{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T22:10:17Z","timestamp":1790633417292,"version":"4.1.0"},"reference-count":60,"publisher":"MDPI AG","issue":"4","license":[{"start":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T00:00:00Z","timestamp":1775779200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100009619","name":"Japan Agency for Medical Research and Development","doi-asserted-by":"publisher","award":["JP23ym0126812"],"award-info":[{"award-number":["JP23ym0126812"]}],"id":[{"id":"10.13039\/100009619","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100009619","name":"Japan Agency for Medical Research and Development","doi-asserted-by":"publisher","award":["JP24ym0126812"],"award-info":[{"award-number":["JP24ym0126812"]}],"id":[{"id":"10.13039\/100009619","id-type":"DOI","asserted-by":"publisher"}]},{"award":["JP23ym0126812"],"award-info":[{"award-number":["JP23ym0126812"]}],"id":[{"id":"https:\/\/ror.org\/004rtk039","id-type":"ROR","asserted-by":"publisher"}]},{"award":["JP24ym0126812"],"award-info":[{"award-number":["JP24ym0126812"]}],"id":[{"id":"https:\/\/ror.org\/004rtk039","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Bioengineering"],"abstract":"<jats:p>Brain\u2013machine interfaces (BMIs) aim to decode motor intentions from neural activity to enable direct control of external devices. However, most existing decoders rely on monolithic architectures that fail to capture the distinct neural representations of different joint movement directions, limiting their generalizability. In this work, we propose a Single-Direction CNN-LSTM decoder inspired by motor cortex encoding mechanisms, which separately models extension and flexion dynamics through parallel CNN-LSTM branches. Each branch extracts spatial\u2013temporal features from neural spike data and predicts directional joint variables, which are then combined by subtraction to yield the net angular velocity and torque of upper-limb joints. Using invasive recordings from a macaque during a 2D center-out reaching task, we demonstrate that our decoder achieves comparable performance to a conventional CNN-LSTM when trained on all tasks, while significantly outperforming both CNN-LSTM and linear regression baselines in cross-target generalization scenarios. Moreover, the model can capture physiologically meaningful co-contraction patterns, providing richer insights into motor control. These results suggest that incorporating neuroscience-inspired modular decoding into deep neural architectures enhances robustness and adaptability across tasks, offering a promising pathway for BMI applications in prosthetics and rehabilitation.<\/jats:p>","DOI":"10.3390\/bioengineering13040440","type":"journal-article","created":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T07:55:12Z","timestamp":1775807712000},"page":"440","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Neuroscience-Inspired Deep Learning Brain\u2013Machine Interface Decoder"],"prefix":"10.3390","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-2766-8993","authenticated-orcid":false,"given":"Hong-Yun","family":"Ou","sequence":"first","affiliation":[{"name":"Graduate School of Life Science and Technology, Institute of Science Tokyo, Yokohama 226-0026, Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Takahiro","family":"Hasegawa","sequence":"additional","affiliation":[{"name":"Graduate School of Life Science and Technology, Institute of Science Tokyo, Yokohama 226-0026, Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2072-2003","authenticated-orcid":false,"given":"Osamu","family":"Fukayama","sequence":"additional","affiliation":[{"name":"Graduate School of Information Science and Technology, The University of Tokyo, Bunkyo, Tokyo 113-8656, Japan"},{"name":"Center for Information and Neural Networks, Advanced ICT Research Institute, National Institute of Information Science and Technology, Suita 565-0871, Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Eizo","family":"Miyashita","sequence":"additional","affiliation":[{"name":"Graduate School of Life Science and Technology, Institute of Science Tokyo, Yokohama 226-0026, Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,4,10]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"536","DOI":"10.1016\/j.tins.2006.07.004","article-title":"Brain\u2013machine interfaces: Past, present and future","volume":"29","author":"Lebedev","year":"2006","journal-title":"TRENDS Neurosci."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"671","DOI":"10.1016\/j.tics.2021.04.003","article-title":"Interface, interaction, and intelligence in generalized brain\u2013computer interfaces","volume":"25","author":"Gao","year":"2021","journal-title":"Trends Cogn. Sci."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"0044","DOI":"10.34133\/cbsystems.0044","article-title":"Neural decoding for intracortical brain\u2013computer interfaces","volume":"4","author":"Dong","year":"2023","journal-title":"Cyborg Bionic Syst."