{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T13:24:20Z","timestamp":1787059460343,"version":"build-2736575974"},"reference-count":24,"publisher":"MDPI AG","issue":"8","license":[{"start":{"date-parts":[[2026,8,8]],"date-time":"2026-08-08T00:00:00Z","timestamp":1786147200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100007224","name":"National Foundation for Science and Technology Development","doi-asserted-by":"publisher","award":["NCUD.01-2025.32"],"award-info":[{"award-number":["NCUD.01-2025.32"]}],"id":[{"id":"10.13039\/100007224","id-type":"DOI","asserted-by":"publisher"}]},{"award":["NCUD.01-2025.32"],"award-info":[{"award-number":["NCUD.01-2025.32"]}],"id":[{"id":"https:\/\/ror.org\/04rw64z44","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J. Imaging"],"abstract":"<jats:p>Accurate segmentation of non-small cell lung cancer (NSCLC) on positron emission tomography\/computed tomography (PET\/CT) is an essential prerequisite for automated metabolic tumor volume (MTV) quantification and staging. Although deep learning models achieve high performance on large-scale datasets, their generalization across different clinical domains is limited by variations in imaging protocols and patient demographics. This study aims to evaluate several deep learning architectures and investigate a transfer learning strategy to mitigate domain shift. Three architectures, including ResNet-backbone 3D U-Net, nnU-Net v2, and Swin UNETR, were benchmarked from scratch and compared with a fine-tuned nnU-Net initialized with AutoPET II weights. Results on the internal dataset showed that the fine-tuned nnU-Net achieved a Dice similarity coefficient (DSC) of 83.4 \u00b1 6.5%, a 95% Hausdorff distance (HD95) of 5.1 \u00b1 3.6 mm, and a precision of 89.6 \u00b1 8.2%. Compared to the nnU-Net v2, the fine-tuned nnU-Net improved the absolute DSC by 6.8% while reducing local training time by 37.5% by bypassing the initial feature-learning phase. The fine-tuned nnU-Net model also demonstrated a high correlation between the MTV and the ground truth (Pearson r = 0.96, p &lt; 0.001), indicating its potential as a reliable automated approach for quantitative MTV extraction and NSCLC prognostic-related analysis.<\/jats:p>","DOI":"10.3390\/jimaging12080364","type":"journal-article","created":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T08:08:53Z","timestamp":1786349333000},"page":"364","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Benchmarking Deep Learning for NSCLC PET\/CT Segmentation on a Histologically Confirmed Vietnamese Dataset: Validation and Generalization"],"prefix":"10.3390","volume":"12","author":[{"given":"Quang Tuan","family":"Ho","sequence":"first","affiliation":[{"name":"Institute for Nuclear Science and Technology, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2272-4619","authenticated-orcid":false,"given":"Ngoc Ha","family":"Bui","sequence":"additional","affiliation":[{"name":"Faculty of Engineering Physics, Hanoi University of Science and Technology, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Thuy Duong","family":"Tran","sequence":"additional","affiliation":[{"name":"Faculty of Engineering Physics, Hanoi University of Science and Technology, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-2490-5362","authenticated-orcid":false,"given":"Quang Huy","family":"Khuat","sequence":"additional","affiliation":[{"name":"Faculty of Engineering Physics, Hanoi University of Science and Technology, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ngoc Toan","family":"Tran","sequence":"additional","affiliation":[{"name":"Institute for Nuclear Science and Technology, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuan Chung","family":"Le","sequence":"additional","affiliation":[{"name":"Institute for Nuclear Science and Technology, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Huu Quyet","family":"Nguyen","sequence":"additional","affiliation":[{"name":"Institute for Nuclear Science and Technology, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tat Thang","family":"Nguyen","sequence":"additional","affiliation":[{"name":"Faculty of Engineering Physics, Hanoi University of Science and Technology, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9352-8984","authenticated-orcid":false,"given":"Van Thai","family":"Nguyen","sequence":"additional","affiliation":[{"name":"Faculty of Engineering Physics, Hanoi University of Science and Technology, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dinh Thuy","family":"Mai","sequence":"additional","affiliation":[{"name":"Faculty of Engineering Physics, Hanoi University of Science and Technology, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Quang