{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,2]],"date-time":"2026-11-02T14:27:32Z","timestamp":1793629652838,"version":"4.5.1"},"reference-count":68,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,6,13]],"date-time":"2026-06-13T00:00:00Z","timestamp":1781308800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T00:00:00Z","timestamp":1781568000000},"content-version":"vor","delay-in-days":3,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Deutsches Krebsforschungszentrum (DKFZ)","id":[{"id":"https:\/\/ror.org\/04cdgtt98","id-type":"ROR","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Sci Rep"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Deep learning (DL) models developed for established medical imaging modalities have shown increasing performance and reliability as a result of scaling efforts. In contrast, model development for emerging modalities such as photoacoustic imaging (PAI) remains challenged by data sparsity, which limits model generalizability and raises the susceptibility to bias. While recent studies in PAI have started to investigate subject-related confounders, the impact of hardware-related confounders remains unexplored, posing a critical risk for failure in multicentric deployment scenarios. We are the first to provide a multicentric analysis of hardware-induced bias in PAI. We analyzed device-specific characteristics in images from four device instances and two peripheral artery disease studies, and trained DL models to classify device origin and disease under varying levels of device\u2013health correlations in the data. We showed that 1) multiple instances of the same PAI device type embed identifiable fingerprints in the images, 2) that DL models can leverage these fingerprints to reach\n                    <jats:inline-formula>\n                      <jats:tex-math>$$100\\,\\%$$<\/jats:tex-math>\n                    <\/jats:inline-formula>\n                    accuracy in device detection and critically, 3) when a correlation between device instance and health status is present, models trained for disease diagnosis exploit these device-specific signatures as shortcuts, thereby producing biased and clinically misleading predictions. This research highlights the risk of overestimating algorithm performance when such confounding is overlooked, emphasizing the importance of bias evaluation and explainable artificial intelligence methods to identify potential shortcuts, finally enabling multicentric PAI studies.\n                  <\/jats:p>","DOI":"10.1038\/s41598-026-53468-6","type":"journal-article","created":{"date-parts":[[2026,6,13]],"date-time":"2026-06-13T15:54:20Z","timestamp":1781366060000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["Photoacoustic device fingerprints induce bias in deep learning models"],"prefix":"10.1038","volume":"16","author":[{"given":"Christoph J.","family":"Bender","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marcel","family":"Knopp","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Niklas","family":"Holzwarth","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tom","family":"Rix","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jan-Hinrich","family":"N\u00f6lke","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kris