{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,13]],"date-time":"2026-08-13T23:26:54Z","timestamp":1786663614314,"version":"build-2736575974"},"reference-count":40,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T00:00:00Z","timestamp":1780444800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,8,14]],"date-time":"2026-08-14T00:00:00Z","timestamp":1786665600000},"content-version":"vor","delay-in-days":72,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"National Research Foundation of Korea (NRF) grants funded by the Korean government","award":["RS-2026-25495438"],"award-info":[{"award-number":["RS-2026-25495438"]}]},{"name":"National Research Foundation of Korea (NRF) grants funded by the Korean government","award":["RS-2025-00553579"],"award-info":[{"award-number":["RS-2025-00553579"]}]},{"name":"Creative-Pioneering Researchers Program funded by Seoul National University"},{"name":"by the AI-Bio Resarch Grant through Seoul National University"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Sci Rep"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Predicting spatial transcriptomics from histology images offers cost-effective insights but faces privacy barriers preventing cross-institutional data sharing. To address this, we present privateST, a homomorphic-encryption-optimized framework that substantiates the feasibility of secure spatial transcriptomics prediction. To accommodate the computational constraints of homomorphic encryption, we downsampled the input images using bilinear interpolation. To improve prediction accuracy for the 100 target genes, we incorporated predictions for an auxiliary set of 150 highly expressed genes. This approach enables the model to learn from a more diverse genomic context, thereby refining the feature representations for the high-priority target genes. To adapt the architecture for homomorphic encryption, we replaced Max-Pooling and ReLU with Average-Pooling and polynomial approximation, respectively. These modifications eliminate non-linear comparison operations, significantly reducing the multiplicative depth. We also implemented multiplexed packing to optimize the efficiency of encrypted data processing. Remarkably, our results demonstrate that despite the reduced input resolution, this proposed approach achieves accuracy comparable to the original ResNet-18 configurations without downsampling.<\/jats:p>","DOI":"10.1038\/s41598-026-55961-4","type":"journal-article","created":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T12:16:20Z","timestamp":1780488980000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["privateST: a feasible framework for privacy-preserving spatial transcriptomics prediction from histopathology images"],"prefix":"10.1038","volume":"16","author":[{"given":"Hakin","family":"Kim","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Miran","family":"Kim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Buhm","family":"Han","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,6,3]]},"reference":[{"key":"55961_CR1","doi-asserted-by":"publisher","first-page":"13604","DOI":"10.1038\/s41598-023-40219-0","volume":"13","author":"MM Rahaman","year":"2023","unstructured":"Rahaman, M. M., Millar, E. K. A. & Meijering, E. Breast cancer histopathology image-based gene expression prediction using spatial transcriptomics data and deep learning. Sci. Rep. 13, 13604. https:\/\/doi.org\/10.1038\/s41598-023-40219-0 (2023).","journal-title":"Sci. Rep."},{"key":"55961_CR2","doi-asserted-by":"publisher","first-page":"1544","DOI":"10.1038\/s41467-025-56618-y","volume":"16","author":"C Wang","year":"2025","unstructured":"Wang, C. et al. Benchmarking the translational potential of spatial gene expression prediction from histology. Nat. Commun. 