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Orban, M., Elsamanty, M., Guo, K., Zhang, S., and Yang, H. (2022). A review of brain activity and EEG-based brain\u2013computer interfaces for rehabilitation application. Bioengineering, 9.","DOI":"10.3390\/bioengineering9120768"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Faye, I., and Islam, M.R. (2022). EEG channel selection techniques in motor imagery applications: A review and new perspectives. Bioengineering, 9.","DOI":"10.3390\/bioengineering9120726"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"2408","DOI":"10.1109\/TNSRE.2024.3421551","article-title":"A review of Motor Brain-Computer interfaces using Intracranial Electroencephalography based on surface electrodes and depth electrodes","volume":"32","author":"Wu","year":"2024","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_7","first-page":"69","article-title":"Effect of artifact removal on EEG based motor imagery BCI applications","volume":"Volume 12984","author":"Anjum","year":"2024","journal-title":"Proceedings of the Fourth International Conference on Computer Vision and Information Technology (CVIT 2023)"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Abu-Rmileh, A., Zakkay, E., Shmuelof, L., and Shriki, O. (2019). Co-adaptive training improves efficacy of a multi-day EEG-based motor imagery BCI training. Front. Hum. Neurosci., 13.","DOI":"10.3389\/fnhum.2019.00362"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Altuwaijri, G.A., and Muhammad, G. (2022). Electroencephalogram-based motor imagery signals classification using a multi-branch convolutional neural network model with attention blocks. Bioengineering, 9.","DOI":"10.3390\/bioengineering9070323"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"702357","DOI":"10.1155\/2010\/702357","article-title":"A survey of stimulation methods used in SSVEP-based BCIs","volume":"2010","author":"Zhu","year":"2010","journal-title":"Comput. Intell. Neurosci."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"2610","DOI":"10.1109\/TBME.2006.886577","article-title":"Frequency recognition based on canonical correlation analysis for SSVEP-based BCIs","volume":"53","author":"Lin","year":"2006","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Gaur, P., Pachori, R.B., Wang, H., and Prasad, G. (2015). An empirical mode decomposition based filtering method for classification of motor-imagery EEG signals for enhancing brain-computer interface. Proceedings of the 2015 International Joint Conference on Neural Networks (IJCNN), IEEE.","DOI":"10.1109\/IJCNN.2015.7280754"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"1437","DOI":"10.1038\/s41562-024-01941-5","article-title":"One hundred years of EEG for brain and behaviour research","volume":"8","author":"Mushtaq","year":"2024","journal-title":"Nat. Hum. Behav."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"577","DOI":"10.3390\/signals3030035","article-title":"A survey on denoising techniques of electroencephalogram signals using wavelet transform","volume":"3","author":"Grobbelaar","year":"2022","journal-title":"Signals"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"407","DOI":"10.1038\/nrn3241","article-title":"The origin of extracellular fields and currents\u2014EEG, ECoG, LFP and spikes","volume":"13","author":"Anastassiou","year":"2012","journal-title":"Nat. Rev. Neurosci."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"1416","DOI":"10.1126\/science.3749885","article-title":"Neuronal population coding of movement direction","volume":"233","author":"Georgopoulos","year":"1986","journal-title":"Science"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"80","DOI":"10.1162\/089976606774841585","article-title":"Bayesian population decoding of motor cortical activity using a Kalman filter","volume":"18","author":"Wu","year":"2006","journal-title":"Neural Comput."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"2105","DOI":"10.1109\/TBME.2021.3137184","article-title":"MIN2Net: End-to-end multi-task learning for subject-independent motor imagery EEG classification","volume":"69","author":"Autthasan","year":"2021","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"056013","DOI":"10.1088\/1741-2552\/aace8c","article-title":"EEGNet: A compact convolutional neural network for EEG-based brain\u2013computer interfaces","volume":"15","author":"Lawhern","year":"2018","journal-title":"J. Neural Eng."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"5391","DOI":"10.1002\/hbm.23730","article-title":"Deep learning with convolutional neural networks for EEG decoding and visualization","volume":"38","author":"Schirrmeister","year":"2017","journal-title":"Hum. Brain Mapp."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Glaser, J.I., Benjamin, A.S., Chowdhury, R.H., Perich, M.G., Miller, L.E., and Kording, K.P. (2020). Machine learning for neural decoding. eneuro, 7.","DOI":"10.1523\/ENEURO.0506-19.2020"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"1031","DOI":"10.1038\/s41586-023-06377-x","article-title":"A high-performance speech neuroprosthesis","volume":"620","author":"Willett","year":"2023","journal-title":"Nature"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Chandrasekaran, S., Wandelt, S.K., Jangam, A., Elias, Z., Ibroci, E., Maffei, C., Rosenthal, I.A., Ramdeo, R., Kim, J.W., and Xu, J. (2025). Restoring Cortically Mediated Movement and Sensation in Complete Tetraplegia. medRxiv.","DOI":"10.1101\/2025.08.19.25330198"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"026023","DOI":"10.1088\/1741-2552\/ac5d69","article-title":"Decoding ECoG signal into 3D hand translation using deep learning","volume":"19","author":"Martin","year":"2022","journal-title":"J. Neural Eng."