{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T10:01:50Z","timestamp":1780999310228,"version":"3.54.1"},"reference-count":70,"publisher":"Association for Computing Machinery (ACM)","issue":"3","license":[{"start":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T00:00:00Z","timestamp":1780963200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Comput. Healthcare"],"published-print":{"date-parts":[[2026,7,30]]},"abstract":"<jats:p>\n                    Photoplethysmography-based Blood Pressure (BP) estimation is a challenging task, particularly on resource-constrained wearable devices. However, fully on-board processing is desirable to ensure user data confidentiality. Recent Deep Neural Networks (DNNs) have achieved high BP estimation accuracy by reconstructing BP waveforms or directly regressing BP values, but their large memory, computation, and energy requirements hinder deployment on wearables. This work introduces a fully automated DNN design pipeline that combines hardware-aware Neural Architecture Search, pruning, and Mixed-Precision Search to generate accurate yet compact BP prediction models optimized for ultra-low-power multi-core Systems-on-Chip (SoCs). Starting from state-of-the-art baseline models on four public datasets, our optimized networks achieve up to 7.99% lower error with a 7.5\n                    <jats:inline-formula content-type=\"math\/tex\">\n                      <jats:tex-math notation=\"LaTeX\" version=\"MathJax\">\\(\\times\\)<\/jats:tex-math>\n                    <\/jats:inline-formula>\n                    parameter reduction, or up to 83\n                    <jats:inline-formula content-type=\"math\/tex\">\n                      <jats:tex-math notation=\"LaTeX\" version=\"MathJax\">\\(\\times\\)<\/jats:tex-math>\n                    <\/jats:inline-formula>\n                    fewer parameters with negligible accuracy loss. All models fit within 512 kB of memory on our target SoC (GreenWaves\u2019 GAP8), requiring less than 55 kB and achieving an average inference latency of 142 ms and energy consumption of 7.25 mJ. Patient-specific fine-tuning further improves accuracy by up to 64%, enabling fully autonomous, low-cost BP monitoring on wearables.\n                  <\/jats:p>","DOI":"10.1145\/3809503","type":"journal-article","created":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T13:32:13Z","timestamp":1777901533000},"page":"1-27","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["End-to-end Automated Deep Neural Network Optimization for PPG-based Blood Pressure Estimation on Wearables"],"prefix":"10.1145","volume":"7","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-5551-9200","authenticated-orcid":false,"given":"Francesco","family":"Carlucci","sequence":"first","affiliation":[{"name":"Politecnico di Torino, Torino, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-3337-1367","authenticated-orcid":false,"given":"Giovanni","family":"Pollo","sequence":"additional","affiliation":[{"name":"Politecnico di Torino, Torino, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3467-5033","authenticated-orcid":false,"given":"Xiaying","family":"Wang","sequence":"additional","affiliation":[{"name":"ETH Zurich, Zurich, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1369-9688","authenticated-orcid":false,"given":"Massimo","family":"Poncino","sequence":"additional","affiliation":[{"name":"Politecnico di Torino, Torino, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9046-5618","authenticated-orcid":false,"given":"Enrico","family":"Macii","sequence":"additional","affiliation":[{"name":"Politecnico di Torino, Torino, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8068-3806","authenticated-orcid":false,"given":"Luca","family":"Benini","sequence":"additional","affiliation":[{"name":"Universit\u00e0 