{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,13]],"date-time":"2026-08-13T22:24:12Z","timestamp":1786659852850,"version":"build-2736575974"},"reference-count":54,"publisher":"The Royal Society","issue":"2326","license":[{"start":{"date-parts":[[2026,8,13]],"date-time":"2026-08-13T00:00:00Z","timestamp":1786579200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"European Union\u2014Next Generation EU\u2014Progetti di Ricerca di Rilevante Interesse Nazionale (PRIN) 2022, PNRR Prot.","award":["P20222MYKE"],"award-info":[{"award-number":["P20222MYKE"]}]},{"name":"A Pilot analysIs of behavioRal and opeRational data for detEcting Socio-emotional PrEcursors of mild CogniTIVE impairments (MCI) and demEntia","award":["B53D23025980001"],"award-info":[{"award-number":["B53D23025980001"]}]},{"name":"Next Generation EU - Progetti di Ricerca di Rilevante Interesse Nazionale (PRIN) 2022","award":["2022ENK9LS"],"award-info":[{"award-number":["2022ENK9LS"]}]},{"name":"EXEGETE: Explainable Generative Deep Learning Methods for Medical Image and Signal Processing","award":["CUP: B53D23013040006"],"award-info":[{"award-number":["CUP: B53D23013040006"]}]},{"name":"Innovative Technologies for Smart Low Emission Mobilities","award":["CZ.02.01.01\/00\/23_020\/0008528"],"award-info":[{"award-number":["CZ.02.01.01\/00\/23_020\/0008528"]}]},{"name":"Programme Johannes Amos Comenius, call Intersectoral cooperation"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,8,13]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Hypercomplex number systems, such as complex numbers and quaternions, have found many uses in the area of signal analysis. Clifford algebras, along with the embedding of (usually) a real vector space, can be used to represent some of the commonly used hypercomplex algebras through subalgebras. Concepts used in hypercomplex algebras can be easily further extended to Clifford algebra. An overview of methods along with their advantages and disadvantages based on Clifford algebra, mostly used in signal analysis, is given in this review, with a focus on applications in neural networks. The use of the described methods is discussed in the context of medical imaging.<\/jats:p>\n                  <jats:p>This article is part of the theme issue \u2018Modern applications of geometric algebra\u2019.<\/jats:p>","DOI":"10.1098\/rsta.2025.0099","type":"journal-article","created":{"date-parts":[[2026,8,13]],"date-time":"2026-08-13T10:44:36Z","timestamp":1786617876000},"source":"Crossref","is-referenced-by-count":1,"title":["Review of Clifford algebra-based AI methods in signal analysis"],"prefix":"10.1098","volume":"384","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7452-0573","authenticated-orcid":false,"given":"Roman","family":"Byrtus","sequence":"first","affiliation":[{"name":"Institute of Mathematics, Brno University of Technology Faculty of Mechanical Engineering 1 , Brno, Czech Republic"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Francesco","family":"Prinzi","sequence":"additional","affiliation":[{"name":"Department of Biomedicine, Neuroscience and Advanced Diagnostics, University of Palermo 2 , Palermo, Sicily, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6725-7378","authenticated-orcid":false,"given":"Petr","family":"Vasik","sequence":"additional","affiliation":[{"name":"Institute of Mathematics, Brno University of Technology Faculty of Mechanical Engineering 1 , Brno, Czech Republic"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Salvatore","family":"Vitabile","sequence":"additional","affiliation":[{"name":"Department of Biomedicine, Neuroscience and Advanced Diagnostics, University of Palermo 2 , Palermo, Sicily, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"175","published-online":{"date-parts":[[2026,8,13]]},"reference":[{"key":"2026081306443452700_bib1","doi-asserted-by":"crossref","first-page":"155","DOI":"10.2307\/1967865","article-title":"On quaternions and their generalization and the history of the eight square theorem","volume":"20","author":"Dickson","year":"1919","journal-title":"Ann. Math."