K.","family":"Dreher","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yi","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Julius","family":"Kempf","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Milenko","family":"Caranovic","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fabian","family":"Schneider","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Melanie","family":"Schellenberg","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Leonie","family":"Boland","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Briain","family":"Haney","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ferdinand","family":"Knieling","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ulrich","family":"Rother","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alexander","family":"Seitel","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lena","family":"Maier-Hein","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,6,13]]},"reference":[{"key":"53468_CR1","doi-asserted-by":"publisher","first-page":"665","DOI":"10.1038\/s42256-020-00257-z","volume":"2","author":"R Geirhos","year":"2020","unstructured":"Geirhos, R. et al. Shortcut learning in deep neural networks. Nat. Mach. Intell. 2, 665\u2013673,\u00a0https:\/\/doi.org\/10.1038\/s42256-020-00257-z (2020).","journal-title":"Nat. Mach. Intell."},{"key":"53468_CR2","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-023-39902-7","volume":"14","author":"A Brown","year":"2023","unstructured":"Brown, A. et al. Detecting shortcut learning for fair medical AI using shortcut testing. Nat. Commun. 14, 4314,\u00a0https:\/\/doi.org\/10.1038\/s41467-023-39902-7 (2023).","journal-title":"Nat. Commun."},{"key":"53468_CR3","doi-asserted-by":"publisher","first-page":"75","DOI":"10.4274\/dir.2024.242854","volume":"31","author":"B Ko\u00e7ak","year":"2025","unstructured":"Ko\u00e7ak, B. et al. Bias in artificial intelligence for medical imaging: Fundamentals, detection, avoidance, mitigation, challenges, ethics, and prospects. Diagn. Interv. Radiol. 31, 75\u201388,\u00a0https:\/\/doi.org\/10.4274\/dir.2024.242854 (2025).","journal-title":"Diagn. Interv. Radiol."},{"key":"53468_CR4","doi-asserted-by":"publisher","unstructured":"Zong, Y., Yang, Y. & Hospedales, T. M. MEDFAIR: Benchmarking Fairness for Medical Imaging. International Conference on Learning Representations (ICLR), https:\/\/doi.org\/10.48550\/arXiv.2210.01725 (2023).","DOI":"10.48550\/arXiv.2210.01725"},{"key":"53468_CR5","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.230060","volume":"5","author":"B Glocker","year":"2023","unstructured":"Glocker, B., Jones, C., Roschewitz, M. & Winzeck, S. Risk of bias in chest radiography deep learning foundation models. Radiol. Artif. Intell. 5, e230060,\u00a0https:\/\/doi.org\/10.1148\/ryai.230060 (2023).","journal-title":"Radiol. Artif. Intell."},{"key":"53468_CR6","doi-asserted-by":"publisher","unstructured":"Jones, C., Roschewitz, M. & Glocker, B. The Role of Subgroup Separability in Group-Fair Medical Image Classification. Medical Image Computing and Computer Assisted Intervention (MICCAI), 179\u2013188, https:\/\/doi.org\/10.1007\/978-3-031-43898-1_18 (2023).","DOI":"10.1007\/978-3-031-43898-1_18"},{"key":"53468_CR7","doi-asserted-by":"publisher","DOI":"10.3389\/fninf.2021.805669","author":"M Bento","year":"2022","unstructured":"Bento, M., Fantini, I., Park, J., Rittner, L. & Frayne, R. Deep learning in large and multi-site structural brain MR imaging datasets. Front. Neuroinform 15, 805669,\u00a0https:\/\/doi.org\/10.3389\/fninf.2021.805669 (2022).","journal-title":"Front. Neuroinform."