{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,29]],"date-time":"2025-12-29T18:54:36Z","timestamp":1767034476961,"version":"build-2065373602"},"reference-count":32,"publisher":"MDPI AG","issue":"20","license":[{"start":{"date-parts":[[2022,10,18]],"date-time":"2022-10-18T00:00:00Z","timestamp":1666051200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"FEDER Funds through the COMPETE 2020\u2014Programa Operacional Competitividade e Internacionaliza\u00e7\u00e3o (POCI) and P2020 with the Reference Project SmartOs","award":["POCI-01-0247-FEDER-039868","UIDB\/04436\/2020","2020.05708.BD","2020.03393.CEECIND"],"award-info":[{"award-number":["POCI-01-0247-FEDER-039868","UIDB\/04436\/2020","2020.05708.BD","2020.03393.CEECIND"]}]},{"DOI":"10.13039\/501100001871","name":"FCT national funds","doi-asserted-by":"publisher","award":["POCI-01-0247-FEDER-039868","UIDB\/04436\/2020","2020.05708.BD","2020.03393.CEECIND"],"award-info":[{"award-number":["POCI-01-0247-FEDER-039868","UIDB\/04436\/2020","2020.05708.BD","2020.03393.CEECIND"]}],"id":[{"id":"10.13039\/501100001871","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001871","name":"Stimulus of Scientific Employment","doi-asserted-by":"publisher","award":["POCI-01-0247-FEDER-039868","UIDB\/04436\/2020","2020.05708.BD","2020.03393.CEECIND"],"award-info":[{"award-number":["POCI-01-0247-FEDER-039868","UIDB\/04436\/2020","2020.05708.BD","2020.03393.CEECIND"]}],"id":[{"id":"10.13039\/501100001871","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Energy expenditure is a key rehabilitation outcome and is starting to be used in robotics-based rehabilitation through human-in-the-loop control to tailor robot assistance towards reducing patients\u2019 energy effort. However, it is usually assessed by indirect calorimetry which entails a certain degree of invasiveness and provides delayed data, which is not suitable for controlling robotic devices. This work proposes a deep learning-based tool for steady-state energy expenditure estimation based on more ergonomic sensors than indirect calorimetry. The study innovates by estimating the energy expenditure in assisted and non-assisted conditions and in slow gait speeds similarly to impaired subjects. This work explores and benchmarks the long short-term memory (LSTM) and convolutional neural network (CNN) as deep learning regressors. As inputs, we fused inertial data, electromyography, and heart rate signals measured by on-body sensors from eight healthy volunteers walking with and without assistance from an ankle-foot exoskeleton at 0.22, 0.33, and 0.44 m\/s. LSTM and CNN were compared against indirect calorimetry using a leave-one-subject-out cross-validation technique. Results showed the suitability of this tool, especially CNN, that demonstrated root-mean-squared errors of 0.36 W\/kg and high correlation (\u03c1 &gt; 0.85) between target and estimation (R\u00af2 = 0.79). CNN was able to discriminate the energy expenditure between assisted and non-assisted gait, basal, and walking energy expenditure, throughout three slow gait speeds. CNN regressor driven by kinematic and physiological data was shown to be a more ergonomic technique for estimating the energy expenditure, contributing to the clinical assessment in slow and robotic-assisted gait and future research concerning human-in-the-loop control.<\/jats:p>","DOI":"10.3390\/s22207913","type":"journal-article","created":{"date-parts":[[2022,10,19]],"date-time":"2022-10-19T00:58:51Z","timestamp":1666141131000},"page":"7913","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":25,"title":["Deep Learning-Based Energy Expenditure Estimation in Assisted and Non-Assisted Gait Using Inertial, EMG, and Heart Rate Wearable Sensors"],"prefix":"10.3390","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4207-4951","authenticated-orcid":false,"given":"Jo\u00e3o M.","family":"Lopes","sequence":"first","affiliation":[{"name":"Center for MicroElectroMechanical Systems (CMEMS), University of Minho, 4800-058 Guimar\u00e3es, Portugal"},{"name":"LABBELS\u2014Associate Laboratory, 4710-057 Braga\/4800-058 Guimar\u00e3es, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9547-3051","authenticated-orcid":false,"given":"Joana","family":"Figueiredo","sequence":"additional","affiliation":[{"name":"Center for MicroElectroMechanical Systems (CMEMS), University of Minho, 4800-058 Guimar\u00e3es, Portugal"},{"name":"LABBELS\u2014Associate Laboratory, 4710-057 Braga\/4800-058 Guimar\u00e3es, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4885-4924","authenticated-orcid":false,"given":"Pedro","family":"Fonseca","sequence":"additional","affiliation":[{"name":"Porto Biomechanics Laboratory (LABIOMEP), Faculty of Sports, University of Porto, 4200-450 Porto, Portugal"}]},{"given":"Jo\u00e3o J.","family":"Cerqueira","sequence":"additional","affiliation":[{"name":"Life and Health Sciences Research Institute (ICVS), School of Medicine, University of Minho, 4710-057 Braga, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4109-2939","authenticated-orcid":false,"given":"Jo\u00e3o P.","family":"Vilas-Boas","sequence":"additional","affiliation":[{"name":"Porto