},{"key":"2026081306443452700_bib2","first-page":"58","article-title":"Theory of communication","volume":"94","author":"Gabor","year":"1946","journal-title":"J. Inst. Electr. Eng."},{"key":"2026081306443452700_bib3","doi-asserted-by":"crossref","first-page":"2844","DOI":"10.1109\/78.960432","article-title":"Hypercomplex signals-a novel extension of the analytic signal to the multidimensional case","volume":"49","author":"Bulow","year":"2001","journal-title":"IEEE Trans. Signal Process."},{"key":"2026081306443452700_bib4","first-page":"167","article-title":"The unified theory of n-dimensional complex and hypercomplex analytic signals","volume":"59","author":"Hahn","year":"2011","journal-title":"Bull. Pol. Acad. Sci."},{"key":"2026081306443452700_bib5","doi-asserted-by":"crossref","first-page":"14","DOI":"10.1109\/MSP.2024.3365463","article-title":"Hypercomplex techniques in signal and image processing using network graph theory: identifying core research directions","volume":"41","author":"Alcayde","year":"2024","journal-title":"IEEE Signal Process. Mag."},{"key":"2026081306443452700_bib6","article-title":"History of quaternion and Clifford-Fourier transforms and wavelets","volume-title":"Quaternion and Clifford Fourier transforms and wavelets","author":"Brackx","year":"2013"},{"key":"2026081306443452700_bib7","doi-asserted-by":"crossref","first-page":"22","DOI":"10.1109\/TIP.2006.884955","article-title":"Hypercomplex Fourier transforms of color images","volume":"16","author":"Ell","year":"2007","journal-title":"IEEE Trans. Image Process."},{"key":"2026081306443452700_bib8","doi-asserted-by":"crossref","first-page":"211","DOI":"10.1007\/978-3-031-66342-0_9","article-title":"Integral Transforms","volume-title":"Geometric Algebra Applications","author":"Bayro-Corrochano","year":"2024"},{"key":"2026081306443452700_bib9","doi-asserted-by":"crossref","first-page":"130195","DOI":"10.1016\/j.neucom.2025.130195","article-title":"Generalizing medical image representations via quaternion wavelet networks","volume":"638","author":"Sigillo","year":"2025","journal-title":"Neurocomputing"},{"key":"2026081306443452700_bib10","doi-asserted-by":"crossref","first-page":"1316","DOI":"10.1109\/TSP.2008.2010600","article-title":"The quaternion LMS algorithm for adaptive filtering of hypercomplex processes","volume":"57","author":"Took","year":"2009","journal-title":"IEEE Trans. Signal Process."},{"key":"2026081306443452700_bib11","first-page":"53","volume-title":"Multi-layer perceptrons","author":"Kruse","year":"2022"},{"key":"2026081306443452700_bib12","doi-asserted-by":"crossref","first-page":"1406","DOI":"10.1109\/JAS.2022.105743","article-title":"Complex-valued neural networks: a comprehensive survey","volume":"9","author":"Lee","year":"2022","journal-title":"IEEE CAA J. Autom. Sinica."},{"key":"2026081306443452700_bib13","doi-asserted-by":"crossref","first-page":"2957","DOI":"10.1007\/s10462-019-09752-1","article-title":"A survey of quaternion neural networks","volume":"53","author":"Parcollet","year":"2019","journal-title":"Artif. Intell. Rev."},{"key":"2026081306443452700_bib14","doi-asserted-by":"crossref","first-page":"99","DOI":"10.1007\/s10462-024-10721-6","article-title":"A review of convolutional neural networks in computer vision","volume":"57","author":"Zhao","year":"2024","journal-title":"Artif. Intell. Rev."