{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,21]],"date-time":"2026-07-21T23:02:56Z","timestamp":1784674976316,"version":"3.55.0"},"reference-count":39,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,5,12]],"date-time":"2026-05-12T00:00:00Z","timestamp":1778544000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,7,22]],"date-time":"2026-07-22T00:00:00Z","timestamp":1784678400000},"content-version":"vor","delay-in-days":71,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["npj Syst Biol Appl"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Rare tumor diseases are difficult to diagnose and there is a lack of routine diagnostic procedures. Approaches must be found that allow comprehensive identification and evaluation of prognostic relevant target proteins or transcripts. Analyzing rare ampullary cancer, respectively, to their prognosis and predictive factors by machine learning (ML) based matrix-assisted laser desorption\/ionization (MALDI) time-of-flight (TOF) imaging is a first step towards providing new solutions for diagnostics of those cancer samples. In this study, we investigated a cohort of ampullary adenocarcinomas, including intestinal, pancreatic and cases of unknown subtypes, to identify differences in the proteome. Human formalin-fixed paraffin-embedded (FFPE) tissues were pathologically assessed, immunohistological stained, MALDI Imaging detected, and ML-related analyzed. We enable MALDI imaging as a diagnostic complement for immunohistochemical analysis and provide a MALDI Imaging neural network for broad application in tumor diagnostics. Moreover, using tools from ML model explainability, we determined a small subset of influential\n                    <jats:italic>m\/z<\/jats:italic>\n                    -values from the trained models. The transformation of locally established ML networks dependent on one proteomic application source to other similar application sources (without peak picking or other pre-processing) is the basis for future rare cancer patient data collection.\n                  <\/jats:p>","DOI":"10.1038\/s41540-026-00705-3","type":"journal-article","created":{"date-parts":[[2026,5,12]],"date-time":"2026-05-12T04:54:37Z","timestamp":1778561677000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Interpreting MALDI imaging data for rare types of ampullary cancer using machine learning"],"prefix":"10.1038","volume":"12","author":[{"given":"Patrick M.","family":"Jensen","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jan","family":"Lellmann","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christian","family":"Sperling","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Frieder","family":"Meier","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marius","family":"Distler","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniela E.","family":"Aust","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Herbert","family":"Thiele","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pia","family":"H\u00f6nscheid","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,5,12]]},"reference":[{"key":"705_CR1","doi-asserted-by":"publisher","first-page":"210","DOI":"10.1016\/j.jamcollsurg.2008.01.028","volume":"207","author":"JT Carter","year":"2008","unstructured":"Carter, J. T., Grenert, J. P., Rubenstein, L., Stewart, L. & Way, L. W. Tumors of the ampulla of Vater: histopathologic classification and predictors of survival. J. Am. Coll. Surg. 207, 210\u2013218 (2008).","journal-title":"J. Am. Coll. Surg."},{"key":"705_CR2","doi-asserted-by":"publisher","first-page":"430","DOI":"10.1245\/s10434-012-2603-0","volume":"20","author":"A Westgaard","year":"2013","unstructured":"Westgaard, A., Pomianowska, E., Clausen, O. P. F. & Gladhaug, I. P. Intestinal-type and pancreatobiliary-type adenocarcinomas: how does ampullary carcinoma differ from other periampullary malignancies?. Ann. Surg. Oncol. 20, 430\u2013439 (2013).","journal-title":"Ann. Surg. Oncol."