16, 1544. https:\/\/doi.org\/10.1038\/s41467-025-56618-y (2025).","journal-title":"Nat. Commun."},{"key":"55961_CR3","doi-asserted-by":"publisher","unstructured":"Zeng, Y. et al. Spatial transcriptomics prediction from histology jointly through Transformer and graph neural networks. Brief. Bioinform. 23, 859. https:\/\/doi.org\/10.1093\/bib\/bbac297 (2022).","DOI":"10.1093\/bib\/bbac297"},{"key":"55961_CR4","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1016\/j.patcog.2023.109966","volume":"145","author":"Y Yang","year":"2024","unstructured":"Yang, Y., Hossain, M. Z., Stone, E. & Rahman, S. Spatial transcriptomics analysis of gene expression prediction using exemplar guided graph neural network. Pattern Recogn. 145, 10. https:\/\/doi.org\/10.1016\/j.patcog.2023.109966 (2024).","journal-title":"Pattern Recogn."},{"key":"55961_CR5","doi-asserted-by":"publisher","unstructured":"Jia, Y., Liu, J., Chen, L., Zhao, T. & Wang, Y. THItoGene: a deep learning method for predicting spatial transcriptomics from histological images. Brief. Bioinform. 25, 586. https:\/\/doi.org\/10.1093\/bib\/bbad464 (2023).","DOI":"10.1093\/bib\/bbad464"},{"key":"55961_CR6","doi-asserted-by":"publisher","first-page":"e12","DOI":"10.1055\/s-0041-1740630","volume":"61","author":"R Torkzadehmahani","year":"2022","unstructured":"Torkzadehmahani, R. et al. Privacy-preserving artificial intelligence techniques in biomedicine. Methods Inf. Med. 61, e12\u2013e27. https:\/\/doi.org\/10.1055\/s-0041-1740630 (2022).","journal-title":"Methods Inf. Med."},{"key":"55961_CR7","doi-asserted-by":"publisher","first-page":"103335","DOI":"10.1016\/j.media.2024.103335","volume":"99","author":"J Ganz","year":"2025","unstructured":"Ganz, J., Ammeling, J., Jabari, S., Breininger, K. & Aubreville, M. Re-identification from histopathology images. Med. Image. Anal. 99, 103335. https:\/\/doi.org\/10.1016\/j.media.2024.103335 (2025).","journal-title":"Med. Image. Anal."},{"key":"55961_CR8","doi-asserted-by":"publisher","first-page":"2577","DOI":"10.1038\/s41467-023-37991-y","volume":"14","author":"P Holub","year":"2023","unstructured":"Holub, P. et al. Privacy risks of whole-slide image sharing in digital pathology. Nat. Commun. 14, 2577. https:\/\/doi.org\/10.1038\/s41467-023-37991-y (2023).","journal-title":"Nat. Commun."},{"key":"55961_CR9","doi-asserted-by":"publisher","unstructured":"Gursoy, G., Chielle, E., Brannon, C. M., Maniatakos, M. & Gerstein, M. Privacy-preserving genotype imputation with fully homomorphic encryption. Cell Syst 13, 173\u2013182 e173. https:\/\/doi.org\/10.1016\/j.cels.2021.10.003 (2022).","DOI":"10.1016\/j.cels.2021.10.003"},{"key":"55961_CR10","doi-asserted-by":"publisher","unstructured":"Scheibner, J. et al. Revolutionizing medical data sharing using advanced privacy-enhancing technologies: technical, legal, and ethical synthesis. J. Med. Internet Res. 23, e25120\u2013e25120. https:\/\/doi.org\/10.2196\/25120 (2021).","DOI":"10.2196\/25120"},{"key":"55961_CR11","doi-asserted-by":"publisher","first-page":"313","DOI":"10.7326\/0003-4819-145-4-200608150-00019","volume":"145","author":"JF Wilson","year":"2006","unstructured":"Wilson, J. F. Health Insurance Portability and Accountability Act Privacy Rule causes ongoing concerns among clinicians and researchers. Ann. Intern. Med. 145, 313\u2013316. https:\/\/doi.org\/10.7326\/0003-4819-145-4-200608150-00019 (2006).","journal-title":"Ann. Intern. Med."