{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T23:58:06Z","timestamp":1771977486618,"version":"3.50.1"},"reference-count":21,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,1,25]],"date-time":"2026-01-25T00:00:00Z","timestamp":1769299200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T00:00:00Z","timestamp":1771891200000},"content-version":"vor","delay-in-days":30,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100009567","name":"Budapest University of Technology and Economics","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100009567","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["EJNMMI Phys"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>\n                      <jats:bold>Background<\/jats:bold>\n                    <\/jats:title>\n                    <jats:p>In positron emission tomography (PET), gamma photons arriving at the detector ring may undergo one or more Compton scattering events, potentially reaching a different scintillation crystal than its initial interaction point. This phenomenon, known as inter-crystal scattering (ICS), can lead to incorrect line of response (LOR) assignments, introducing spatial blurring and noise in the reconstructed image. Existing ICS correction methods include energy-based heuristics, Compton kinematics, Monte Carlo simulation, and deep learning to recover the first interaction position of the photons in the detector. These corrections are performed in the LOR domain, since that is where ICS manifests. However, in ordered-subset expectation-maximization (OS-EM) reconstruction, only a subset of LORs is processed at each iteration, making LOR-domain corrections difficult to apply on-the-fly.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>\n                      <jats:bold>Methods<\/jats:bold>\n                    <\/jats:title>\n                    <jats:p>We propose a novel ICS correction technique performed entirely in the image domain. A neural network is trained to predict spatially-varying 3D filter kernels that model the blurring effect of ICS across the image volume. These kernels are applied to the PET image estimate prior to each OS-EM forward projection. We evaluated our method on simulated and real PET data using two network variants: one predicting full 3D kernels (ICS-Net-direct), and another using a compact skew normal representation (ICS-Net-skewnorm).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>\n                      <jats:bold>Results<\/jats:bold>\n                    <\/jats:title>\n                    <jats:p>Both ICS-Net-direct and ICS-Net-skewnorm significantly improved the spatial resolution and the contrast of the output image, while also being computationally efficient. We found that ICS-Net-skewnorm is better suited for structural, symmetric reconstruction objects, while ICS-Net-direct performs best in complex, real-world scenarios.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>\n                      <jats:bold>Conclusions<\/jats:bold>\n                    <\/jats:title>\n                    <jats:p>The proposed image-domain ICS correction technique enables efficient and effective compensation of inter-crystal scattering for OS-EM reconstruction. Network selection should be guided by the target imaging scenario to maximize performance.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1186\/s40658-026-00837-9","type":"journal-article","created":{"date-parts":[[2026,1,25]],"date-time":"2026-01-25T04:52:50Z","timestamp":1769316770000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Image-space compensation of inter-crystal scattering in PET using a neural network based filter"],"prefix":"10.1186","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9220-5868","authenticated-orcid":false,"given":"D\u00f3ra","family":"Varny\u00fa","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"\u00c1kos","family":"R\u00e1bely","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"L\u00e1szl\u00f3","family":"Szirmay-Kalos","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2026,1,25]]},"reference":[{"key":"837_CR1","doi-asserted-by":"publisher","first-page":"1769","DOI":"10.1007\/s00415-012-6428-3","volume":"259","author":"M Politis","year":"2012","unstructured":"Politis M, Piccini P. Positron emission tomography imaging in neurological disorders. J Neurol. 