},{"key":"705_CR3","doi-asserted-by":"publisher","first-page":"87","DOI":"10.1097\/00000658-199807000-00013","volume":"228","author":"JR Howe","year":"1998","unstructured":"Howe, J. R., Klimstra, D. S., Moccia, R. D., Conlon, K. C. & Brennan, M. F. Factors predictive of survival in ampullary carcinoma. Ann. Surg. 228, 87\u201394 (1998).","journal-title":"Ann. Surg."},{"key":"705_CR4","doi-asserted-by":"publisher","first-page":"266","DOI":"10.1002\/jso.22090","volume":"105","author":"WS Kim","year":"2012","unstructured":"Kim, W. S. et al. Clinical significance of pathologic subtype in curatively resected ampulla of Vater cancer. J. Surg. Oncol. 105, 266\u2013272 (2012).","journal-title":"J. Surg. Oncol."},{"key":"705_CR5","doi-asserted-by":"publisher","first-page":"598","DOI":"10.1002\/jso.21374","volume":"100","author":"J Albores-Saavedra","year":"2009","unstructured":"Albores-Saavedra, J., Schwartz, A. M., Batich, K. & Henson, D. E. Cancers of the ampulla of Vater: demographics, morphology, and survival based on 5,625 cases from the SEER program. J. Surg. Oncol. 100, 598\u2013605 (2009).","journal-title":"J. Surg. Oncol."},{"key":"705_CR6","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-019-49179-w","volume":"9","author":"C Zimmermann","year":"2019","unstructured":"Zimmermann, C. et al. The pathohistological subtype strongly predicts survival in patients with ampullary carcinoma. Sci. Rep. 9, 12676 (2019).","journal-title":"Sci. Rep."},{"key":"705_CR7","doi-asserted-by":"publisher","first-page":"6996","DOI":"10.3390\/curroncol30070507","volume":"30","author":"G Nappo","year":"2023","unstructured":"Nappo, G. et al. Ampullary cancer: histological subtypes, markers, and clinical behaviour-state of the art and perspectives. Curr. Oncol. 30, 6996\u20137006 (2023).","journal-title":"Curr. Oncol."},{"key":"705_CR8","doi-asserted-by":"publisher","first-page":"100181","DOI":"10.1016\/j.modpat.2023.100181","volume":"36","author":"J Kawata","year":"2023","unstructured":"Kawata, J. et al. Clinicopathologic features and genetic alterations in mixed-type ampullary carcinoma. Mod. Pathol. 36, 100181 (2023).","journal-title":"Mod. Pathol."},{"key":"705_CR9","doi-asserted-by":"publisher","first-page":"858","DOI":"10.1016\/j.bbapap.2016.11.018","volume":"1865","author":"M Kriegsmann","year":"2017","unstructured":"Kriegsmann, M. et al. Typing of colon and lung adenocarcinoma by high throughput imaging mass spectrometry. Biochim Biophys. Acta Proteins Proteom. 1865, 858\u2013864 (2017).","journal-title":"Biochim Biophys. Acta Proteins Proteom."},{"key":"705_CR10","doi-asserted-by":"publisher","first-page":"67","DOI":"10.1016\/j.tips.2023.11.003","volume":"45","author":"P Rajbhandari","year":"2024","unstructured":"Rajbhandari, P., Neelakantan, T. V., Hosny, N. & Stockwell, B. R. Spatial pharmacology using mass spectrometry imaging. Trends Pharm. Sci. 45, 67\u201380 (2024).","journal-title":"Trends Pharm. Sci."},{"key":"705_CR11","doi-asserted-by":"publisher","first-page":"2249","DOI":"10.1002\/pmic.201400088","volume":"14","author":"O Klein","year":"2014","unstructured":"Klein, O. et al. MALDI imaging mass spectrometry: discrimination of pathophysiological regions in traumatized skeletal muscle by characteristic peptide signatures. Proteomics 14, 2249\u20132260 (2014).","journal-title":"Proteomics"},{"key":"705_CR12","doi-asserted-by":"publisher","first-page":"3184","DOI":"10.3390\/cancers13133184","volume":"13","author":"Z Wu","year":"2021","unstructured":"Wu, Z. et al. Discovery of spatial peptide signatures for neuroblastoma risk assessment by MALDI mass spectrometry imaging. Cancers 13, 3184 (2021).","journal-title":"Cancers"},{"key":"705_CR13","doi-asserted-by":"publisher","first-page":"278","DOI":"10.1109\/ICDAR.1995.598994","volume":"1","author":"HoTin Kam","year":"1995","unstructured":"HoTin Kam Random decision forests. Int. Conf. Doc. Anal. Recognit. 1, 278\u2013282 (1995).","journal-title":"Int. Conf. Doc. Anal. Recognit."},{"key":"705_CR14","doi-asserted-by":"crossref","unstructured":"Hosmer Jr, D. W., Lemeshow, S. & Sturdivant, R. X. Applied Logistic Regression (John Wiley & Sons, 2013).","DOI":"10.1002\/9781118548387"},{"key":"705_CR15","doi-asserted-by":"publisher","first-page":"273","DOI":"10.1023\/A:1022627411411","volume":"20","author":"C Cortes","year":"1995","unstructured":"Cortes, C. & Vapnik, V. Support-vector networks. Mach. Learn. 