},{"key":"55961_CR12","doi-asserted-by":"publisher","first-page":"65","DOI":"10.1080\/13600834.2019.1573501","volume":"28","author":"CJ Hoofnagle","year":"2019","unstructured":"Hoofnagle, C. J., van der Sloot, B. & Borgesius, F. Z. The European Union general data protection regulation: what it is and what it means*. Inform. Commun. Technol. Law. 28, 65\u201398. https:\/\/doi.org\/10.1080\/13600834.2019.1573501 (2019).","journal-title":"Inform. Commun. Technol. Law"},{"key":"55961_CR13","unstructured":"Cheon, J. H., Kim, A., Kim, M. & Song Y 409\u2013437 (Springer International Publishing, 2025)."},{"key":"55961_CR14","doi-asserted-by":"publisher","first-page":"34","DOI":"10.1007\/s00145-019-09319-x","volume":"33","author":"I Chillotti","year":"2020","unstructured":"Chillotti, I., Gama, N., Georgieva, M. & Izabach\u00e8ne, M. T. F. H. E. Fast fully homomorphic encryption over the Torus. J. Cryptol. 33, 34\u201391. https:\/\/doi.org\/10.1007\/s00145-019-09319-x (2020).","journal-title":"J. Cryptol."},{"key":"55961_CR15","unstructured":"Fan, J. & Vercauteren, F. Somewhat practical fully homomorphic encryption. IACR Cryptol. ePrint Arch. 144 (2012). (2012)."},{"key":"55961_CR16","unstructured":"Cheon, J. H., Kim, A., Kim, M. & Song, Y. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 409\u2013437 (2025)."},{"key":"55961_CR17","doi-asserted-by":"publisher","first-page":"btae343","DOI":"10.1093\/bioinformatics\/btae343","volume":"40","author":"X Chen","year":"2024","unstructured":"Chen, X. et al. HE2Gene: image-to-RNA translation via multi-task learning for spatial transcriptomics data. Bioinformatics 40, btae343. https:\/\/doi.org\/10.1093\/bioinformatics\/btae343 (2024).","journal-title":"Bioinformatics"},{"key":"55961_CR18","doi-asserted-by":"publisher","first-page":"3877","DOI":"10.1038\/s41467-020-17678-4","volume":"11","author":"B Schmauch","year":"2020","unstructured":"Schmauch, B. et al. A deep learning model to predict RNA-Seq expression of tumours from whole slide images. Nat. Commun. 11, 3877. https:\/\/doi.org\/10.1038\/s41467-020-17678-4 (2020).","journal-title":"Nat. Commun."},{"key":"55961_CR19","doi-asserted-by":"publisher","first-page":"e240798","DOI":"10.1148\/ryai.240798","volume":"7","author":"SW Lee","year":"2025","unstructured":"Lee, S. W. et al. Development of privacy-preserving deep learning model with homomorphic encryption: a technical feasibility study in kidney CT imaging. Radiology: Artif. Intell. 7, e240798. https:\/\/doi.org\/10.1148\/ryai.240798 (2025).","journal-title":"Radiology: Artif. Intell."},{"key":"55961_CR20","doi-asserted-by":"crossref","unstructured":"Ebel, A., Garimella, K. & Reagen, B. In Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2 734\u2013749 (Association for Computing Machinery, 2025).","DOI":"10.1145\/3676641.3716008"},{"key":"55961_CR21","doi-asserted-by":"publisher","first-page":"78","DOI":"10.1126\/science.aaf2403","volume":"353","author":"PL Stahl","year":"2016","unstructured":"Stahl, P. L. et al. Visualization and analysis of gene expression in tissue sections by spatial transcriptomics. Science 353, 78\u201382. https:\/\/doi.org\/10.1126\/science.aaf2403 (2016).","journal-title":"Science"},{"key":"55961_CR22","doi-asserted-by":"publisher","first-page":"R844","DOI":"10.1186\/bcr1312","volume":"7","author":"JP Sokol","year":"2005","unstructured":"Sokol, J. P., Neil, J. R., Schiemann, B. J. & Schiemann, W. P. The use of cystatin C to inhibit epithelial-mesenchymal transition and morphological transformation stimulated by transforming growth factor-beta. Breast Cancer Res. 7, R844\u2013853. https:\/\/doi.org\/10.1186\/bcr1312 (2005).","journal-title":"Breast Cancer Res."