2012;259:1769\u201380. https:\/\/doi.org\/10.1007\/s00415-012-6428-3.","journal-title":"J Neurol"},{"key":"837_CR2","doi-asserted-by":"publisher","first-page":"IV55","DOI":"10.1007\/s003920070063","volume":"89","author":"HR Schelbert","year":"2000","unstructured":"Schelbert HR. Positron emission tomography and the changing paradigm in coronary artery disease. Clin Res Cardiol. 2000;89:IV55\u201360. https:\/\/doi.org\/10.1007\/s003920070063.","journal-title":"Clin Res Cardiol."},{"issue":"6","key":"837_CR3","doi-asserted-by":"publisher","first-page":"1243","DOI":"10.1007\/s13534-024-00427-7","volume":"14","author":"MS Lee","year":"2024","unstructured":"Lee MS, Seok H, Lee JS. Strategies for mitigating inter-crystal scattering effects in positron emission tomography: a comprehensive review. Biomed Eng Lett. 2024;14(6):1243\u201358. https:\/\/doi.org\/10.1007\/s13534-024-00427-7.","journal-title":"Biomed Eng Lett."},{"issue":"4","key":"837_CR4","doi-asserted-by":"publisher","first-page":"2213","DOI":"10.1109\/23.531884","volume":"43","author":"KA Comanor","year":"1996","unstructured":"Comanor KA, Virador PRG, Moses WW. Algorithms to identify detector Compton scatter in PET modules. IEEE Trans Nucl Sci. 1996;43(4):2213\u20138. https:\/\/doi.org\/10.1109\/23.531884.","journal-title":"IEEE Trans Nucl Sci"},{"issue":"7","key":"837_CR5","doi-asserted-by":"publisher","first-page":"821","DOI":"10.1088\/0031-9155\/48\/7\/302","volume":"48","author":"M Rafecas","year":"2003","unstructured":"Rafecas M, B\u00f6ning G, Pichler BJ, Lorenz E, Schwaiger M, Ziegler SI. Inter-crystal scatter in a dual layer, high resolution LSO-APD positron emission Tomograph. Phys Med Biol. 2003;48(7):821. https:\/\/doi.org\/10.1088\/0031-9155\/48\/7\/302.","journal-title":"Phys Med Biol"},{"issue":"6Part1","key":"837_CR6","doi-asserted-by":"publisher","first-page":"3049","DOI":"10.1118\/1.4950719","volume":"43","author":"N Gross-Weege","year":"2016","unstructured":"Gross-Weege N, Schug D, Hallen P, Schulz V. Maximum likelihood positioning algorithm for high-resolution PET scanners. Med Phys. 2016;43(6Part1):3049\u201361. https:\/\/doi.org\/10.1118\/1.4950719.","journal-title":"Med Phys"},{"key":"837_CR7","doi-asserted-by":"crossref","unstructured":"Szirmay-Kalos L, Varny\u00fa D, Magdics M, T\u00f3th B. Modeling of depth of interaction with inter-crystal scattering for PET reconstruction. In: 2020 IEEE nuclear science symposium and medical imaging conference (NSS\/MIC); 2020. p. 1\u20136.","DOI":"10.1109\/NSS\/MIC42677.2020.9507795"},{"issue":"1","key":"837_CR8","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1109\/TNS.2014.2372788","volume":"62","author":"JB Michaud","year":"2015","unstructured":"Michaud JB, T\u00e9trault MA, Beaudoin JF, Cadorette J, Leroux JD, Brunet CA, et al. Sensitivity increase through a neural network method for LOR recovery of ICS triple coincidences in high-resolution pixelated- detectors PET scanners. IEEE Trans Nucl Sci. 2015;62(1):82\u201394. https:\/\/doi.org\/10.1109\/TNS.2014.2372788.","journal-title":"IEEE Trans Nucl Sci"},{"issue":"18","key":"837_CR9","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ac215d","volume":"66","author":"S Lee","year":"2021","unstructured":"Lee S, Lee JS. Inter-crystal scattering recovery of light-sharing PET detectors using convolutional neural networks. Phys Med Biol. 2021;66(18):185004. https:\/\/doi.org\/10.1088\/1361-6560\/ac215d.","journal-title":"Phys Med Biol"},{"issue":"4","key":"837_CR10","doi-asserted-by":"publisher","first-page":"601","DOI":"10.1109\/42.363108","volume":"13","author":"HM Hudson","year":"1994","unstructured":"Hudson HM, Larkin RS. Accelerated image reconstruction using ordered subsets of projection data. IEEE Trans Med Imaging. 