di Bologna, Bologna, Italy and ETH Zurich, Z\u00fcrich, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9666-5194","authenticated-orcid":false,"given":"Sara","family":"Vinco","sequence":"additional","affiliation":[{"name":"Politecnico di Torino, Torino, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6215-8220","authenticated-orcid":false,"given":"Alessio","family":"Burrello","sequence":"additional","affiliation":[{"name":"Politecnico di Torino, Torino, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2872-7071","authenticated-orcid":false,"given":"Daniele","family":"Jahier Pagliari","sequence":"additional","affiliation":[{"name":"Politecnico di Torino, Torino, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2026,6,9]]},"reference":[{"key":"e_1_3_1_2_2","doi-asserted-by":"publisher","DOI":"10.1109\/IEEESTD.2014.6882122"},{"key":"e_1_3_1_3_2","doi-asserted-by":"publisher","DOI":"10.3390\/s21062167"},{"key":"e_1_3_1_4_2","doi-asserted-by":"publisher","DOI":"10.1097\/MBP.0000000000000639"},{"key":"e_1_3_1_5_2","doi-asserted-by":"publisher","DOI":"10.1016\/J.BSPC.2023.104782"},{"key":"e_1_3_1_6_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41440-023-01258-2"},{"key":"e_1_3_1_7_2","doi-asserted-by":"crossref","unstructured":"Amir Arjomand Amin Boudesh Farnoush Bayatmakou Kenneth B. Kent and Arash Mohammadi. 2025. TransfoRhythm A transformer architecture conductive to blood pressure estimation via solo PPG signal capturing. arXiv:2404.15352. Retrieved from https:\/\/arxiv.org\/abs\/2404.15352","DOI":"10.2139\/ssrn.5371308"},{"key":"e_1_3_1_8_2","doi-asserted-by":"publisher","DOI":"10.1145\/3587256"},{"key":"e_1_3_1_9_2","unstructured":"Luca Benfenati Sofia Belloni Alessio Burrello Panagiotis Kasnesis Xiaying Wang Luca Benini Massimo Poncino Enrico Macii and Daniele Jahier Pagliari. 2024. EnhancePPG: Improving PPG-based heart rate estimation with self-supervision and augmentation. arXiv:2412.17860. Retrieved from https:\/\/arxiv.org\/abs\/2412.17860"},{"key":"e_1_3_1_10_2","doi-asserted-by":"publisher","DOI":"10.1109\/EMBC44109.2020.9176231"},{"key":"e_1_3_1_11_2","doi-asserted-by":"publisher","DOI":"10.1145\/3387902.3394038"},{"key":"e_1_3_1_12_2","doi-asserted-by":"publisher","DOI":"10.1109\/BioCAS61083.2024.10798404"},{"key":"e_1_3_1_13_2","unstructured":"Alessio Burrello Angelo Garofalo Nazareno Bruschi Giuseppe Tagliavini Davide Rossi and Francesco Conti. 2020. DORY: Automatic end-to-end deployment of real-world DNNs on low-cost IoT MCUs. arXiv:2008.07127. Retrieved from https:\/\/arxiv.org\/abs\/2008.07127"},{"key":"e_1_3_1_14_2","doi-asserted-by":"crossref","unstructured":"Alessio Burrello Francesco Bianco Morghet Moritz Scherer Simone Benatti Luca Benini Enrico Macii Massimo Poncino and Daniele Jahier Pagliari. 2022. Bioformers: Embedding transformers for ultra-low power sEMG-based gesture recognition. arXiv:2203.12932. Retrieved from https:\/\/arxiv.org\/abs\/2203.12932","DOI":"10.23919\/DATE54114.2022.9774639"},{"key":"e_1_3_1_15_2","doi-asserted-by":"publisher","DOI":"10.1145\/3487910"},{"key":"e_1_3_1_16_2","doi-asserted-by":"publisher","DOI":"10.1109\/TBCAS.2021.3122017"},{"key":"e_1_3_1_17_2","unstructured":"Zhaowei Cai and Nuno Vasconcelos. 2020. Rethinking differentiable search for mixed-precision neural networks. arXiv:2004.05795. Retrieved from https:\/\/arxiv.org\/abs\/2004.05795"},{"key":"e_1_3_1_18_2","doi-asserted-by":"publisher","DOI":"10.3390\/s21010156"},{"key":"e_1_3_1_19_2","doi-asserted-by":"publisher","DOI":"10.1016\/J.COMPBIOMED.2021.104877"},{"key":"e_1_3_1_20_2","doi-asserted-by":"publisher","DOI":"10.3390\/s25010040"},{"key":"e_1_3_1_21_2","unstructured":"Jungwook Choi Zhuo Wang Swagath Venkataramani Pierce I-Jen Chuang Vijayalakshmi Srinivasan and Kailash Gopalakrishnan. 