},{"key":"2026081306443452700_bib15","doi-asserted-by":"crossref","first-page":"195","DOI":"10.1007\/978-3-030-42227-1_9","article-title":"Recurrent networks","volume-title":"Computational intelligence : a methodological introduction","author":"Kruse","year":"2022","edition":"3rd"},{"key":"2026081306443452700_bib16","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1109\/MSP.2024.3401621","article-title":"Understanding vector-valued neural networks and their relationship with real and hypercomplex-valued neural networks: incorporating intercorrelation between features into neural networks","volume":"41","author":"Valle","year":"2024","journal-title":"IEEE Signal Process. Mag."},{"key":"2026081306443452700_bib17","doi-asserted-by":"crossref","first-page":"59","DOI":"10.1109\/MSP.2024.3401622","article-title":"Demystifying the hypercomplex: inductive biases in hypercomplex deep learning","volume":"41","author":"Comminiello","year":"2024","journal-title":"IEEE Signal Process. Mag."},{"key":"2026081306443452700_bib18","doi-asserted-by":"crossref","first-page":"136","DOI":"10.1016\/j.mri.2020.02.002","article-title":"DeepcomplexMRI: exploiting deep residual network for fast parallel MR imaging with complex convolution","volume":"68","author":"Wang","year":"2020","journal-title":"Magn. Reson. Imaging"},{"key":"2026081306443452700_bib19","doi-asserted-by":"crossref","first-page":"2101","DOI":"10.1109\/78.134446","article-title":"The complex backpropagation algorithm","volume":"39","author":"Leung","year":"1991","journal-title":"IEEE Trans. Signal Process."},{"key":"2026081306443452700_bib20","doi-asserted-by":"crossref","first-page":"75","DOI":"10.1002\/9781118590072","article-title":"Learning algorithms in complex-valued neural networks using Wirtinger calculus","volume-title":"Complex-valued neural networks : advances and applications","author":"Hirose","year":"2013"},{"key":"2026081306443452700_bib21","doi-asserted-by":"crossref","first-page":"23713","DOI":"10.1038\/s41598-025-08957-5","article-title":"4D hypercomplex-valued neural network in multivariate time series forecasting","volume":"15","author":"Kycia","year":"2025","journal-title":"Sci. Rep."},{"key":"2026081306443452700_bib22","doi-asserted-by":"crossref","first-page":"8293","DOI":"10.1109\/TNNLS.2022.3226772","article-title":"PHNNs: lightweight neural networks via parameterized hypercomplex convolutions","volume":"35","author":"Grassucci","year":"2024","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"2026081306443452700_bib23","doi-asserted-by":"crossref","first-page":"19","DOI":"10.1007\/s00006-024-01313-2","article-title":"Exploring quaternion neural network loss surfaces","volume":"34","author":"Bill","year":"2024","journal-title":"Adv. Appl. Clifford Algebras"},{"key":"2026081306443452700_bib24","first-page":"1","article-title":"Hypercomplex multimodal emotion recognition from EEG and peripheral physiological signals","volume-title":"2023 IEEE Int. Conf. Acoust. Speech Sig. Process. Workshops","author":"Lopez","year":"2023"},{"key":"2026081306443452700_bib25","volume-title":"Geometric algebra with applications in engineering","author":"Perwass","year":"2009"},{"key":"2026081306443452700_bib26","volume-title":"Geometric algebra for computer science: an object-oriented approach to geometry","author":"Dorst","year":"2007"},{"key":"2026081306443452700_bib27","doi-asserted-by":"crossref","DOI":"10.1201\/b18273","article-title":"Hamilton, Grassmann, and Clifford for computer vision and graphics","volume-title":"Understanding geometric algebra","author":"Kanatani","year":"2015"},{"key":"2026081306443452700_bib28","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511526022","volume-title":"Clifford algebras and spinors","author":"Lounesto","year":"2001"},{"key":"2026081306443452700_bib29","volume-title":"Introduction to geometric algebra computing","author":"Hildenbrand","year":"2019"},{"key":"2026081306443452700_bib30","doi-asserted-by":"crossref","first-page":"132","DOI":"10.1007\/s10851-010-0197-3","article-title":"The signal multi-vector","volume":"37","author":"Wietzke","year":"2010","journal-title":"J. Math. Imaging Vis."