},{"key":"55961_CR23","doi-asserted-by":"publisher","first-page":"497","DOI":"10.1016\/j.cell.2020.05.039","volume":"182","author":"AL Ji","year":"2020","unstructured":"Ji, A. L. et al. Multimodal analysis of composition and spatial architecture in human squamous cell carcinoma. Cell 182, 497\u2013514e422. https:\/\/doi.org\/10.1016\/j.cell.2020.05.039 (2020).","journal-title":"Cell"},{"key":"55961_CR24","doi-asserted-by":"publisher","unstructured":"Kim, H., Hwang, I., Song, Y. & Han, B. A privacy-preserving HLA imputation method with homomorphic encryption. iScience 28, 2563. https:\/\/doi.org\/10.1016\/j.isci.2025.113442 (2025).","DOI":"10.1016\/j.isci.2025.113442"},{"key":"55961_CR25","doi-asserted-by":"publisher","first-page":"103386","DOI":"10.1016\/j.jisa.2022.103386","volume":"72","author":"J Zhou","year":"2023","unstructured":"Zhou, J. et al. Secure genotype imputation using homomorphic encryption. J. Inform. Secur. Appl. 72, 103386. https:\/\/doi.org\/10.1016\/j.jisa.2022.103386 (2023).","journal-title":"J. Inform. Secur. Appl."},{"key":"55961_CR26","doi-asserted-by":"publisher","unstructured":"Jung, W., Kim, S., Ahn, J. H., Cheon, J. H. & Lee, Y. Over 100x faster bootstrapping in fully homomorphic encryption through memory-centric optimization with GPUs. In IACR Transactions on Cryptographic Hardware and Embedded Systems 114\u2013148 (2021). https:\/\/doi.org\/10.46586\/tches.v2021.i4.114-148 (2021).","DOI":"10.46586\/tches.v2021.i4.114-148"},{"key":"55961_CR27","doi-asserted-by":"crossref","unstructured":"Moon, J., Yoo, D., Jiang, X. & Kim, M. THOR: Secure Transformer Inference with Homomorphic Encryption (Springer, 2025).","DOI":"10.1145\/3719027.3765150"},{"key":"55961_CR28","doi-asserted-by":"publisher","first-page":"1372","DOI":"10.1038\/s41587-023-02019-9","volume":"42","author":"D Zhang","year":"2024","unstructured":"Zhang, D. et al. Inferring super-resolution tissue architecture by integrating spatial transcriptomics with histology. Nat. Biotechnol. 42, 1372\u20131377. https:\/\/doi.org\/10.1038\/s41587-023-02019-9 (2024).","journal-title":"Nat. Biotechnol."},{"key":"55961_CR29","doi-asserted-by":"publisher","first-page":"131880","DOI":"10.1016\/j.eswa.2026.131880","volume":"316","author":"W Wu","year":"2026","unstructured":"Wu, W. et al. Contrastive prompt clustering for weakly supervised semantic segmentation. Expert Syst. Appl. 316, 131880. https:\/\/doi.org\/10.1016\/j.eswa.2026.131880 (2026).","journal-title":"Expert Syst. Appl."},{"key":"55961_CR30","doi-asserted-by":"crossref","unstructured":"Wu, W., Chen, Z., Huang, X., Ma, F. & Xiao, J. SynthSeg-Agents: Multi-Agent Synthetic Data Generation for Zero-Shot Weakly Supervised Semantic Segmentation (Springer, 2025).","DOI":"10.1016\/j.neucom.2025.131314"},{"key":"55961_CR31","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TCE.2025.3620527","volume":"PP","author":"W Wu","year":"2025","unstructured":"Wu, W. et al. Tag-enriched multi-attention with large language models for cross-domain sequential recommendation. IEEE Trans. Consum. Electron. PP, 1\u20131. https:\/\/doi.org\/10.1109\/TCE.2025.3620527 (2025).","journal-title":"IEEE Trans. Consum. Electron."},{"key":"55961_CR32","doi-asserted-by":"crossref","unstructured":"Cai, H., Wu, W., Chai, B. & Zhang, Y. Neural Information Processing. (eds. Mufti, M. et al.) 285\u2013299 (Springer, 2025).","DOI":"10.1007\/978-981-96-6951-6_20"},{"key":"55961_CR33","doi-asserted-by":"publisher","first-page":"13801","DOI":"10.1109\/TNNLS.2023.3271859","volume":"35","author":"S Han","year":"2024","unstructured":"Han, S. et al. Practical and robust federated learning with highly scalable regression training. IEEE Trans. Neural Networks Learn. Syst. 35, 13801\u201313815. https:\/\/doi.org\/10.1109\/TNNLS.2023.3271859 (2024).","journal-title":"IEEE Trans. Neural Networks Learn. Syst."},{"key":"55961_CR34","doi-asserted-by":"publisher","first-page":"305","DOI":"10.1038\/s42256-020-0186-1","volume":"2","author":"GA Kaissis","year":"2020","unstructured":"Kaissis, G. A., Makowski, M. R., R\u00fcckert, D. & Braren, R. F. Secure, privacy-preserving and federated machine learning in medical imaging. Nat. Mach. Intell. 