},{"key":"53468_CR8","doi-asserted-by":"publisher","unstructured":"Drenkow, N. Detecting Dataset Bias in Medical AI: A Generalized and Modality-Agnostic Auditing Framework, https:\/\/doi.org\/10.48550\/arXiv.2503.09969 (2025).","DOI":"10.48550\/arXiv.2503.09969"},{"key":"53468_CR9","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2020.101879","volume":"67","author":"C Wachinger","year":"2021","unstructured":"Wachinger, C., Rieckmann, A. & P\u00f6lsterl, S. Detect and correct bias in multi-site neuroimaging datasets. Med. Image Anal. 67, 101879,\u00a0https:\/\/doi.org\/10.1016\/j.media.2020.101879 (2021).","journal-title":"Med. Image Anal."},{"key":"53468_CR10","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pmed.1002683","volume":"15","author":"JR Zech","year":"2018","unstructured":"Zech, J. R. et al. Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: A cross-sectional study. PLoS Med. 15, e1002683,\u00a0https:\/\/doi.org\/10.1371\/journal.pmed.1002683 (2018).","journal-title":"PLoS Med."},{"key":"53468_CR11","doi-asserted-by":"publisher","first-page":"602","DOI":"10.1098\/rsfs.2011.0028","volume":"1","author":"P Beard","year":"2011","unstructured":"Beard, P. Biomedical photoacoustic imaging. Interface Focus 1, 602\u2013631,\u00a0https:\/\/doi.org\/10.1098\/rsfs.2011.0028 (2011).","journal-title":"Interface Focus"},{"key":"53468_CR12","doi-asserted-by":"publisher","first-page":"65","DOI":"10.1016\/j.pacs.2018.05.001","volume":"10","author":"M Li","year":"2018","unstructured":"Li, M., Tang, Y. & Yao, J. Photoacoustic tomography of blood oxygenation: A mini review. Photoacoustics 10, 65\u201373,\u00a0https:\/\/doi.org\/10.1016\/j.pacs.2018.05.001 (2018).","journal-title":"Photoacoustics"},{"key":"53468_CR13","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab1ede","volume":"64","author":"S Iskander-Rizk","year":"2019","unstructured":"Iskander-Rizk, S., van der Steen, A. F. W. & van Soest, G. Photoacoustic imaging for guidance of interventions in cardiovascular medicine. Phys. Med. Biol. 64, 16TR01,\u00a0https:\/\/doi.org\/10.1088\/1361-6560\/ab1ede (2019).","journal-title":"Phys. Med. Biol."},{"key":"53468_CR14","doi-asserted-by":"publisher","DOI":"10.1038\/s44222-024-00240-y","author":"J Park","year":"2024","unstructured":"Park, J. et al. Clinical translation of photoacoustic imaging. Nat. Rev. Bioeng.\u00a03, 193\u2013212, https:\/\/doi.org\/10.1038\/s44222-024-00240-y (2024).","journal-title":"Nat. Rev. Bioeng."},{"key":"53468_CR15","doi-asserted-by":"publisher","first-page":"3366","DOI":"10.1109\/TMI.2024.3403417","volume":"43","author":"J-H N\u00f6lke","year":"2024","unstructured":"N\u00f6lke, J.-H. et al. Photoacoustic quantification of tissue oxygenation using conditional invertible neural networks. IEEE Trans. Med. Imaging\u00a043, 3366\u20133376,\u00a0https:\/\/doi.org\/10.1109\/TMI.2024.3403417 (2024).","journal-title":"IEEE Trans. Med. Imaging."},{"key":"53468_CR16","doi-asserted-by":"publisher","DOI":"10.1117\/1.JBO.29.S1.S11506","volume":"29","author":"TR Else","year":"2023","unstructured":"Else, T. R. et al. Effects of skin tone on photoacoustic imaging and oximetry. J. Biomed. Opt. 29, S11506,\u00a0https:\/\/doi.org\/10.1117\/1.JBO.29.S1.S11506 (2023).","journal-title":"J. Biomed. Opt."},{"key":"53468_CR17","doi-asserted-by":"publisher","first-page":"1325","DOI":"10.1007\/s11548-025-03370-9","volume":"20","author":"M Knopp","year":"2025","unstructured":"Knopp, M. et al. Shortcut learning leads to sex bias in deep learning models for photoacoustic tomography. Int. J. Comput. Assist. Radiol. Surg. 20, 1325\u20131333,\u00a0https:\/\/doi.org\/10.1007\/s11548-025-03370-9 (2025).","journal-title":"Int. J. Comput. Assist. Radiol. Surg."