{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,17]],"date-time":"2026-10-17T00:40:41Z","timestamp":1792197641726,"version":"4.4.0"},"reference-count":50,"publisher":"Springer Science and Business Media LLC","license":[{"start":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T00:00:00Z","timestamp":1792108800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T00:00:00Z","timestamp":1792108800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100003125","name":"Suomen Kulttuurirahasto","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100003125","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003125","name":"Suomen Kulttuurirahasto","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100003125","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Valtion tutkimusrahasto"},{"DOI":"10.13039\/100018472","name":"Juhani Ahon L\u00e4\u00e4ketieteen Tutkimuss\u00e4\u00e4ti\u00f6","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100018472","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["J Digit Imaging. Inform. med."],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Accurate assessment of different types of prostate cancer (PCa) lesions is a crucial part of treatment planning. Following a computed tomography (CT) or a positron emission tomography (PET)\/CT imaging scan, a significant amount of time is required from the clinicians to detect lesions and characterize them by their anatomical locations. To facilitate the development of new image analysis applications, we investigated the potential of automated classification of PCa lesions between bone metastases, lymph node metastases, local lesions, and other PCa lesions in CT images via machine learning (ML) and deep learning models. We annotated a total of 4949 PCa lesions by their type in a private dataset of\n                    <jats:inline-formula>\n                      <jats:alternatives>\n                        <jats:tex-math>$$^{18}$$<\/jats:tex-math>\n                        <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                          <mml:mmultiscripts>\n                            <mml:mrow\/>\n                            <mml:mrow\/>\n                            <mml:mn>18<\/mml:mn>\n                          <\/mml:mmultiscripts>\n                        <\/mml:math>\n                      <\/jats:alternatives>\n                    <\/jats:inline-formula>\n                    F-prostate-specific membrane antigen (PSMA) PET\/CT images of 532 PCa patients and an external test set of publicly available PSMA PET\/CT images from 30 patients. Using information of coordinates of the detected lesions, we automatically extracted 15 different features from the CT and the results of the automated 3D anatomical segmentation tool TotalSegmentator for each lesion, and trained several traditional ML models to classify lesions by their anatomical type. Additionally, we experimented with convolutional neural networks (CNNs) of different dimensionalities with CT and TotalSegmentator inputs. According to our results, the most accurate model was XGBoost with an overall accuracy of 96.0%. It identified bone metastases, lymph node metastases, and local lesions with sensitivities of 97.9\u00b11.2%, 95.2\u00b11.3%, and 93.1\u00b12.8% in the five-fold cross-validation experiments of the primary data and 96.3%, 94.0%, and 80.0% on the external test set, respectively. Conversely, the CNNs underperformed the best ML models in distinguishing local lesions. To conclude, a simple ML model can accurately classify PCa lesions between bone, lymph node, and local lesions based on CT images, lesion locations, and automated anatomical segmentation data, but more research is required for distinguishing PCa lesions outside these categories.