},{"key":"2026081306443452700_bib31","article-title":"Gradient of Clifford neural networks","volume-title":"Adv. Neural Inf. Process. Syst. Workshop on Data-driven and Differ. Simul., Surrog., and Solv.","author":"Maruyama","year":"2024"},{"key":"2026081306443452700_bib32","doi-asserted-by":"crossref","first-page":"1425","DOI":"10.1002\/mma.8639","article-title":"Normalization, square roots, and the exponential and logarithmic maps in geometric algebras of less than 6D","volume":"47","author":"De Keninck","year":"2022","journal-title":"Math. Meth. Appl. Sci."},{"key":"2026081306443452700_bib33","first-page":"29 306","article-title":"Geometric Clifford algebra networks","volume-title":"Proc. 40th Int. Conf. Mach. Learn.","author":"Ruhe","year":"2023"},{"key":"2026081306443452700_bib34","doi-asserted-by":"crossref","first-page":"291","DOI":"10.1007\/978-3-662-04621-0_12","article-title":"Clifford algebra multilayer perceptrons","volume-title":"Geometric computing with Clifford algebras","author":"Buchholz","year":"2001","edition":"ed."},{"key":"2026081306443452700_bib35","doi-asserted-by":"crossref","first-page":"291","DOI":"10.1007\/978-3-662-04621-0_12","article-title":"Introduction to neural computation in Clifford algebra","volume-title":"Geometric computing with Clifford algebras","author":"Buchholz","year":"2001","edition":"ed."},{"key":"2026081306443452700_bib36","article-title":"STAResNet: a network in spacetime algebra to solve Maxwell's PDEs","author":"Pepe","year":"2024"},{"key":"2026081306443452700_bib37","article-title":"Clifford neural layers for PDE modeling","author":"Brandstetter","year":"2023"},{"key":"2026081306443452700_bib38","doi-asserted-by":"crossref","first-page":"417","DOI":"10.1007\/978-3-319-74830-6_13","article-title":"Geometric Neurocomputing","volume-title":"Geometric algebra applications","author":"Bayro Corrochano","year":"2019"},{"key":"2026081306443452700_bib39","doi-asserted-by":"crossref","first-page":"925","DOI":"10.1016\/j.neunet.2008.03.004","article-title":"On Clifford neurons and Clifford multi-layer perceptrons","volume":"21","author":"Buchholz","year":"2008","journal-title":"Neural Netw."},{"key":"2026081306443452700_bib40","doi-asserted-by":"crossref","first-page":"377","DOI":"10.1007\/s00006-013-0378-4","article-title":"Applications of Clifford\u015b geometric algebra","volume":"23","author":"Hitzer","year":"2013","journal-title":"Adv. Appl. Clifford Algebr."},{"key":"2026081306443452700_bib41","doi-asserted-by":"crossref","first-page":"11 368","DOI":"10.1002\/mma.9575","article-title":"Survey of new applications of geometric algebra","volume":"47","author":"Hitzer","year":"2024","journal-title":"Math. Meth. Appl. Sci."},{"key":"2026081306443452700_bib42","first-page":"1","article-title":"GA-CNN: convolutional neural network based on geometric algebra for hyperspectral image classification","volume":"60","author":"Li","year":"2022","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"2026081306443452700_bib43","first-page":"62 922","article-title":"Clifford group equivariant neural networks","volume-title":"Proc. 37th Conf. Adv. Neural Inf. Process. Syst.","author":"Ruhe","year":"2023"},{"key":"2026081306443452700_bib44","doi-asserted-by":"crossref","first-page":"955","DOI":"10.1109\/TC.2014.2315652","article-title":"ConformalALU: a conformal geometric algebra coprocessor for medical image processing","volume":"64","author":"Franchini","year":"2015","journal-title":"IEEE Trans. Comput."