},{"key":"53468_CR18","doi-asserted-by":"publisher","unstructured":"Gretton, A., Bousquet, O., Smola, A. & Sch\u00f6lkopf, B. Measuring Statistical Dependence with Hilbert-Schmidt Norms, Algorithmic Learning Theory (ALT), 3734, https:\/\/doi.org\/10.1007\/11564089_7 (2005).","DOI":"10.1007\/11564089_7"},{"key":"53468_CR19","unstructured":"Gretton, A. et al. A Kernel Statistical Test of Independence. Advances in Neural Information Processing Systems (NIPS) 20 (2007)."},{"key":"53468_CR20","doi-asserted-by":"publisher","unstructured":"Kornblith, S., Norouzi, M., Lee, H. & Hinton, G. Similarity of Neural Network Representations Revisited. International Conference on Machine Learning (ICML) 97, 3519\u20133529, https:\/\/doi.org\/10.48550\/arXiv.1905.00414 (2019).","DOI":"10.48550\/arXiv.1905.00414"},{"key":"53468_CR21","unstructured":"Pearson, K. Mathematical Contributions to the Theory of Evolution. XIII. On the Theory of Contingency and Its Relation to Association and Normal Correlation. (Drapers Company Research Memoirs, Biometric Series I, 1904)."},{"key":"53468_CR22","volume-title":"Mathematical Methods of Statistics","author":"H Cram\u00e9r","year":"1946","unstructured":"Cram\u00e9r, H. Mathematical Methods of Statistics (Princeton University Press, 1946)."},{"key":"53468_CR23","doi-asserted-by":"publisher","first-page":"336","DOI":"10.1007\/s11263-019-01228-7","volume":"128","author":"RR Selvaraju","year":"2020","unstructured":"Selvaraju, R. R. et al. Grad-CAM: Visual explanations from deep networks via gradient-based localization. Int. J. Comput. Vis. 128, 336\u2013359,\u00a0https:\/\/doi.org\/10.1007\/s11263-019-01228-7 (2020).","journal-title":"Int. J. Comput. Vis."},{"key":"53468_CR24","doi-asserted-by":"publisher","unstructured":"Draelos, R. L. & Carin, L. Use HiResCAM instead of Grad-CAM for faithful explanations of convolutional neural networks, https:\/\/doi.org\/10.48550\/arXiv.2011.08891 (2020).","DOI":"10.48550\/arXiv.2011.08891"},{"key":"53468_CR25","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2024.108238","volume":"253","author":"V Lamprou","year":"2024","unstructured":"Lamprou, V., Kallipolitis, A. & Maglogiannis, I. On the evaluation of deep learning interpretability methods for medical images under the scope of faithfulness. Comput. Methods Programs Biomed. 253, 108238,\u00a0https:\/\/doi.org\/10.1016\/j.cmpb.2024.108238 (2024).","journal-title":"Comput. Methods Programs Biomed."},{"key":"53468_CR26","doi-asserted-by":"publisher","DOI":"10.1117\/1.JBO.28.9.097001","volume":"28","author":"MT Graham","year":"2023","unstructured":"Graham, M. T. et al. Optical absorption spectra and corresponding in vivo photoacoustic visualization of exposed peripheral nerves. J. Biomed. Opt. 28, 097001,\u00a0https:\/\/doi.org\/10.1117\/1.JBO.28.9.097001 (2023).","journal-title":"J. Biomed. Opt."},{"key":"53468_CR27","doi-asserted-by":"publisher","first-page":"433","DOI":"10.2214\/AJR.06.0409","volume":"188","author":"RN Uppot","year":"2007","unstructured":"Uppot, R. N., Sahani, D. V., Hahn, P. F., Gervais, D. & Mueller, P. R. Impact of obesity on medical imaging and image-guided intervention. AJR Am. J. Roentgenol. 188, 433\u2013440,\u00a0https:\/\/doi.org\/10.2214\/AJR.06.0409 (2007).","journal-title":"AJR Am. J. Roentgenol."},{"key":"53468_CR28","doi-asserted-by":"publisher","unstructured":"Dietrich, M. et al. Machine learning-based analysis of hyperspectral images for automated sepsis diagnosis, https:\/\/doi.org\/10.48550\/arXiv.2106.08445 (2021).","DOI":"10.48550\/arXiv.2106.08445"},{"key":"53468_CR29","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-025-95137-0","volume":"15","author":"N Holzwarth","year":"2025","unstructured":"Holzwarth, N. et al. Photoacoustic imaging for monitoring radiotherapy treatment response in head and neck tumors. Sci. Rep. 15, 16344,\u00a0https:\/\/doi.org\/10.1038\/s41598-025-95137-0 (2025).","journal-title":"Sci. Rep."