Duy","family":"To","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology and Radiosurgery, 108 Military Central Hospital, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9699-0675","authenticated-orcid":false,"given":"Dinh Chau","family":"Nguyen","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology and Radiosurgery, 108 Military Central Hospital, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nguyen Huong Giang","family":"Trinh","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology and Radiosurgery, 108 Military Central Hospital, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Van Chinh","family":"Cao","sequence":"additional","affiliation":[{"name":"Department of Medical Imaging Technology, Hanoi Medical College, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-9831-2928","authenticated-orcid":false,"given":"Tien Hung","family":"Bui","sequence":"additional","affiliation":[{"name":"Hong Ngoc-Phuc Truong Minh General Hospital, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Thu Trang","family":"Vu","sequence":"additional","affiliation":[{"name":"Faculty of English, Thuong Mai University, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Khac Nam","family":"Vo","sequence":"additional","affiliation":[{"name":"Nuclear Medicine Department, HCM Oncology Hospital, Ho Chi Minh City 700000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7096-3318","authenticated-orcid":false,"given":"Hai Quan","family":"Ho","sequence":"additional","affiliation":[{"name":"Viettel Institute for Strategic Technology, Viettel Group, Hanoi 100000, Vietnam"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,8,8]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"1237","DOI":"10.1109\/JBHI.2022.3186275","article-title":"Modality-specific segmentation network for lung tumor segmentation in PET-CT images","volume":"27","author":"Xiang","year":"2022","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Hershman, M., Yousefi, B., Serletti, L., Galperin-Aizenberg, M., Roshkovan, L., Luna, J.M., Thompson, J.C., Aggarwal, C., Carpenter, E.L., and Kontos, D. (2021). Impact of Interobserver Variability in Manual Segmentation of Non-Small Cell Lung Cancer (NSCLC) Applying Low-Rank Radiomic Representation on Computed Tomography. Cancers, 13.","DOI":"10.3390\/cancers13235985"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"27","DOI":"10.1007\/s00259-011-1934-6","article-title":"Prognostic value of metabolic tumor burden on 18F-FDG PET in nonsurgical patients with non-small cell lung cancer","volume":"39","author":"Liao","year":"2012","journal-title":"Eur. J. Nucl. Med. Mol. Imaging"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1038\/s41592-020-01008-z","article-title":"nnU-Net: A self-configuring method for deep learning-based biomedical image segmentation","volume":"18","author":"Isensee","year":"2021","journal-title":"Nat. Methods"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1396","DOI":"10.1038\/s42256-024-00912-9","article-title":"Results From the AutoPET Challenge on Fully Automated Lesion Segmentation in Oncologic PET\/CT Imaging","volume":"6","author":"Gatidis","year":"2024","journal-title":"Nat. Mach. Intell."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"481","DOI":"10.2967\/jnumed.125.270260","article-title":"AutoPET Challenge on Fully Automated Lesion Segmentation in Oncologic PET\/CT Imaging, Part 2: Domain Generalization","volume":"67","author":"Dexl","year":"2026","journal-title":"J. Nucl. Med."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"1173","DOI":"10.1109\/TBME.2021.3117407","article-title":"Domain adaptation for medical image analysis: A survey","volume":"69","author":"Guan","year":"2021","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1186\/s13550-025-01333-4","article-title":"Global and regional accuracy of deep learning-based tumor segmentation from whole-body [18F] fluorodeoxyglucose PET\/CT images","volume":"16","author":"Ciarmiello","year":"2026","journal-title":"EJNMMI Res."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"137","DOI":"10.55981\/aij.2025.1645","article-title":"Development of a Vietnamese PET\/CT Dataset for Machine Learning-Based Analysis of Non-Small Cell Lung Cancer Images","volume":"51","author":"Tuan","year":"2025","journal-title":"At. Indones."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"903","DOI":"10.1109\/TMI.2004.828354","article-title":"Simultaneous truth and performance level estimation (STAPLE): An algorithm for the validation of image segmentation","volume":"23","author":"Warfield","year":"2004","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"9473","DOI":"10.1088\/0031-9155\/60\/24\/9473","article-title":"Is STAPLE algorithm confident to assess segmentation methods in PET imaging?","volume":"60","author":"Betrouni","year":"2015","journal-title":"Phys. Med. Biol."