\n                  <\/jats:p>","DOI":"10.1007\/s10278-026-02402-3","type":"journal-article","created":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T23:56:43Z","timestamp":1792195003000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Investigation on Automatic Classification of Bone, Lymph Node, and Local Lesions in CT Images of Prostate Cancer Patients"],"prefix":"10.1007","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7775-7656","authenticated-orcid":false,"given":"Oona","family":"Rainio","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-4375-2861","authenticated-orcid":false,"given":"Ahti","family":"Vehmanen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-2170-9188","authenticated-orcid":false,"given":"Anting","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7199-0062","authenticated-orcid":false,"given":"Maria K.","family":"Jaakkola","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9390-1198","authenticated-orcid":false,"given":"Jukka","family":"Kemppainen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0982-8360","authenticated-orcid":false,"given":"Riku","family":"Kl\u00e9n","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,10,16]]},"reference":[{"key":"2402_CR1","unstructured":"J. Ferlay et al. Global Cancer Observatory: Cancer Today (version 1.1). 2024."},{"key":"2402_CR2","doi-asserted-by":"crossref","unstructured":"R. Raychaudhuri, D. W. Lin, and R. B. Montgomery. \u201cProstate cancer: a review\". In: Jama 333.16 (2025), pp. 1433\u20131446.","DOI":"10.1001\/jama.2025.0228"},{"key":"2402_CR3","doi-asserted-by":"crossref","unstructured":"W. Xie et al. \u201cMetastasis-free survival is a strong surrogate of overall survival in localized prostate cancer\". In: Journal of Clinical Oncology 35.27 (2017), pp. 3097\u2013 3104.","DOI":"10.1200\/JCO.2017.73.9987"},{"key":"2402_CR4","unstructured":"S. Halabi et al. \u201cMeta-analysis evaluating the impact of site of metastasis on overall survival in men with castration-resistant prostate cancer\". In: Journal of clinical oncology 34.14 (2016), pp. 1652\u20131659."},{"key":"2402_CR5","doi-asserted-by":"crossref","unstructured":"S. Buelens et al. \u201cImportance of metastatic volume in prognostic models to predict survival in newly diagnosed metastatic prostate cancer\". In: World Journal of Urology 37.12 (2019), pp. 2565\u20132571.","DOI":"10.1007\/s00345-018-2449-6"},{"key":"2402_CR6","doi-asserted-by":"crossref","unstructured":"T. Otani et al. \u201cPSMA PET\/CT imaging and its application to prostate cancer treatment\". In: Japanese journal of radiology 43.1 (2025), pp. 1\u201312.","DOI":"10.1007\/s11604-024-01646-9"},{"key":"2402_CR7","doi-asserted-by":"crossref","unstructured":"Z. Chen, X. Chen, and R.Wang. \u201cApplication of SPECT and PET\/CT with computeraided diagnosis in bone metastasis of prostate cancer: a review\". In: Cancer Imaging 22.1 (2022), p. 18.","DOI":"10.1186\/s40644-022-00456-4"},{"key":"2402_CR8","doi-asserted-by":"crossref","unstructured":"I. Bhattacharya et al. \u201cA review of artificial intelligence in prostate cancer detection on imaging\". In: Therapeutic advances in urology 14 (2022), p. 17562872221128791.","DOI":"10.1177\/17562872221128791"},{"key":"2402_CR9","doi-asserted-by":"crossref","unstructured":"B. G\u00fclmez. \u201cConvolutional neural networks for prostate cancer detection, classification, and segmentation: A systematic review and bibliometric analysis\". In: European Journal of Radiology Open 16 (2026), p. 100741.","DOI":"10.1016\/j.ejro.2026.100741"},{"key":"2402_CR10","doi-asserted-by":"crossref","unstructured":"O. Rainio and R. Kl\u00e9n. \u201cConvolutional neural networks for tumor segmentation by cancer type and imaging modality: a systematic review\". In: Network Modeling Analysis in Health Informatics and Bioinformatics 14.1 (2025), p. 58.","DOI":"10.1007\/s13721-025-00556-8"},{"key":"2402_CR11","doi-asserted-by":"crossref","unstructured":"S. Iqbal et al. \u201cProstate cancer detection using deep learning and traditional techniques\". In: IEEE Access 9 (2021), pp. 27085\u201327100.","DOI":"10.1109\/ACCESS.2021.3057654"},{"key":"2402_CR12","doi-asserted-by":"crossref","unstructured":"N. Papandrianos et al. \u201cBone metastasis classification using whole body images from prostate cancer patients based on convolutional neural networks application\". In: PloS one 15.8 (2020), e0237213.","DOI":"10.1371\/journal.pone.0237213"},{"key":"2402_CR13","doi-asserted-by":"crossref","unstructured":"O. Rainio