},{"key":"53468_CR30","doi-asserted-by":"publisher","DOI":"10.1002\/jbio.202400106","volume":"17","author":"Y Li","year":"2024","unstructured":"Li, Y. et al. Teachability of multispectral optoacoustic tomography. J. Biophotonics 17, e202400106,\u00a0https:\/\/doi.org\/10.1002\/jbio.202400106 (2024).","journal-title":"J. Biophotonics"},{"key":"53468_CR31","doi-asserted-by":"publisher","DOI":"10.1117\/1.3247155","volume":"14","author":"M Pramanik","year":"2009","unstructured":"Pramanik, M. & Wang, L. V. Thermoacoustic and photoacoustic sensing of temperature. J. Biomed. Opt. 14, 054024,\u00a0https:\/\/doi.org\/10.1117\/1.3247155 (2009).","journal-title":"J. Biomed. Opt."},{"key":"53468_CR32","unstructured":"Longo, A. Image Quality Improvement in Optoacoustic Tomography (Munich, Germany, 2022) (Ph.D. thesis)."},{"key":"53468_CR33","doi-asserted-by":"publisher","unstructured":"Tan, M. & Le, Q. EfficientNetV2: Smaller Models and Faster Training. International Conference on Machine Learning (ICML), 10096\u201310106, https:\/\/doi.org\/10.48550\/arXiv.2104.00298 (2021).","DOI":"10.48550\/arXiv.2104.00298"},{"key":"53468_CR34","doi-asserted-by":"publisher","unstructured":"Vo, A. et al. Vision Language Models are Biased, https:\/\/doi.org\/10.48550\/arXiv.2505.23941 (2025).","DOI":"10.48550\/arXiv.2505.23941"},{"key":"53468_CR35","doi-asserted-by":"publisher","unstructured":"Mayer, L. et al. 6 Fingers, 1 Kidney: Natural Adversarial Medical Images Reveal Critical Weaknesses of Vision-Language Models, https:\/\/doi.org\/10.48550\/arXiv.2512.04238 (2025).","DOI":"10.48550\/arXiv.2512.04238"},{"key":"53468_CR36","doi-asserted-by":"publisher","unstructured":"Liu, Z. et al. Swin Transformer V2: Scaling Up Capacity and Resolution. In International Conference on Computer Vision and Pattern Recognition (CVPR), 11999-12009, https:\/\/doi.org\/10.48550\/arXiv.2111.09883 (2022).","DOI":"10.48550\/arXiv.2111.09883"},{"key":"53468_CR37","doi-asserted-by":"publisher","unstructured":"Pavlak, M., Drenkow, N., Petrick, N., Farhangi, M. M. & Unberath, M. Data AUDIT: Identifying Attribute Utility- and Detectability-Induced Bias in Task Models. Medical Image Computing and Computer Assisted Intervention (MICCAI), 442\u2013452, https:\/\/doi.org\/10.1007\/978-3-031-43898-1_43 (2023).","DOI":"10.1007\/978-3-031-43898-1_43"},{"key":"53468_CR38","doi-asserted-by":"publisher","first-page":"3182","DOI":"10.1109\/TMI.2022.3180115","volume":"41","author":"C Dehner","year":"2022","unstructured":"Dehner, C., Olefir, I., Chowdhury, K. B., Justel, D. & Ntziachristos, V. Deep-learning-based electrical noise removal enables high spectral optoacoustic contrast in deep tissue. IEEE Trans. Med. Imaging 41, 3182\u20133193,\u00a0https:\/\/doi.org\/10.1109\/TMI.2022.3180115 (2022).","journal-title":"IEEE Trans. Med. Imaging"},{"key":"53468_CR39","doi-asserted-by":"publisher","first-page":"1166","DOI":"10.1038\/s41591-024-02838-6","volume":"30","author":"I Ktena","year":"2024","unstructured":"Ktena, I. et al. Generative models improve fairness of medical classifiers under distribution shifts. Nat. Med. 30, 1166\u20131173,\u00a0https:\/\/doi.org\/10.1038\/s41591-024-02838-6 (2024).","journal-title":"Nat. Med."