2018. PACT: Parameterized clipping activation for quantized neural networks. arXiv:1805.06085. Retrieved from http:\/\/arxiv.org\/abs\/1805.06085"},{"key":"e_1_3_1_22_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2023.3284458"},{"key":"e_1_3_1_23_2","doi-asserted-by":"publisher","DOI":"10.1109\/EMBC44109.2020.9175699"},{"key":"e_1_3_1_24_2","doi-asserted-by":"publisher","DOI":"10.1016\/J.COMPBIOMED.2019.103392"},{"key":"e_1_3_1_25_2","doi-asserted-by":"publisher","DOI":"10.1161\/HYPERTENSIONAHA.119.14240"},{"key":"e_1_3_1_26_2","doi-asserted-by":"crossref","unstructured":"Angelo Garofalo Manuele Rusci Francesco Conti Davide Rossi and Luca Benini. 2019. PULP-NN: Accelerating quantized neural networks on parallel ultra-low-power RISC-V processors. arXiv:1908.11263. Retrieved from http:\/\/arxiv.org\/abs\/1908.11263","DOI":"10.1098\/rsta.2019.0155"},{"key":"e_1_3_1_27_2","doi-asserted-by":"publisher","DOI":"10.2196\/13048"},{"key":"e_1_3_1_28_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41597-023-02020-6"},{"key":"e_1_3_1_29_2","doi-asserted-by":"publisher","DOI":"10.1109\/BSN.2016.7516258"},{"key":"e_1_3_1_30_2","unstructured":"Andrew G. Howard Menglong Zhu Bo Chen Dmitry Kalenichenko Weijun Wang Tobias Weyand Marco Andreetto and Hartwig Adam. 2017. MobileNets: Efficient convolutional neural networks for mobile vision applications. arXiv:1704.04861. Retrieved from http:\/\/arxiv.org\/abs\/1704.04861"},{"key":"e_1_3_1_31_2","doi-asserted-by":"publisher","DOI":"10.1016\/J.BSPC.2021.103404"},{"key":"e_1_3_1_32_2","unstructured":"iMDsoft. 2025. iMDsoft MetaVision ICU. Retrieved March 27 2025 from https:\/\/imd-soft.com\/metavision\/icu\/"},{"key":"e_1_3_1_33_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11892-019-1161-2"},{"key":"e_1_3_1_34_2","doi-asserted-by":"crossref","unstructured":"D. Jahier Pagliari M. Risso B. A. Motetti and A. Burrello. 2023. PLiNIO: A user-friendly library of gradient-based methods for complexity-aware DNN optimization. arXiv:2307.09488. Retrieved from https:\/\/arxiv.org\/abs\/2307.09488","DOI":"10.1109\/FDL59689.2023.10272045"},{"key":"e_1_3_1_35_2","doi-asserted-by":"publisher","DOI":"10.1038\/SDATA.2016.35"},{"key":"e_1_3_1_36_2","doi-asserted-by":"publisher","DOI":"10.1109\/AICERA.2014.6908199"},{"key":"e_1_3_1_37_2","doi-asserted-by":"publisher","DOI":"10.1145\/3699512"},{"key":"e_1_3_1_38_2","doi-asserted-by":"publisher","DOI":"10.1109\/ISCAS.2015.7168806"},{"key":"e_1_3_1_39_2","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2016.2580904"},{"key":"e_1_3_1_40_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2020.3040257"},{"key":"e_1_3_1_41_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2024.3417940"},{"key":"e_1_3_1_42_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cose.2023.103488"},{"key":"e_1_3_1_43_2","doi-asserted-by":"publisher","DOI":"10.3390\/s20195606"},{"key":"e_1_3_1_44_2","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2018.20"},{"key":"e_1_3_1_45_2","doi-asserted-by":"publisher","DOI":"10.1016\/J.BSPC.2025.107585"},{"key":"e_1_3_1_46_2","doi-asserted-by":"publisher","DOI":"10.1109\/EMBC46164.2021.9630488"},{"key":"e_1_3_1_47_2","unstructured":"Hanxiao Liu Karen Simonyan and Yiming Yang. 2018. DARTS: Differentiable architecture search. arXiv:1806.09055. Retrieved from https:\/\/arxiv.org\/abs\/1806.09055"},{"issue":"1","key":"e_1_3_1_48_2","doi-asserted-by":"crossref","first-page":"42","DOI":"10.1016\/j.amjmed.2005.08.051","article-title":"Trends in blood pressure, hypertension control, and stroke mortality: The Minnesota heart survey","volume":"119","author":"Luepker Russell V.","year":"2006","unstructured":"Russell V. Luepker, Donna K. Arnett, David R. Jacobs, Susan J. Duval, Aaron R. Folsom, Christopher Armstrong, and Henry Blackburn. 