20, 273\u2013297 (1995).","journal-title":"Mach. Learn."},{"key":"705_CR16","doi-asserted-by":"publisher","first-page":"4962","DOI":"10.1016\/j.jprot.2012.06.014","volume":"75","author":"EA Jones","year":"2012","unstructured":"Jones, E. A. et al. Imaging mass spectrometry statistical analysis. J. Proteom. 75, 4962\u20134989 (2012).","journal-title":"J. Proteom."},{"key":"705_CR17","unstructured":"Kanter, F. G. Deep Learning for Mass Spectrometry Imaging and Image Registration. Diss. Universit\u00e4t zu L\u00fcbeck (2023)."},{"key":"705_CR18","doi-asserted-by":"publisher","first-page":"1700181","DOI":"10.1002\/prca.201700181","volume":"13","author":"O Klein","year":"2019","unstructured":"Klein, O. et al. MALDI-imaging for classification of epithelial ovarian cancer histotypes from a tissue microarray using machine learning methods. PROTEOMICS\u2013Clin. Appl. 13, 1700181 (2019).","journal-title":"PROTEOMICS\u2013Clin. Appl."},{"key":"705_CR19","doi-asserted-by":"publisher","unstructured":"Winter, M. et al. MALDI mass spectrometry imaging: a novel tool for the identification and classification of amyloidosis. Proteomics https:\/\/doi.org\/10.1002\/pmic.201700236 (2017).","DOI":"10.1002\/pmic.201700236"},{"key":"705_CR20","doi-asserted-by":"publisher","first-page":"559","DOI":"10.1080\/14786440109462720","volume":"2","author":"K Pearson","year":"1901","unstructured":"Pearson, K. LIII. On lines and planes of closest fit to systems of points in space. Lond., Edinb., Dublin Philos. Mag. J. Sci. 2, 559\u2013572 (1901).","journal-title":"Lond., Edinb., Dublin Philos. Mag. J. Sci."},{"key":"705_CR21","doi-asserted-by":"publisher","first-page":"1900131","DOI":"10.1002\/prca.201900131","volume":"14","author":"O Klein","year":"2020","unstructured":"Klein, O. et al. Classification of inflammatory bowel disease from formalin-fixed, paraffin-embedded tissue biopsies via imaging mass spectrometry. PROTEOMICS\u2013Clin. Appl. 14, 1900131 (2020).","journal-title":"PROTEOMICS\u2013Clin. Appl."},{"key":"705_CR22","doi-asserted-by":"publisher","first-page":"686","DOI":"10.3390\/cancers15030686","volume":"15","author":"F Kanter","year":"2023","unstructured":"Kanter, F. et al. Classification of pancreatic ductal adenocarcinoma using MALDI mass spectrometry imaging combined with neural networks. Cancers 15, 686 (2023).","journal-title":"Cancers"},{"key":"705_CR23","doi-asserted-by":"publisher","first-page":"1215","DOI":"10.1093\/bioinformatics\/btx724","volume":"34","author":"J Behrmann","year":"2018","unstructured":"Behrmann, J. et al. Deep learning for tumor classification in imaging mass spectrometry. Bioinformatics 34, 1215\u20131223 (2018).","journal-title":"Bioinformatics"},{"key":"705_CR24","doi-asserted-by":"publisher","first-page":"5388","DOI":"10.3390\/cancers13215388","volume":"13","author":"P Mittal","year":"2021","unstructured":"Mittal, P. et al. Cancer tissue classification using supervised machine learning applied to MALDI mass spectrometry imaging. Cancers 13, 5388 (2021).","journal-title":"Cancers"},{"key":"705_CR25","doi-asserted-by":"publisher","first-page":"164","DOI":"10.1038\/s41591-021-01619-9","volume":"28","author":"C Weis","year":"2022","unstructured":"Weis, C. et al. Direct antimicrobial resistance prediction from clinical MALDI-TOF mass spectra using machine learning. Nat. Med 28, 164\u2013174 (2022).","journal-title":"Nat. Med"},{"key":"705_CR26","unstructured":"Pedregosa, F. et al. Scikit-learn: machine learning in Python. J. Mach. Learn. Res. 12, 2825\u20132830 (2011)."},{"key":"705_CR27","unstructured":"Sundararajan, M., Taly, A. & Yan, Q. Axiomatic attribution for deep networks. In International Conference On Machine Learning (ICML) (2017)."},{"key":"705_CR28","doi-asserted-by":"publisher","first-page":"152476","DOI":"10.1016\/j.anndiagpath.2025.152476","volume":"77","author":"D Ozyigit Buyuktalanci","year":"2025","unstructured":"Ozyigit Buyuktalanci, D., Gun, E., Dilek, O. N. & Dilek, F. H. Histopathological and prognostic variability of ampullary tumors: A comprehensive study on tumor location, histological subtypes, and survival outcomes. Ann. Diagn. Pathol. 