},{"key":"53468_CR40","doi-asserted-by":"publisher","unstructured":"Stanley, E. A. M. Synthetic Ground Truth Counterfactuals for Comprehensive Evaluation of Causal Generative Models in Medical Imaging. In Medical Image Computing and Computer Assisted Intervention (MICCAI), 541\u2013550, https:\/\/doi.org\/10.1007\/978-3-032-04984-1_52 (2025).","DOI":"10.1007\/978-3-032-04984-1_52"},{"key":"53468_CR41","doi-asserted-by":"publisher","unstructured":"Sagawa, S., Koh, P. W., Hashimoto, T. B. & Liang, P. Distributionally Robust Neural Networks. International Conference on Learning Representations (ICLR), https:\/\/doi.org\/10.48550\/arXiv.1911.08731 (2020).","DOI":"10.48550\/arXiv.1911.08731"},{"key":"53468_CR42","doi-asserted-by":"publisher","first-page":"2838","DOI":"10.1038\/s41591-024-03113-4","volume":"30","author":"Y Yang","year":"2024","unstructured":"Yang, Y., Zhang, H., Gichoya, J. W., Katabi, D. & Ghassemi, M. The limits of fair medical imaging AI in real-world generalization. Nat. Med. 30, 2838\u20132848,\u00a0https:\/\/doi.org\/10.1038\/s41591-024-03113-4 (2024).","journal-title":"Nat. Med."},{"key":"53468_CR43","doi-asserted-by":"publisher","unstructured":"Nguyen, M., Wang, A. Q., Kim, H. & Sabuncu, M. R. Robust Learning via Conditional Prevalence Adjustment. Winter Conference on Applications of Computer Vision (WACV), 2729\u20132738, https:\/\/doi.org\/10.1109\/WACV57701.2024.00272 (2024).","DOI":"10.1109\/WACV57701.2024.00272"},{"key":"53468_CR44","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-023-00805-y","volume":"6","author":"J Yang","year":"2023","unstructured":"Yang, J., Soltan, A. A. S., Eyre, D. W., Yang, Y. & Clifton, D. A. An adversarial training framework for mitigating algorithmic biases in clinical machine learning. npj Digit. Med. 6, 55,\u00a0https:\/\/doi.org\/10.1038\/s41746-023-00805-y (2023).","journal-title":"npj Digit. Med."},{"key":"53468_CR45","doi-asserted-by":"publisher","unstructured":"Tirindelli, M. Rethinking Ultrasound Augmentation: A Physics-Inspired Approach. Medical Image Computing and Computer Assisted Intervention (MICCAI), 690\u2013700, https:\/\/doi.org\/10.1007\/978-3-030-87237-3_66 (2021)","DOI":"10.1007\/978-3-030-87237-3_66"},{"key":"53468_CR46","doi-asserted-by":"publisher","DOI":"10.1117\/1.JBO.27.8.083010","author":"J Gr\u00f6hl","year":"2022","unstructured":"Gr\u00f6hl, J. et al. SIMPA: An open-source toolkit for simulation and image processing for photonics and acoustics. J. Biomed. Opt. https:\/\/doi.org\/10.1117\/1.JBO.27.8.083010 (2022).","journal-title":"J. Biomed. Opt."},{"key":"53468_CR47","doi-asserted-by":"publisher","unstructured":"Ilse, M., Tomczak, J. M., Louizos, C. & Welling, M. DIVA: Domain Invariant Variational Autoencoders. Medical Imaging with Deep Learning (MIDL) 121, 322\u2013348, https:\/\/doi.org\/10.48550\/arXiv.1905.10427 (2020).","DOI":"10.48550\/arXiv.1905.10427"},{"key":"53468_CR48","doi-asserted-by":"publisher","unstructured":"Roschewitz, M., de Sousa Ribeiro, F., Xia, T., Khara, G. & Glocker, B. Counterfactual Contrastive Learning: Robust Representations via Causal Image Synthesis. Data Engineering in Medical Imaging (DEMI), 22\u201332, https:\/\/doi.org\/10.1007\/978-3-031-73748-0_3 (2025).","DOI":"10.1007\/978-3-031-73748-0_3"},{"key":"53468_CR49","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2025.103668","volume":"105","author":"M Roschewitz","year":"2025","unstructured":"Roschewitz, M., De Sousa Ribeiro, F., Xia, T., Khara, G. & Glocker, B. Robust image representations with counterfactual contrastive learning. Med. Image Anal. 105, 103668,\u00a0https:\/\/doi.org\/10.1016\/j.media.2025.103668 (2025).","journal-title":"Med. Image Anal."