et al. \u201cNew method of using a convolutional neural network for 2D intraprostatic tumor segmentation from PET images\". In: Research on biomedical engineering 39.4 (2023), pp. 905\u2013913.","DOI":"10.1007\/s42600-023-00314-7"},{"key":"2402_CR14","doi-asserted-by":"crossref","unstructured":"L. A. Matkovic et al. \u201cProstate and dominant intraprostatic lesion segmentation on PET\/CT using cascaded regional-net\". In: Physics in Medicine & Biology 66.24 (2021), p. 245006.","DOI":"10.1088\/1361-6560\/ac3c13"},{"key":"2402_CR15","doi-asserted-by":"crossref","unstructured":"E. Tr\u00e4g\u00e5rdh et al. \u201cA fully automated AI-based method for tumour detection and quantification on [18F] PSMA-1007 PET\u2013CT images in prostate cancer\". In: EJN- MMI physics 12.1 (2025), p. 78.","DOI":"10.1186\/s40658-025-00786-9"},{"key":"2402_CR16","doi-asserted-by":"crossref","unstructured":"O. J. Pellicer-Valero et al. \u201cDeep learning for fully automatic detection, segmentation, and Gleason grade estimation of prostate cancer in multiparametric magnetic resonance images\". In: Scientific reports 12.1 (2022), p. 2975.","DOI":"10.1038\/s41598-022-06730-6"},{"key":"2402_CR17","doi-asserted-by":"crossref","unstructured":"R. N. Kandalan et al. \u201cDose prediction with deep learning for prostate cancer radiation therapy: model adaptation to different treatment planning practices\". In: Radiotherapy and Oncology 153 (2020), pp. 228\u2013235.","DOI":"10.1016\/j.radonc.2020.10.027"},{"key":"2402_CR18","doi-asserted-by":"crossref","unstructured":"H. Abdollahi et al. \u201cMachine learning-based radiomic models to predict intensitymodulated radiation therapy response, Gleason score and stage in prostate cancer\". In: La radiologia medica 124.6 (2019), pp. 555\u2013567.","DOI":"10.1007\/s11547-018-0966-4"},{"key":"2402_CR19","doi-asserted-by":"crossref","unstructured":"L. G. Kerkmeijer et al. \u201cPREDICT: model for prediction of survival in localized prostate cancer\". In: World journal of urology 34.6 (2016), pp. 789\u2013795.","DOI":"10.1007\/s00345-015-1691-4"},{"key":"2402_CR20","doi-asserted-by":"crossref","unstructured":"A. Hartenstein et al. \u201cProstate cancer nodal staging: using deep learning to predict 68Ga-PSMA-positivity from CT imaging alone\". In: Scientific reports 10.1 (2020), p. 3398.","DOI":"10.1038\/s41598-020-60311-z"},{"key":"2402_CR21","doi-asserted-by":"crossref","unstructured":"S. Masoudi et al. \u201cDeep learning based staging of bone lesions from computed tomography scans\". In: IEEE access 9 (2021), pp. 87531\u201387542.","DOI":"10.1109\/ACCESS.2021.3074051"},{"key":"2402_CR22","doi-asserted-by":"crossref","unstructured":"M. J. Belue et al. \u201cDeep learning-based detection and classification of bone lesions on staging computed tomography in prostate cancer: a development study\". In: Academic radiology 31.6 (2024), pp. 2424\u20132433.","DOI":"10.1016\/j.acra.2024.01.009"},{"key":"2402_CR23","doi-asserted-by":"crossref","unstructured":"Y. Zhao et al. \u201cDeep neural network for automatic characterization of lesions on 68Ga-PSMA PET\/CT images\". In: 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEEE. 2019, pp. 951\u2013 954.","DOI":"10.1109\/EMBC.2019.8857955"},{"key":"2402_CR24","doi-asserted-by":"crossref","unstructured":"Y. Zhao et al. \u201cDeep neural network for automatic characterization of lesions on 68Ga-PSMA-11 PET\/CT\". In: European journal of nuclear medicine and molecular imaging 47.3 (2020), pp. 603\u2013613.","DOI":"10.1007\/s00259-019-04606-y"},{"key":"2402_CR25","doi-asserted-by":"crossref","unstructured":"A. Toosi et al. \u201cComputer-Aided Detection (CADe) of Small Metastatic Prostate Cancer Lesions on 3D PSMA PET Volumes Using Multi-Angle Maximum Intensity Projections\". In: Cancers 17.9 (2025), p. 1563.","DOI":"10.3390\/cancers17091563"},{"key":"2402_CR26","doi-asserted-by":"crossref","unstructured":"E. Tr\u00e4g\u00e5rdh et al. \u201cFreely available, fully automated ai-based analysis of primary tumour and metastases of prostate cancer in whole-body [18f]-psma-1007 pet-ct\". In: Diagnostics 12.9 (2022), p. 2101.","DOI":"10.3390\/diagnostics12092101"},{"key":"2402_CR27","doi-asserted-by":"crossref","unstructured":"N. Nickols et al. \u201caPROMISE: a novel automated PROMISE platform to standardize evaluation of tumor burden in 18F-DCFPyL images of veterans with prostate cancer\". In: Journal of Nuclear Medicine 63.2 (2022), pp. 233\u2013239.","DOI":"10.2967\/jnumed.120.261863"},{"key":"2402_CR28","doi-asserted-by":"crossref","unstructured":"B. Schott et al. \u201cComparison of automated full-body bone metastases delineation methods and their corresponding prognostic power\". In: Physics in Medicine & Biology 68.3 (2023), p. 035011.","DOI":"10.1088\/1361-6560\/acaf22"},{"key":"2402_CR29","doi-asserted-by":"crossref","unstructured":"M. Anttinen et al. \u201cA Prospective Comparison of 18F-prostate-specific Membrane Antigen-1007 Positron Emission Tomography Computed Tomography, Whole-body 1.5 T Magnetic Resonance Imaging with Difusion-weighted Imaging, and Singlephoton Emission Computed Tomography\/Computed Tomography with Traditional Imaging in Primary Distant Metastasis Staging of Prostate Cancer (PROSTAGE)\". In: European Urology Oncology 4.4 (2021), pp. 635\u2013644.","DOI":"10.1016\/j.euo.2020.06.012"},{"key":"2402_CR30","doi-asserted-by":"crossref","unstructured":"K. Jeblick et al. A whole-body PSMA-PET\/CT dataset with manually annotated tumor lesions (Version 2) [dataset]. 2026.","DOI":"10.1038\/s41597-026-07821-z"},{"key":"2402_CR31","doi-asserted-by":"crossref","unstructured":"O. Rainio et al. \u201cCarimas: an extensive medical imaging data processing tool for research\". In: Journal of Digital Imaging 36.4 (2023), pp. 1885\u20131893.","DOI":"10.1007\/s10278-023-00812-1"},{"key":"2402_CR32","doi-asserted-by":"crossref","unstructured":"R. Seifert et al. \u201cSecond version of the prostate cancer molecular imaging standardized evaluation framework including response evaluation for clinical trials (PROMISE V2)\". In: European urology 83.5 (2023), pp. 405\u2013412.","DOI":"10.1016\/j.eururo.2023.02.002"},{"key":"2402_CR33","doi-asserted-by":"crossref","unstructured":"J. Wasserthal et al. \u201cTotalSegmentator: robust segmentation of 104 anatomic structures in CT images\". In: Radiology: Artificial Intelligence 5.5 (2023), e230024.","DOI":"10.1148\/ryai.230024"},{"key":"2402_CR34","doi-asserted-by":"publisher","unstructured":"O. Rainio et al. \u201cDeep learning for automatic detection of prostatectomy in CT images of prostate cancer patients\". In: Physical and Engineering Sciences in Medicine (2026). https:\/\/doi.org\/10.1007\/s13246-026-01801-8.","DOI":"10.1007\/s13246-026-01801-8."},{"key":"2402_CR35","unstructured":"G. Van Rossum and F. L. Drake. Python 3 Reference Manual. Scotts Valley, CA: CreateSpace, 2009. isbn: 1441412697."},{"key":"2402_CR36","unstructured":"F. Pedregosa et al. \u201cScikit-learn: Machine Learning in Python\". In: Journal of Ma- chine Learning Research 12 (2011), pp. 2825\u20132830."},{"key":"2402_CR37","unstructured":"M. Abadi et al. TensorFlow: Large-scale machine learning on heterogeneous systems. 2015."},{"key":"2402_CR38","unstructured":"F. Chollet et al. Keras. https:\/\/github.com\/fchollet\/keras. 2015."},{"key":"2402_CR39","doi-asserted-by":"crossref","unstructured":"O. Ronneberger, P. Fischer, and T. Brox. \u201cU-net: Convolutional networks for biomedical image segmentation\". In: International Conference on Medical image computing and computer-assisted intervention. Springer. 2015, pp. 234\u2013241","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"2402_CR40","doi-asserted-by":"crossref","unstructured":"O. Rainio et al. \u201cDeep learning for medical ultrasound image segmentation: A systematic review of the current research\". In: Journal of Imaging Informatics in Medicine (2026), pp. 1\u201314.","DOI":"10.1007\/s10278-026-01931-1"},{"key":"2402_CR41","doi-asserted-by":"crossref","unstructured":"H. Hellstr\u00f6m et al. \u201cClassification of head and neck cancer from PET images using convolutional neural networks\". In: Scientific Reports 13.1 (2023), p. 