2, 305\u2013311. https:\/\/doi.org\/10.1038\/s42256-020-0186-1 (2020).","journal-title":"Nat. Mach. Intell."},{"key":"55961_CR35","doi-asserted-by":"publisher","first-page":"119","DOI":"10.1038\/s41746-020-00323-1","volume":"3","author":"N Rieke","year":"2020","unstructured":"Rieke, N. et al. The future of digital health with federated learning. npj Digit. Med. 3, 119. https:\/\/doi.org\/10.1038\/s41746-020-00323-1 (2020).","journal-title":"npj Digit. Med."},{"key":"55961_CR36","doi-asserted-by":"publisher","unstructured":"Fei, S., Yan, Z., Ding, W. & Xie, H. Security vulnerabilities of SGX and countermeasures: a survey. ACM Comput. Surv. 54, 126. https:\/\/doi.org\/10.1145\/3456631 (2021).","DOI":"10.1145\/3456631"},{"key":"55961_CR37","doi-asserted-by":"publisher","first-page":"e1013526","DOI":"10.1371\/journal.pcbi.1013526","volume":"21","author":"R Yuan","year":"2025","unstructured":"Yuan, R. et al. Harmony-based data integration for distributed single-cell multi-omics data. PLoS Comput. Biol. 21, e1013526. https:\/\/doi.org\/10.1371\/journal.pcbi.1013526 (2025).","journal-title":"PLoS Comput. Biol."},{"key":"55961_CR38","first-page":"626","volume":"2020","author":"S Soni","year":"2020","unstructured":"Soni, S. & Roberts, K. Paraphrasing to improve the performance of Electronic Health Records Question Answering. AMIA Jt. Summits Transl Sci. Proc. 2020, 626\u2013635 (2020).","journal-title":"AMIA Jt. Summits Transl Sci. Proc."},{"key":"55961_CR39","doi-asserted-by":"publisher","first-page":"1962","DOI":"10.1109\/tmi.2016.2529665","volume":"35","author":"A Vahadane","year":"2016","unstructured":"Vahadane, A. et al. Structure-preserving color normalization and sparse stain separation for histological images. IEEE Trans. Med. Imaging. 35, 1962\u20131971. https:\/\/doi.org\/10.1109\/tmi.2016.2529665 (2016).","journal-title":"IEEE Trans. Med. Imaging"},{"key":"55961_CR40","doi-asserted-by":"publisher","first-page":"347","DOI":"10.1007\/978-3-030-10970-7_16","volume":"11349","author":"JH Cheon","year":"2018","unstructured":"Cheon, J. H., Han, K., Kim, A., Kim, M. & Song, Y. A full RNS variant of approximate homomorphic encryption. Sel. Areas Cryptogr. 11349, 347\u2013368. https:\/\/doi.org\/10.1007\/978-3-030-10970-7_16 (2018).","journal-title":"Sel. Areas Cryptogr"}],"container-title":["Scientific Reports"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-55961-4","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-55961-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-55961-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,13]],"date-time":"2026-08-13T22:31:45Z","timestamp":1786660305000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-55961-4"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,3]]},"references-count":40,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["55961"],"URL":"https:\/\/doi.org\/10.1038\/s41598-026-55961-4","relation":{},"ISSN":["2045-2322"],"issn-type":[{"value":"2045-2322","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,3]]},"assertion":[{"value":"3 March 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"28 May 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"3 June 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"Buhm Han is the CEO of SpintoAI Inc.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Declarartions"}},{"value":"During the preparation of this work the authors used Gemini in order to refine the wording. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.","order":2,"name":"Ethics","label":"Use of generative AI","group":{"name":"EthicsHeading","label":"Declarartions"}}],"article-number":"25330"}}