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"2176","DOI":"10.1038\/s41591-021-01595-0","article-title":"Underdiagnosis bias of artificial intelligence algorithms applied to chest radiographs in under-served patient populations","volume":"27","author":"Zhang","year":"2021","journal-title":"Nat. Med."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Celi, L.A., Cellini, J., Charpignon, M.-L., Dee, E.C., Dernoncourt, F., Eber, R., Mitchell, W.G., Moukheiber, L., Schirmer, J., and Situ, J. (2022). Sources of bias in artificial intelligence that perpetuate healthcare disparities\u2014A global review. PLoS Digit. Health, 1.","DOI":"10.1371\/journal.pdig.0000022"},{"key":"ref_14","unstructured":"Li, P., Wang, S., Li, T., Lu, J., HuangFu, Y., and Wang, D. (2020). A Large-Scale CT and PET\/CT Dataset for Lung Cancer Diagnosis (Lung-PET-CT-Dx) [Data Set], The Cancer Imaging Archive."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"195","DOI":"10.1038\/s41592-023-02150-0","article-title":"Metrics reloaded: Recommendations for image analysis validation","volume":"21","author":"Reinke","year":"2024","journal-title":"Nat. Methods"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Isensee, F., Wald, T., Ulrich, C., Baumgartner, M., Roy, S., Maier-Hein, K., and Jaeger, P.F. (2024). nnU-net revisited: A call for rigorous validation in 3d medical image segmentation. International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer Nature.","DOI":"10.1007\/978-3-031-72114-4_47"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Hatamizadeh, A., Nath, V., Tang, Y., Yang, D., Roth, H.R., and Xu, D. (2021). Swin unetr: Swin transformers for semantic segmentation of brain tumors in MRI images. International MICCAI Brainlesion Workshop, Springer International Publishing.","DOI":"10.1007\/978-3-031-08999-2_22"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"\u00c7i\u00e7ek, \u00d6., Abdulkadir, A., Lienkamp, S.S., Brox, T., and Ronneberger, O. (2016). 3D U-Net: Learning dense volumetric segmentation from sparse annotation. International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer International Publishing.","DOI":"10.1007\/978-3-319-46723-8_49"},{"key":"ref_19","unstructured":"Cardoso, M.J., Li, W., Brown, R., Ma, N., Kerfoot, E., Wang, Y., Murrey, B., Myronenko, A., Zhao, C., and Yang, D. (2022). MONAI: An open-source framework for deep learning in healthcare. arXiv."},{"key":"ref_20","unstructured":"Isensee, F., and Maier-Hein, K.H. (2023). Look Ma, no code: Fine tuning nnU-Net for the AutoPET II challenge by only adjusting its JSON plans. arXiv."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"307","DOI":"10.1016\/S0140-6736(86)90837-8","article-title":"Statistical methods for assessing agreement between two methods of clinical measurement","volume":"1","author":"Bland","year":"1986","journal-title":"Lancet"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"771","DOI":"10.1016\/j.ijrobp.2006.12.067","article-title":"Pet-ct\u2013based auto-contouring in non\u2013small-cell lung cancer correlates with pathology and reduces interobserver variability in the delineation of the primary tumor and involved nodal volumes","volume":"68","author":"Bosmans","year":"2007","journal-title":"Int. J. Radiat. Oncol. Biol. Phys."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"182","DOI":"10.1016\/j.radonc.2005.09.017","article-title":"Observer variation in target volume delineation of lung cancer related to radiation oncologist\u2013computer interaction: A \u2018Big Brother\u2019evaluation","volume":"77","author":"Steenbakkers","year":"2005","journal-title":"Radiother. Oncol."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Wang, H., Chen, W., Luo, X., Xing, Z., Liu, L., Qin, J., Wu, S.C., and Zhu, L. (2025). Toward fair and accurate cross-domain medical image segmentation: A vlm-driven active domain adaptation paradigm. Proceedings of the IEEE\/CVF International Conference on Computer Vision, IEEE.","DOI":"10.1109\/ICCV51701.2025.02234"}],"container-title":["Journal of Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2313-433X\/12\/8\/364\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T12:27:52Z","timestamp":1787056072000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2313-433X\/12\/8\/364"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,8]]},"references-count":24,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2026,8]]}},"alternative-id":["jimaging12080364"],"URL":"https:\/\/doi.org\/10.3390\/jimaging12080364","relation":{},"ISSN":["2313-433X"],"issn-type":[{"value":"2313-433X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,8]]}}}