},{"key":"ref_25","unstructured":"Ji, C. (2024). Explainable mst-ecognet decode visual information from ecog signal. arXiv."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"036009","DOI":"10.1088\/1741-2552\/aa9dbe","article-title":"Decoding of finger trajectory from ECoG using deep learning","volume":"15","author":"Xie","year":"2018","journal-title":"J. Neural Eng."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"826","DOI":"10.1152\/jn.1997.77.2.826","article-title":"Reaching movements with similar hand paths but different arm orientations. I. Activity of individual cells in motor cortex","volume":"77","author":"Scott","year":"1997","journal-title":"J. Neurophysiol."},{"key":"ref_28","unstructured":"Dey, S., Yoshida, T., Foerster, R.H., Ernst, M., Schmalz, T., Carnier, R.M., and Schilling, A.F. (2021). A hybrid approach for dynamically training a torque prediction model for devising a human-machine interface control strategy. arXiv."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"1943","DOI":"10.1109\/TIE.2015.2504579","article-title":"State variables of the arm may be encoded by single neuron activity in the monkey motor cortex","volume":"63","author":"Miyashita","year":"2015","journal-title":"IEEE Trans. Ind. Electron."},{"key":"ref_30","unstructured":"Tian, K., Zhao, S., Zhang, Y., and Yu, S. (2025). Multi-dimensional Neural Decoding with Orthogonal Representations for Brain-Computer Interfaces. arXiv."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"1527","DOI":"10.1523\/JNEUROSCI.02-11-01527.1982","article-title":"On the relations between the direction of two-dimensional arm movements and cell discharge in primate motor cortex","volume":"2","author":"Georgopoulos","year":"1982","journal-title":"J. Neurosci."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"026011","DOI":"10.1088\/1741-2552\/abde8a","article-title":"Robust and accurate decoding of hand kinematics from entire spiking activity using deep learning","volume":"18","author":"Ahmadi","year":"2021","journal-title":"J. Neural Eng."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"6180","DOI":"10.1109\/LRA.2021.3092269","article-title":"Combining unsupervised muscle co-contraction estimation with bio-feedback allows augmented kinesthetic teaching","volume":"6","author":"Meattini","year":"2021","journal-title":"IEEE Robot. Autom. Lett."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"387","DOI":"10.1016\/S1364-6613(02)01963-0","article-title":"Neural delays, visual motion and the flash-lag effect","volume":"6","author":"Nijhawan","year":"2002","journal-title":"Trends Cogn. Sci."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"056038","DOI":"10.1088\/1741-2552\/ac97c3","article-title":"Validation of a non-invasive, real-time, human-in-the-loop model of intracortical brain-computer interfaces","volume":"19","author":"Awasthi","year":"2022","journal-title":"J. Neural Eng."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Noh, S.H. (2021). Analysis of gradient vanishing of RNNs and performance comparison. Information, 12.","DOI":"10.3390\/info12110442"},{"key":"ref_38","first-page":"1875","article-title":"Nonparametric regression using deep neural networks with ReLU activation function","volume":"48","year":"2020","journal-title":"Ann. Statist."},{"key":"ref_39","unstructured":"Bhanja, S., and Das, A. (2018). Impact of data normalization on deep neural network for time series forecasting. arXiv."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"917","DOI":"10.1038\/s41551-024-01297-1","article-title":"Enhanced control of a brain\u2013computer interface by tetraplegic participants via neural-network-mediated feature extraction","volume":"9","author":"Haghi","year":"2024","journal-title":"Nat. Biomed. Eng."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"805","DOI":"10.1038\/s41592-018-0109-9","article-title":"Inferring single-trial neural population dynamics using sequential auto-encoders","volume":"15","author":"Pandarinath","year":"2018","journal-title":"Nat. Methods"},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"600","DOI":"10.1109\/TNSRE.2022.3156786","article-title":"Lower-limb joint torque prediction using LSTM neural networks and transfer learning","volume":"30","author":"Zhang","year":"2022","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Shah, S., Haghi, B., Kellis, S., Bashford, L., Kramer, D., Lee, B., Liu, C., Andersen, R., and Emami, A. (2019). Decoding kinematics from human parietal cortex using neural networks. Proceedings of the 2019 9th International IEEE\/EMBS Conference on Neural Engineering (NER), IEEE.","DOI":"10.1109\/NER.2019.8717137"},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"De Feo, V., Boi, F., Safaai, H., Onken, A., Panzeri, S., and Vato, A. (2017). State-dependent decoding algorithms improve the performance of a bidirectional bmi in anesthetized rats. Front. Neurosci., 11.","DOI":"10.3389\/fnins.2017.00269"},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"289","DOI":"10.1111\/j.2517-6161.1995.tb02031.x","article-title":"Controlling the false discovery rate: A practical and powerful approach to multiple testing","volume":"57","author":"Benjamini","year":"1995","journal-title":"J. R. Stat. Soc. Ser. B (Methodol.)"