77, 152476 (2025).","journal-title":"Ann. Diagn. Pathol."},{"key":"705_CR29","doi-asserted-by":"publisher","first-page":"301","DOI":"10.1007\/s00534-004-0898-3","volume":"11","author":"HP Fischer","year":"2004","unstructured":"Fischer, H. P. & Zhou, H. Pathogenesis of carcinoma of the papilla of Vater. J. Hepatobiliary Pancreat. Surg. 11, 301\u2013309 (2004).","journal-title":"J. Hepatobiliary Pancreat. Surg."},{"key":"705_CR30","doi-asserted-by":"publisher","unstructured":"Casadonte, R. et al. Imaging mass spectrometry to discriminate breast from pancreatic cancer metastasis in formalin-fixed paraffin-embedded tissues. Proteomics https:\/\/doi.org\/10.1002\/pmic.201300430 (2014).","DOI":"10.1002\/pmic.201300430"},{"key":"705_CR31","doi-asserted-by":"publisher","unstructured":"Stillger, M. N., Li, M. J., H\u00f6nscheid, P., von Neubeck, C. & F\u00f6ll, M. C. Advancing rare cancer research by MALDI mass spectrometry imaging: applications, challenges, and future perspectives in sarcoma. Proteomics https:\/\/doi.org\/10.1002\/pmic.202300001 (2024).","DOI":"10.1002\/pmic.202300001"},{"key":"705_CR32","doi-asserted-by":"publisher","first-page":"4342","DOI":"10.3390\/cancers14174342","volume":"14","author":"D Pertzborn","year":"2022","unstructured":"Pertzborn, D. et al. Multi-class cancer subtyping in salivary gland carcinomas with MALDI imaging and deep learning. Cancers 14, 4342 (2022).","journal-title":"Cancers"},{"key":"705_CR33","unstructured":"Bengio, Y., Goodfellow, I. & Courville, A. Deep Learning Vol. 1 (MIT Press, 2017)."},{"key":"705_CR34","unstructured":"Ioffe, S. & Szegedy, C. Batch normalization: accelerating deep network training by reducing internal covariate shift. In International Conference On Machine Learning (ICML) (2015)."},{"key":"705_CR35","unstructured":"Loshchilov, I. & Hutter, F. Decoupled weight decay regularization. In International Conference on Learning Representations (ICLR) (2019)."},{"key":"705_CR36","unstructured":"Dosovitskiy, A. et al. An image is worth 16x16 words: transformers for image recognition at scale. In International Conference on Learning Representations (ICLR) (2021)."},{"key":"705_CR37","doi-asserted-by":"publisher","unstructured":"Beyer, L., Zhai, X. & Kolesnikov, A. Better plain vit baselines for imagenet-1k. Preprint at https:\/\/doi.org\/10.48550\/arXiv.2205.01580 (2022).","DOI":"10.48550\/arXiv.2205.01580"},{"key":"705_CR38","doi-asserted-by":"publisher","unstructured":"Kokhlikyan, N. et al. Captum: a unified and generic model interpretability library for PyTorch. Preprint at https:\/\/doi.org\/10.48550\/arXiv.2009.07896 (2020).","DOI":"10.48550\/arXiv.2009.07896"},{"key":"705_CR39","unstructured":"Ancona, M. et al. Towards better understanding of gradient-based attribution methods for Deep Neural Networks. In International Conference on Learning Representations (ICLR) (2018)."}],"container-title":["npj Systems Biology and Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41540-026-00705-3","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41540-026-00705-3.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41540-026-00705-3.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,21]],"date-time":"2026-07-21T22:42:09Z","timestamp":1784673729000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41540-026-00705-3"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,12]]},"references-count":39,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["705"],"URL":"https:\/\/doi.org\/10.1038\/s41540-026-00705-3","relation":{},"ISSN":["2056-7189"],"issn-type":[{"value":"2056-7189","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,12]]},"assertion":[{"value":"18 September 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"25 March 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"12 May 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Ethics"}}],"article-number":"111"}}