10528.","DOI":"10.1038\/s41598-023-37603-1"},{"key":"2402_CR42","doi-asserted-by":"crossref","unstructured":"O. Rainio and R. Kl\u00e9n. \u201cComparison of simple augmentation transformations for a convolutional neural network classifying medical images\". In: Signal, image and video processing 18.4 (2024), pp. 3353\u20133360.","DOI":"10.1007\/s11760-024-02998-5"},{"key":"2402_CR43","doi-asserted-by":"crossref","unstructured":"O. Rainio et al. \u201cDeep learning detects retropharyngeal edema on MRI in patients with acute neck infections\". In: European Radiology Experimental 9.1 (2025), p. 60.","DOI":"10.1186\/s41747-025-00599-6"},{"key":"2402_CR44","unstructured":"D. Gopstein. clust.bin.pair: Statistical Methods for Analyzing Clustered Matched Pair Data. R package version 0.1.2. 2018. https:\/\/CRAN.R-project.org\/package=clust.bin.pair."},{"key":"2402_CR45","unstructured":"R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. Vienna, Austria, 2024. https:\/\/www.R-project.org\/."},{"key":"2402_CR46","doi-asserted-by":"crossref","unstructured":"M. K. Buyyounouski et al. \u201cProstate cancer\u2013major changes in the American Joint Committee on Cancer eighth edition cancer staging manual\". In: CA: a cancer jour- nal for clinicians 67.3 (2017), pp. 245\u2013253.","DOI":"10.3322\/caac.21391"},{"key":"2402_CR47","doi-asserted-by":"crossref","unstructured":"S. Jeong et al. \u201cDeep-learning-based segmentation using individual patient data on prostate cancer radiation therapy\". In: Plos one 19.7 (2024), e0308181.","DOI":"10.1371\/journal.pone.0308181"},{"key":"2402_CR48","doi-asserted-by":"crossref","unstructured":"G. Palazzo et al. \u201cReal-world validation of artificial intelligence-based computed tomography auto-contouring for prostate cancer radiotherapy planning\". In: Physics and Imaging in Radiation Oncology 28 (2023), p. 100501.","DOI":"10.1016\/j.phro.2023.100501"},{"key":"2402_CR49","doi-asserted-by":"crossref","unstructured":"E. Tr\u00e4g\u00e5rdh et al. \u201cRECOMIA\u2014a cloud-based platform for artificial intelligence research in nuclear medicine and radiology\". In: EJNMMI physics 7.1 (2020), p. 51.","DOI":"10.1186\/s40658-020-00316-9"},{"key":"2402_CR50","doi-asserted-by":"crossref","unstructured":"S. Gonz\u00e1lez Izard et al. \u201cNextmed: automatic imaging segmentation, 3D reconstruction, and 3D model visualization platform using augmented and virtual reality\". In: Sensors 20.10 (2020), p. 2962.","DOI":"10.3390\/s20102962"}],"container-title":["Journal of Imaging Informatics in Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10278-026-02402-3.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s10278-026-02402-3","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10278-026-02402-3.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T23:56:47Z","timestamp":1792195007000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s10278-026-02402-3"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,16]]},"references-count":50,"alternative-id":["2402"],"URL":"https:\/\/doi.org\/10.1007\/s10278-026-02402-3","relation":{},"ISSN":["2948-2933"],"issn-type":[{"value":"2948-2933","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,16]]},"assertion":[{"value":"30 June 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"1 October 2026","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"2 October 2026","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 October 2026","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"A subset of the data was from an earlier prospective study, during which the patients consented to the research use of their data. In accordance with Finnish legislation, the requirement for individual informed consent was waived for the rest of patients due to the use of retrospective, register-based data. The research was approved by the Ethics Committee of the Hospital District of Southwest Finland.","order":1,"name":"Ethics","label":"Ethics Approval and Consent to Participate","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":2,"name":"Ethics","label":"Conflict of Interest","group":{"name":"EthicsHeading","label":"Declarations"}}]}}