2006. Trends in blood pressure, hypertension control, and stroke mortality: The Minnesota heart survey. The American Journal of Medicine 119, 1 (Jan. 2006), 42\u201349.","journal-title":"The American Journal of Medicine"},{"key":"e_1_3_1_49_2","doi-asserted-by":"publisher","DOI":"10.3390\/jcm7100316"},{"key":"e_1_3_1_50_2","doi-asserted-by":"publisher","unstructured":"Agnes S. Meidert and Bernd Saugel. 2017. Techniques for non-invasive monitoring of arterial blood pressure. Frontiers in Medicine 4 (Jan. 2017) 231. DOI: 10.3389\/fmed.2017.00231","DOI":"10.3389\/fmed.2017.00231"},{"key":"e_1_3_1_51_2","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2024.3449084"},{"key":"e_1_3_1_52_2","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2015.2441951"},{"key":"e_1_3_1_53_2","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.i4482"},{"key":"e_1_3_1_54_2","doi-asserted-by":"publisher","DOI":"10.3390\/bioengineering10040400"},{"key":"e_1_3_1_55_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICMLA52953.2021.00241"},{"key":"e_1_3_1_56_2","doi-asserted-by":"crossref","unstructured":"Olaf Ronneberger Philipp Fischer and Thomas Brox. 2015. U-Net: Convolutional networks for biomedical image segmentation. arXiv:1505.04597. Retrieved from https:\/\/arxiv.org\/abs\/1505.04597","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_3_1_57_2","doi-asserted-by":"publisher","DOI":"10.1109\/TIM.2022.3212750"},{"key":"e_1_3_1_58_2","doi-asserted-by":"publisher","DOI":"10.1109\/CIC.2002.1166854"},{"key":"e_1_3_1_59_2","unstructured":"Nordic Semiconductor. 2024. Nordic II. Retrieved May 16 2024 from https:\/\/www.nordicsemi.com\/"},{"key":"e_1_3_1_60_2","doi-asserted-by":"publisher","DOI":"10.1109\/BioCAS61083.2024.10798356"},{"key":"e_1_3_1_61_2","unstructured":"Daniel Stoller Sebastian Ewert and Simon Dixon. 2018. Wave-U-Net: A multi-scale neural network for end-to-end audio source separation. arXiv:1806.03185. Retrieved from https:\/\/arxiv.org\/abs\/1806.03185"},{"key":"e_1_3_1_62_2","doi-asserted-by":"publisher","DOI":"10.1145\/3555776.3577747"},{"key":"e_1_3_1_63_2","unstructured":"Greenwaves Technologies. 2024. Gap8. Retrieved May 16 2025 from https:\/\/github.com\/GreenWaves-Technologies\/gap_sdk"},{"key":"e_1_3_1_64_2","doi-asserted-by":"publisher","DOI":"10.1016\/J.BSPC.2024.106741"},{"key":"e_1_3_1_65_2","doi-asserted-by":"publisher","DOI":"10.5555\/3295222.3295349"},{"key":"e_1_3_1_66_2","doi-asserted-by":"publisher","DOI":"10.1093\/bjaceaccp\/mkm022"},{"key":"e_1_3_1_67_2","doi-asserted-by":"publisher","DOI":"10.1161\/01.HYP.21.4.504"},{"key":"e_1_3_1_68_2","doi-asserted-by":"publisher","DOI":"10.1109\/EMBC48229.2022.9871215"},{"key":"e_1_3_1_69_2","doi-asserted-by":"publisher","DOI":"10.1007\/s40820-025-02003-9"},{"key":"e_1_3_1_70_2","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2025.3535788"},{"key":"e_1_3_1_71_2","doi-asserted-by":"publisher","DOI":"10.3390\/electronics11091378"}],"container-title":["ACM Transactions on Computing for Healthcare"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3809503","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T09:37:25Z","timestamp":1780997845000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3809503"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,9]]},"references-count":70,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,7,30]]}},"alternative-id":["10.1145\/3809503"],"URL":"https:\/\/doi.org\/10.1145\/3809503","relation":{},"ISSN":["2691-1957","2637-8051"],"issn-type":[{"value":"2691-1957","type":"print"},{"value":"2637-8051","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,9]]},"assertion":[{"value":"2025-10-21","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-04-03","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-06-09","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}