},{"key":"2026081306443452700_bib45","doi-asserted-by":"crossref","first-page":"48","DOI":"10.1007\/s10851-008-0123-0","article-title":"The use of geometric algebra for 3D modeling and registration of medical data","volume":"34","author":"Bayro-Corrochano","year":"2009","journal-title":"J. Math. Imaging Vis."},{"key":"2026081306443452700_bib46","doi-asserted-by":"crossref","first-page":"729","DOI":"10.1007\/11578079_76","article-title":"Medical image segmentation and the use of geometric algebras in medical applications","volume-title":"Progress in pattern recognition, image analysis and applications","author":"Orozco-Aguirre","year":"2005"},{"key":"2026081306443452700_bib47","doi-asserted-by":"crossref","first-page":"19","DOI":"10.1007\/s00006-021-01197-6","article-title":"Multi-modal medical image fusion based on geometric algebra discrete cosine transform","volume":"32","author":"Wang","year":"2022","journal-title":"Adv. Appl. Clifford Algebr."},{"key":"2026081306443452700_bib48","doi-asserted-by":"crossref","first-page":"327","DOI":"10.1007\/s11227-024-06833-z","article-title":"A geometric algebra-enhanced network for skin lesion detection with diagnostic prior","volume":"81","author":"Wang","year":"2025","journal-title":"J. Supercomput."},{"key":"2026081306443452700_bib49","first-page":"185","article-title":"LaB-GATr: geometric algebra transformers for\u00c2\u00a0large biomedical surface and\u00c2\u00a0volume meshes","volume-title":"Med. Image Comput. Comput. Assist. Interv.","author":"Suk","year":"2024"},{"key":"2026081306443452700_bib50","first-page":"13771","article-title":"Clifford convolutional neural networks for lymphoblast image classification","volume-title":"Adv. Comp. Appl. Geom. Alg.","author":"Vieira","year":"2024"},{"key":"2026081306443452700_bib51","first-page":"1","article-title":"A methodology of three-dimensional medical image registration based on conformal geometric invariant","volume":"2014","author":"Feng","year":"2014","journal-title":"Math. Probl. Eng."},{"key":"2026081306443452700_bib52","article-title":"The Pygae Team. pygae\/clifford: v1.4.0","author":"Hadfield","year":"2021","journal-title":"Zenodo"},{"key":"2026081306443452700_bib53","article-title":"Torch-GA: building geometric algebra networks in PyTorch","author":"Pepe","year":"2025","journal-title":"Zenodo"},{"key":"2026081306443452700_bib54","article-title":"TensorFlow geometric algebra","author":"Kahlow","year":"2020","journal-title":"Zenodo"}],"container-title":["Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/royalsocietypublishing.org\/rsta\/article-pdf\/doi\/10.1098\/rsta.2025.0099\/6158494\/rsta.2025.0099.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/royalsocietypublishing.org\/rsta\/article-pdf\/doi\/10.1098\/rsta.2025.0099\/6158494\/rsta.2025.0099.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,13]],"date-time":"2026-08-13T10:44:44Z","timestamp":1786617884000},"score":1,"resource":{"primary":{"URL":"https:\/\/royalsocietypublishing.org\/rsta\/article\/384\/2326\/20250099\/482768\/Review-of-Clifford-algebra-based-AI-methods-in"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,13]]},"references-count":54,"journal-issue":{"issue":"2326","published-online":{"date-parts":[[2026,8,13]]},"published-print":{"date-parts":[[2026,8,13]]}},"URL":"https:\/\/doi.org\/10.1098\/rsta.2025.0099","relation":{},"ISSN":["1364-503X","1471-2962"],"issn-type":[{"value":"1364-503X","type":"print"},{"value":"1471-2962","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,13]]},"article-number":"20250099"}}