},{"key":"53468_CR50","doi-asserted-by":"publisher","unstructured":"Liu, E. Z. et al. Just Train Twice: Improving Group Robustness without Training Group Information. International Conference on Machine Learning (ICML) 139, 6781\u20136792, \t https:\/\/doi.org\/10.48550\/arXiv.2107.09044 (2021).","DOI":"10.48550\/arXiv.2107.09044"},{"key":"53468_CR51","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcmg.2022.11.010","author":"JS G\u00fcnther","year":"2023","unstructured":"G\u00fcnther, J. S. Targeting muscular hemoglobin content for classification of peripheral arterial disease by noninvasive multispectral optoacoustic tomography. JACC Cardiovasc. Imaging https:\/\/doi.org\/10.1016\/j.jcmg.2022.11.010 (2023).","journal-title":"JACC Cardiovasc. Imaging"},{"key":"53468_CR52","doi-asserted-by":"publisher","DOI":"10.1016\/j.pacs.2023.100579","volume":"35","author":"AP Tr\u00e4ger","year":"2024","unstructured":"Tr\u00e4ger, A. P. et al. Hybrid ultrasound and single wavelength optoacoustic imaging reveals muscle degeneration in peripheral artery disease. Photoacoustics 35, 100579,\u00a0https:\/\/doi.org\/10.1016\/j.pacs.2023.100579 (2024).","journal-title":"Photoacoustics"},{"key":"53468_CR53","doi-asserted-by":"publisher","DOI":"10.1038\/s43856-025-00801-1","volume":"5","author":"M Caranovic","year":"2025","unstructured":"Caranovic, M. et al. Derivation and validation of a non-invasive optoacoustic imaging biomarker for detection of patients with intermittent claudication. Commun. Med. 5, 88,\u00a0https:\/\/doi.org\/10.1038\/s43856-025-00801-1 (2025).","journal-title":"Commun. Med."},{"key":"53468_CR54","doi-asserted-by":"publisher","first-page":"378","DOI":"10.1055\/s-0034-1393976","volume":"31","author":"R Hardman","year":"2014","unstructured":"Hardman, R., Jazaeri, O., Yi, J., Smith, M. & Gupta, R. Overview of classification systems in peripheral artery disease. Semin. Intervent. Radiol. 31, 378\u2013388,\u00a0https:\/\/doi.org\/10.1055\/s-0034-1393976 (2014).","journal-title":"Semin. Intervent. Radiol."},{"key":"53468_CR55","doi-asserted-by":"publisher","first-page":"3218","DOI":"10.1109\/TMI.2020.2989236","volume":"39","author":"KB Chowdhury","year":"2020","unstructured":"Chowdhury, K. B., Prakash, J., Karlas, A., J\u00fcstel, D. & Ntziachristos, V. A synthetic total impulse response characterization method for correction of hand-held optoacoustic images. Trans. Med. Imaging 39, 3218\u20133230,\u00a0https:\/\/doi.org\/10.1109\/TMI.2020.2989236 (2020).","journal-title":"IEEE Trans. Med. Imaging"},{"key":"53468_CR56","doi-asserted-by":"publisher","DOI":"10.1016\/j.pacs.2025.100745","volume":"45","author":"MT Rietberg","year":"2025","unstructured":"Rietberg, M. T. et al. Artifacts in photoacoustic imaging: Origins and mitigations. Photoacoustics 45, 100745,\u00a0https:\/\/doi.org\/10.1016\/j.pacs.2025.100745 (2025).","journal-title":"Photoacoustics"},{"key":"53468_CR57","doi-asserted-by":"crossref","unstructured":"Bender, C. J. Hardware-Related Biases in Machine Learning Algorithms for Photoacoustic Image Analysis (Heidelberg, Germany, 2023) (Master\u2019s thesis).","DOI":"10.1117\/12.3000348"},{"key":"53468_CR58","doi-asserted-by":"publisher","DOI":"10.1364\/BOE.8.000068","volume":"8","author":"ER Hill","year":"2017","unstructured":"Hill, E. R., Xia, W., Clarkson, M. J. & Desjardins, A. E. Identification and removal of laser-induced noise in photoacoustic imaging using singular value decomposition. Biomed. Opt. Express 8, 68,\u00a0https:\/\/doi.org\/10.1364\/BOE.8.000068 (2017).","journal-title":"Biomed. Opt. Express"},{"key":"53468_CR59","unstructured":"Laha, A. Automatische Erkennung und Entfernung von Artefakten in der photoakustischen Bildgebung (Heidelberg, Germany, 2018) (Master\u2019s thesis)."