1994;13(4):601\u20139. https:\/\/doi.org\/10.1109\/42.363108.","journal-title":"IEEE Trans Med Imaging"},{"issue":"2","key":"837_CR11","first-page":"171","volume":"12","author":"A Azzalini","year":"1985","unstructured":"Azzalini A. A class of distributions which includes the normal ones. Scand J Stat. 1985;12(2):171\u20138.","journal-title":"Scand J Stat"},{"key":"837_CR12","unstructured":"Mediso Medical Imaging Systems.: nanoScan\u00ae PET\/CT. Accessed: 2025-05-04. https:\/\/mediso.com\/global\/en\/product\/pre-clinical-products\/nanoscanr-petct."},{"key":"837_CR13","unstructured":"Girshick R. Fast R-CNN. Available from: https:\/\/arxiv.org\/abs\/1504.08083."},{"key":"837_CR14","unstructured":"Paszke Aea. PyTorch: An Imperative Style, High-Performance Deep Learning Library. In: Advances in Neural Information Processing Systems 32. Curran Associates, Inc.; 2019. p. 8024\u20138035. Available from: http:\/\/papers.neurips.cc\/paper\/9015-pytorch-an-imperative-style-high-performance-deep-learning-library.pdf."},{"key":"837_CR15","unstructured":"Misra D. Mish: A Self Regularized Non-Monotonic Activation Function. Available from: https:\/\/arxiv.org\/abs\/1908.08681."},{"key":"837_CR16","unstructured":"Agarap AF. Deep Learning using Rectified Linear Units (ReLU). Available from: https:\/\/arxiv.org\/abs\/1803.08375."},{"key":"837_CR17","doi-asserted-by":"publisher","first-page":"546","DOI":"10.1109\/TNS.1986.4337166","volume":"33","author":"SE Derenzo","year":"1986","unstructured":"Derenzo SE. Mathematical removal of positron range blurring in high resolution tomography. IEEE Trans Nucl Sci. 1986;33:546\u20139.","journal-title":"IEEE Trans Nucl Sci"},{"key":"837_CR18","doi-asserted-by":"publisher","first-page":"4543","DOI":"10.1088\/0031-9155\/49\/19\/007","volume":"49","author":"S Jan","year":"2004","unstructured":"Jan S, Santin G, Strul D, Staelens S, Assi\u00e9 K, Autret D, et al. GATE: a simulation toolkit for PET and SPECT. Phys Med Biol. 2004;49:4543\u201361. https:\/\/doi.org\/10.1088\/0031-9155\/49\/19\/007.","journal-title":"Phys Med Biol"},{"issue":"2","key":"837_CR19","doi-asserted-by":"publisher","first-page":"40","DOI":"10.1145\/1365490.1365500","volume":"6","author":"J Nickolls","year":"2008","unstructured":"Nickolls J, Buck I, Garland M, Skadron K. Scalable parallel programming with CUDA: is CUDA the parallel programming model that application developers have been waiting for? Queue. 2008;6(2):40\u201353. https:\/\/doi.org\/10.1145\/1365490.1365500.","journal-title":"Queue"},{"key":"837_CR20","unstructured":"Liu L, Jiang H, He P, Chen W, Liu X, Gao J, et\u00a0al. On the variance of the adaptive learning rate and beyond. Available from: https:\/\/arxiv.org\/abs\/1908.03265."},{"key":"837_CR21","unstructured":"Magdics M, Szirmay-Kalos L. Crystal Scattering Simulation for PET on the GPU. In: Avis N, Lefebvre S, editors. Eurographics 2011 - Short Papers. The Eurographics Association; 2011."}],"container-title":["EJNMMI Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s40658-026-00837-9","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s40658-026-00837-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s40658-026-00837-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T23:03:32Z","timestamp":1771974212000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1186\/s40658-026-00837-9"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,25]]},"references-count":21,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["837"],"URL":"https:\/\/doi.org\/10.1186\/s40658-026-00837-9","relation":{},"ISSN":["2197-7364"],"issn-type":[{"value":"2197-7364","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,25]]},"assertion":[{"value":"15 November 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"6 January 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"25 January 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical approval"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare that they have no conflict of interest.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"18"}}