectivelymphnodeirradiationinheadandneckcancerpatients AT ericaliotta automaticsegmentationformagneticresonanceimagingguidedindividualelectivelymphnodeirradiationinheadandneckcancerpatients AT aneeshrangnekar automaticsegmentationformagneticresonanceimagingguidedindividualelectivelymphnodeirradiationinheadandneckcancerpatients AT hariniveeraraghavan automaticsegmentationformagneticresonanceimagingguidedindividualelectivelymphnodeirradiationinheadandneckcancerpatients AT marielleepphilippens automaticsegmentationformagneticresonanceimagingguidedindividualelectivelymphnodeirradiationinheadandneckcancerpatients |