Biomechanics Laboratory (LABIOMEP), Faculty of Sports, University of Porto, 4200-450 Porto, Portugal"},{"name":"Faculty of Sports and CIFI2D, University of Porto, 4200-450 Porto, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0023-7203","authenticated-orcid":false,"given":"Cristina P.","family":"Santos","sequence":"additional","affiliation":[{"name":"Center for MicroElectroMechanical Systems (CMEMS), University of Minho, 4800-058 Guimar\u00e3es, Portugal"},{"name":"LABBELS\u2014Associate Laboratory, 4710-057 Braga\/4800-058 Guimar\u00e3es, Portugal"}]}],"member":"1968","published-online":{"date-parts":[[2022,10,18]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1177\/1687814018783627","article-title":"Advanced technology for gait rehabilitation: An overview","volume":"10","author":"Mikolajczyk","year":"2018","journal-title":"Adv. Mech. Eng."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"619","DOI":"10.1016\/j.apmr.2015.11.007","article-title":"Energy Expenditure and Cost during Walking after Stroke: A Systematic Review","volume":"97","author":"Kramer","year":"2016","journal-title":"Arch. Phys. Med. Rehabil."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"206","DOI":"10.1097\/01.PEP.0000145930.84009.23","article-title":"Comparison of the Energy Expenditure Index and Oxygen Consumption Index During Self-Paced Walking in Children with Spastic Diplegia Cerebral Palsy and Children Without Physical Disabilities","volume":"16","author":"Norman","year":"2004","journal-title":"Pediatr. Phys. Ther."},{"key":"ref_4","first-page":"463","article-title":"Derivation of formulae used to calculate energy expenditure in man","volume":"41","author":"Brockway","year":"1987","journal-title":"Hum. Nutr. Clin. Nutr."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1555","DOI":"10.1016\/j.medengphy.2014.08.005","article-title":"Control strategies for effective robot assisted gait rehabilitation: The state of art and future prospects","volume":"36","author":"Cao","year":"2014","journal-title":"Med. Eng. Phys."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"132","DOI":"10.1016\/j.mechatronics.2015.04.005","article-title":"Recent development of mechanisms and control strategies for robot-assisted lower limb rehabilitation","volume":"31","author":"Meng","year":"2015","journal-title":"Mechatronics"},{"key":"ref_7","first-page":"406","article-title":"Towards human-knee orthosis interaction based on adaptive impedance control through stiffness adjustment","volume":"2017","author":"Figueiredo","year":"2017","journal-title":"IEEE Int. Conf. Rehabil. Robot."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"444","DOI":"10.1177\/0954411919898293","article-title":"Robotic orthoses for gait rehabilitation: An overview of mechanical design and control strategies","volume":"234","author":"Jamwal","year":"2020","journal-title":"Proc. Inst. Mech. Eng. Part H J. Eng. Med."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Lopes, J.M., Pinheiro, C., Figueiredo, J., Reis, L.P., and Santos, C.P. (2020, January 15\u201317). Assist-as-needed Impedance Control Strategy for a Wearable Ankle Robotic Orthosis. Proceedings of the IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC), Ponta Delgada, Portugal.","DOI":"10.1109\/ICARSC49921.2020.9096186"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1186\/s12984-016-0111-3","article-title":"Biomechanical walking mechanisms underlying the metabolic reduction caused by an autonomous exoskeleton","volume":"13","author":"Mooney","year":"2016","journal-title":"J. Neuroeng. Rehabil."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Seo, K., Lee, J., Lee, Y., Ha, T., and Shim, Y. (2016, January 16\u201321). Fully Autonomous Hip Exoskeleton Saves Metabolic Cost of Walking. Proceedings of the IEEE International Conference on Rehabilitation Robotics and Automation (ICRA), Stockholm, Sweden.","DOI":"10.1109\/ICRA.2016.7487663"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"eaai9084","DOI":"10.1126\/scitranslmed.aai9084","article-title":"A soft robotic exosuit improves walking in patients after stroke","volume":"9","author":"Awad","year":"2017","journal-title":"Sci. Transl. Med."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"74","DOI":"10.1186\/s12984-018-0381-z","article-title":"Exoskeleton assistance symmetry matters: Unilateral assistance reduces metabolic cost, but relatively less than bilateral assistance","volume":"15","author":"Malcolm","year":"2018","journal-title":"J. Neuroeng. Rehabil."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"7157","DOI":"10.1038\/s41598-019-43628-2","article-title":"Gait training using a robotic hip exoskeleton improves metabolic gait efficiency in the elderly","volume":"9","author":"Martini","year":"2019","journal-title":"Sci. Rep."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"67","DOI":"10.1186\/s12984-019-0535-7","article-title":"Rapid energy expenditure estimation for ankle assisted and inclined loaded walking","volume":"16","author":"Slade","year":"2019","journal-title":"J. Neuroeng. Rehabil."