},{"key":"53468_CR60","doi-asserted-by":"publisher","DOI":"10.1118\/1.4893530","volume":"41","author":"S Tzoumas","year":"2014","unstructured":"Tzoumas, S., Rosenthal, A., Lutzweiler, C., Razansky, D. & Ntziachristos, V. Spatiospectral denoising framework for multispectral optoacoustic imaging based on sparse signal representation. Med. Phys. 41, 113301,\u00a0https:\/\/doi.org\/10.1118\/1.4893530 (2014).","journal-title":"Med. Phys."},{"key":"53468_CR61","doi-asserted-by":"publisher","first-page":"2011","DOI":"10.1109\/CVPR.2011.5995347","volume":"1521\u20131528","author":"A Torralba","year":"2011","unstructured":"Torralba, A. & Efros, A. A. Unbiased look at dataset bias. International Conference on Computer Vision and Pattern Recognition (CVPR),\u00a01521\u20131528,\u00a0https:\/\/doi.org\/10.1109\/CVPR.2011.5995347 (2011).","journal-title":"In CVPR"},{"key":"53468_CR62","unstructured":"TorchVision maintainers & contributors. TorchVision: PyTorch\u2019s computer vision library. https:\/\/github.com\/pytorch\/vision (2016)."},{"key":"53468_CR63","doi-asserted-by":"publisher","unstructured":"Loshchilov, I. & Hutter, F. Decoupled Weight Decay Regularization, https:\/\/doi.org\/10.48550\/arXiv.1711.05101 (2019).","DOI":"10.48550\/arXiv.1711.05101"},{"key":"53468_CR64","unstructured":"Yadan, O. Hydra - A framework for elegantly configuring complex applications. https:\/\/github.com\/facebookresearch\/hydra (2019)."},{"key":"53468_CR65","doi-asserted-by":"publisher","unstructured":"Akiba, T., Sano, S., Yanase, T., Ohta, T. & Koyama, M. Optuna: A Next-generation Hyperparameter Optimization Framework. International Conference on Knowledge Discovery and Data Mining (KDD), 2623\u20132631, https:\/\/doi.org\/10.1145\/3292500.3330701 (2019).","DOI":"10.1145\/3292500.3330701"},{"key":"53468_CR66","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/S0169-7439(00)00122-2","volume":"56","author":"Q-S Xu","year":"2001","unstructured":"Xu, Q.-S. & Liang, Y.-Z. Monte Carlo cross validation. Chemom. Intell. Lab. Syst. 56, 1\u201311,\u00a0https:\/\/doi.org\/10.1016\/S0169-7439(00)00122-2 (2001).","journal-title":"Chemom. Intell. Lab. Syst."},{"key":"53468_CR67","doi-asserted-by":"publisher","first-page":"195","DOI":"10.1038\/s41592-023-02151-z","volume":"21","author":"L Maier-Hein","year":"2024","unstructured":"Maier-Hein, L. et al. Metrics reloaded: Recommendations for image analysis validation. Nat. Methods 21, 195\u2013212,\u00a0https:\/\/doi.org\/10.1038\/s41592-023-02151-z (2024).","journal-title":"Nat. Methods"},{"key":"53468_CR68","unstructured":"Gildenblat, J. & contributors. PyTorch library for CAM methods. https:\/\/github.com\/jacobgil\/pytorch-grad-cam (2021)."}],"container-title":["Scientific Reports"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-53468-6","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-53468-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-53468-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,11,2]],"date-time":"2026-11-02T13:44:52Z","timestamp":1793627092000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-53468-6"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,13]]},"references-count":68,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["53468"],"URL":"https:\/\/doi.org\/10.1038\/s41598-026-53468-6","relation":{},"ISSN":["2045-2322"],"issn-type":[{"value":"2045-2322","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,13]]},"assertion":[{"value":"10 January 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"12 May 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"13 June 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":"Declarations"}}],"article-number":"18695"}}