},{"key":"ref_46","unstructured":"Kingma, D.P., and Ba, J. (2014). Adam: A method for stochastic optimization. arXiv."},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Rao, C.R., Rao, C.R., Statistiker, M., Rao, C.R., and Rao, C.R. (1973). Linear Statistical Inference and Its Applications, Wiley.","DOI":"10.1002\/9780470316436"},{"key":"ref_48","doi-asserted-by":"crossref","first-page":"026014","DOI":"10.1088\/1741-2552\/abda0b","article-title":"Generalized neural decoders for transfer learning across participants and recording modalities","volume":"18","author":"Peterson","year":"2021","journal-title":"J. Neural Eng."},{"key":"ref_49","doi-asserted-by":"crossref","first-page":"4235","DOI":"10.1152\/jn.00095.2007","article-title":"Temporal complexity and heterogeneity of single-neuron activity in premotor and motor cortex","volume":"97","author":"Churchland","year":"2007","journal-title":"J. Neurophysiol."},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"518","DOI":"10.1109\/TBME.2025.3586870","article-title":"Intracortical Brain-Machine Interfaces with High-Performance Neural Decoding through Efficient Transfer Meta-learning","volume":"73","author":"Chen","year":"2025","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_51","doi-asserted-by":"crossref","unstructured":"Iman, M., Arabnia, H.R., and Rasheed, K. (2023). A review of deep transfer learning and recent advancements. Technologies, 11.","DOI":"10.3390\/technologies11020040"},{"key":"ref_52","doi-asserted-by":"crossref","first-page":"556","DOI":"10.1109\/TNSRE.2021.3059166","article-title":"Dynamic joint domain adaptation network for motor imagery classification","volume":"29","author":"Hong","year":"2021","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_53","doi-asserted-by":"crossref","first-page":"1363","DOI":"10.1109\/TNSRE.2021.3096874","article-title":"Decoding imagined speech based on deep metric learning for intuitive BCI communication","volume":"29","author":"Lee","year":"2021","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_54","doi-asserted-by":"crossref","unstructured":"Halmich, C., H\u00f6schler, L., Schranz, C., and Borgelt, C. (2025). Data augmentation of time-series data in human movement biomechanics: A scoping review. PLoS ONE, 20.","DOI":"10.1371\/journal.pone.0327038"},{"key":"ref_55","doi-asserted-by":"crossref","unstructured":"Banks, C.L., Huang, H.J., Little, V.L., and Patten, C. (2017). Electromyography exposes heterogeneity in muscle co-contraction following stroke. Front. Neurol., 8.","DOI":"10.3389\/fneur.2017.00699"},{"key":"ref_56","doi-asserted-by":"crossref","first-page":"pgae076","DOI":"10.1093\/pnasnexus\/pgae076","article-title":"Transfer learning promotes acquisition of individual BCI skills","volume":"3","author":"Kumar","year":"2024","journal-title":"PNAS Nexus"},{"key":"ref_57","doi-asserted-by":"crossref","first-page":"e70024","DOI":"10.1002\/brx2.70024","article-title":"Brain\u2013computer interfaces in 2023\u20132024","volume":"3","author":"Chen","year":"2025","journal-title":"Brain-x"},{"key":"ref_58","doi-asserted-by":"crossref","first-page":"nwaf086","DOI":"10.1093\/nsr\/nwaf086","article-title":"Canine EEG helps human: Cross-species and cross-modality epileptic seizure detection via multi-space alignment","volume":"12","author":"Wang","year":"2025","journal-title":"Natl. Sci. Rev."},{"key":"ref_59","doi-asserted-by":"crossref","first-page":"1309","DOI":"10.1109\/JBHI.2023.3347672","article-title":"A physics-informed low-shot adversarial learning for semg-based estimation of muscle force and joint kinematics","volume":"28","author":"Shi","year":"2023","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_60","doi-asserted-by":"crossref","unstructured":"Kim, Y., Kim, C., and Hwangbo, J. (2022). Learning forward dynamics model and informed trajectory sampler for safe quadruped navigation. arXiv.","DOI":"10.15607\/RSS.2022.XVIII.069"}],"container-title":["Bioengineering"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2306-5354\/13\/4\/440\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T08:40:22Z","timestamp":1775810422000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2306-5354\/13\/4\/440"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,10]]},"references-count":60,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2026,4]]}},"alternative-id":["bioengineering13040440"],"URL":"https:\/\/doi.org\/10.3390\/bioengineering13040440","relation":{"has-preprint":[{"id-type":"doi","id":"10.64898\/2026.02.07.703641","asserted-by":"object"}]},"ISSN":["2306-5354"],"issn-type":[{"value":"2306-5354","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,10]]}}}