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"1280","DOI":"10.1126\/science.aal5054","article-title":"Human-in-the-loop optimization of exoskeleton assistance during walking","volume":"356","author":"Zhang","year":"2017","journal-title":"Science (80-)"},{"key":"ref_17","first-page":"1","article-title":"\u201cBody-in-the-Loop\u201d Optimization of Assistive Robotic Devices: A Validation Study","volume":"2016","author":"Koller","year":"2016","journal-title":"Robot. Sci. Syst."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"eaar5438","DOI":"10.1126\/scirobotics.aar5438","article-title":"Human-in-the-loop optimization of hip assistance with a soft exosuit during walking","volume":"3","author":"Ding","year":"2018","journal-title":"Sci. Robot."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"717","DOI":"10.1152\/japplphysiol.00714.2018","article-title":"Evaluating physiological signal salience for estimating metabolic energy cost from wearable sensors","volume":"126","author":"Ingraham","year":"2019","journal-title":"J. Appl. Physiol."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"45738","DOI":"10.1038\/srep45738","article-title":"Prediction of oxygen uptake dynamics by machine learning analysis of wearable sensors during activities of daily living","volume":"7","author":"Beltrame","year":"2017","journal-title":"Sci. Rep."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"1226","DOI":"10.1152\/japplphysiol.00600.2016","article-title":"Estimating oxygen uptake and energy expenditure during treadmill walking by neural network analysis of easy-to-obtain inputs","volume":"121","author":"Beltrame","year":"2016","journal-title":"J. Appl. Physiol."},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Zhu, J., Pande, A., Mohapatra, P., and Han, J.J. (2015, January 14\u201317). Using Deep Learning for Energy Expenditure Estimation with wearable sensors. Proceedings of the 2015 17th International Conference on E-health Networking, Application & Services (HealthCom), Boston, MA, USA.","DOI":"10.1109\/HealthCom.2015.7454554"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"311","DOI":"10.1016\/j.gaitpost.2009.11.011","article-title":"Differences in self-selected and fastest-comfortable walking in post-stroke hemiparetic persons","volume":"31","author":"Beaman","year":"2010","journal-title":"Gait Posture"},{"key":"ref_24","unstructured":"(2022, October 17). Polar Research and Technology. Polar H10 Heart Rate Sensor System. Available online: https:\/\/www.polar.com\/en\/img\/static\/whitepapers\/pdf\/polar-h10-heart-rate-sensor-white-paper.pdf."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"361","DOI":"10.1016\/S1050-6411(00)00027-4","article-title":"Development of recommendations for SEMG sensors and sensor placement procedures","volume":"10","author":"Hermens","year":"2000","journal-title":"J. Electromyogr. Kinesiol."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Al-Amri, M., Nicholas, K., Button, K., Sparkes, V., Sheeran, L., and Davies, J.L. (2018). Inertial measurement units for clinical movement analysis: Reliability and concurrent validity. Sensors, 18.","DOI":"10.3390\/s18030719"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"280","DOI":"10.1055\/s-2001-13816","article-title":"Validation of the COSMED K4b2 Portable Metabolic System","volume":"22","author":"McLaughlin","year":"2001","journal-title":"Int. J. Sports Med."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"130","DOI":"10.1152\/japplphysiol.01212.2003","article-title":"Reducing the time period of steady state does not affect the accuracy of energy expenditure measurements by indirect calorimetry","volume":"97","author":"Reeves","year":"2004","journal-title":"J. Appl. Physiol."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Winter, D.A. (2009). Biomechanics and Motor Control of Human Movement, John Wiley & Sons, Inc.. [4th ed.].","DOI":"10.1002\/9780470549148"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"1573","DOI":"10.1016\/j.jbiomech.2010.01.027","article-title":"Filtering the surface EMG signal: Movement artifact and baseline noise contamination","volume":"43","author":"Kuznetsov","year":"2010","journal-title":"J. Biomech."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Gerdle, B., Karlsson, S., Day, S., and Djupsj\u00f6backa, M. (1999). Acquisition, Processing and Analysis of the Surface Electromyogram. Modern Techniques in Neuroscience Research, Springer.","DOI":"10.1007\/978-3-642-58552-4_26"},{"key":"ref_32","unstructured":"Goodfellow, I., Bengio, Y., and Courville, A. (2016). Deep Learning, The MIT Press."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/22\/20\/7913\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T00:56:12Z","timestamp":1760144172000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/22\/20\/7913"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,10,18]]},"references-count":32,"journal-issue":{"issue":"20","published-online":{"date-parts":[[2022,10]]}},"alternative-id":["s22207913"],"URL":"https:\/\/doi.org\/10.3390\/s22207913","relation":{},"ISSN":["1424-8220"],"issn-type":[{"type":"electronic","value":"1424-8220"}],"subject":[],"published":{"date-parts":[[2022,10,18]]}}}