{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:43:26Z","timestamp":1785545006495,"version":"3.56.0"},"reference-count":39,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Med Inform"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Irregularly sampled data, as a common data structure in the medical field, is frequently observed in emergency clinical datasets. It poses problems such as unequal sampling time intervals and frequencies, making it difficult to align the data without losing information for input into models. Meanwhile, due to information loss in the dataset, it is also difficult for the model to effectively analyze the overall data variation and predict the complex and dynamic conditions of emergency patients in the future.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to address the impact of issues such as missing values and irregularly sampled time intervals on prediction results. This paper proposes a method that converts time-series data into a graph network structure and develops a prediction model for graph data to better support complex tasks.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n In this paper, feature channels and measurement times in time-series data are constructed as nodes, and feature measurements at time points are constructed as edge weights. The point feature space and edge weight space are extracted by a graph convolutional neural network and a gated convolutional attention mechanism network, respectively, and feature fusion is carried out under the action of the learnable head.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n The proposed method was tested on 4 datasets and compared with both irregularly sampled time-series predictors and state-of-the-art methods for complete data. In addition, a dedicated analysis was conducted leveraging data from 5 key clinical treatment phases. In single-target classification and single-target regression tasks, the method proposed in this study outperforms other comparative methods in most experiments (accuracy=0.74\u20100.88, area under the curve=0.70\u20100.87, and mean absolute error=0.038\u20100.061). Meanwhile, this study constructs clinical phenotypes during patients\u2019 hospitalization as multitarget prediction tasks, and the proposed model also outperforms comparative methods in multitarget classification tasks (mean accuracy=0.75\u20100.91 and mean area under the curve=0.70\u20100.75). Ablation experiments verify the effectiveness of fusion between the node feature space and the edge feature space. The interpretability analysis of time points identifies key features that require focused attention for patients across different target tasks and time periods.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n The proposed model can effectively analyze medical data with missing values and uneven sampling. It achieves high prediction accuracy in both regression prediction tasks and classification tasks. The graph network-based model structure endows the model with better interpretability, enabling effective exploration of factors that trigger changes.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/81145","type":"journal-article","created":{"date-parts":[[2026,7,3]],"date-time":"2026-07-03T19:22:56Z","timestamp":1783106576000},"page":"e81145-e81145","source":"Crossref","is-referenced-by-count":0,"title":["Graph Network Feature Space Fusion for Predicting Irregularly Sampled Medical Time-Series Data: Deep Learning Model Development and Validation Study"],"prefix":"10.2196","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2220-4621","authenticated-orcid":false,"given":"Tianle","family":"Hong","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-4263-6422","authenticated-orcid":false,"given":"Zedong","family":"Ren","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-6249-9541","authenticated-orcid":false,"given":"Junfei","family":"Fang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5013-606X","authenticated-orcid":false,"given":"Shichao","family":"Quan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-2508-2150","authenticated-orcid":false,"given":"Jingye","family":"Pan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3992-7738","authenticated-orcid":false,"given":"Yezhi","family":"Lin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,7,3]]},"reference":[{"issue":"1","key":"R1","doi-asserted-by":"publisher","DOI":"10.1186\/s13054-024-04860-z","article-title":"Use of artificial intelligence in critical care: opportunities and obstacles","volume":"28","author":"Pinsky","journal-title":"Crit Care"},{"issue":"2","key":"R2","doi-asserted-by":"publisher","first-page":"132","DOI":"10.15441\/ceem.23.041","article-title":"Current challenges in adopting machine learning to critical care and emergency medicine","volume":"10","author":"Kang","journal-title":"Clin Exp Emerg Med"},{"issue":"8","key":"R3","doi-asserted-by":"publisher","first-page":"1749","DOI":"10.1097\/CCM.0000000000000330","article-title":"Empiric antibiotic treatment reduces mortality in severe sepsis and septic shock from the first hour: results from a guideline-based performance improvement program","volume":"42","author":"Ferrer","journal-title":"Crit Care Med"},{"key":"R4","doi-asserted-by":"publisher","DOI":"10.2196\/45846","article-title":"Analyzing and forecasting pediatric fever clinic visits in high frequency using ensemble time-series methods after the COVID-19 pandemic in Hangzhou, China: retrospective study","volume":"11","author":"Zhang","journal-title":"JMIR Med Inform"},{"key":"R5","doi-asserted-by":"publisher","DOI":"10.2196\/49643","article-title":"Data-driven identification of factors that influence the quality of adverse event reports: 15-year interpretable machine learning and time-series analyses of VigiBase and QUEST","volume":"12","author":"Choo","journal-title":"JMIR Med Inform"},{"key":"R6","doi-asserted-by":"publisher","DOI":"10.2196\/45515","article-title":"Clinical timing-sequence warning models for serious bacterial infections in adults based on machine learning: retrospective study","volume":"25","author":"Liu","journal-title":"J Med Internet Res"},{"issue":"1","key":"R7","doi-asserted-by":"publisher","DOI":"10.1186\/s13054-020-03379-3","article-title":"Prediction of hypotension events with physiologic vital sign signatures in the intensive care unit","volume":"24","author":"Yoon","journal-title":"Crit Care"},{"issue":"4","key":"R8","doi-asserted-by":"publisher","first-page":"334","DOI":"10.1016\/j.annemergmed.2018.11.036","article-title":"Development and evaluation of a machine learning model for the early identification of patients at risk for sepsis","volume":"73","author":"Delahanty","journal-title":"Ann Emerg Med"},{"issue":"1","key":"R9","doi-asserted-by":"publisher","DOI":"10.1186\/s12873-021-00501-8","article-title":"A machine learning model to predict critical care outcomes in patient with chest pain visiting the emergency department","volume":"21","author":"Wu","journal-title":"BMC Emerg Med"},{"issue":"8","key":"R10","doi-asserted-by":"publisher","first-page":"4270","DOI":"10.1109\/JBHI.2022.3172549","article-title":"Multi-view integrative attention-based deep representation learning for irregular clinical time-series data","volume":"26","author":"Lee","journal-title":"IEEE J Biomed Health Inform"},{"key":"R11","doi-asserted-by":"publisher","DOI":"10.1016\/j.artmed.2021.102112","article-title":"Improving clinical outcome predictions using convolution over medical entities with multimodal learning","volume":"117","author":"Bardak","journal-title":"Artif Intell Med"},{"issue":"7","key":"R12","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TKDE.2022.3181780","article-title":"Time-aware context-gated graph attention network for clinical risk prediction","volume":"35","author":"Xu","journal-title":"IEEE Trans Knowl Data Eng"},{"issue":"2","key":"R13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3470889","article-title":"Graph sequence neural network with an attention mechanism for traffic speed prediction","volume":"13","author":"Lu","journal-title":"ACM Trans Intell Syst Technol"},{"issue":"2","key":"R14","doi-asserted-by":"publisher","DOI":"10.3390\/e25020247","article-title":"Deep spatio-temporal graph network with self-optimization for air quality prediction","volume":"25","author":"Jin","journal-title":"Entropy (Basel)"},{"key":"R15","doi-asserted-by":"publisher","DOI":"10.1016\/j.bdr.2021.100237","article-title":"Time-series graph network for sea surface temperature prediction","volume":"25","author":"Sun","journal-title":"Big Data Research"},{"key":"R16","doi-asserted-by":"publisher","DOI":"10.1016\/j.ins.2023.119056","article-title":"Spatiotemporal dynamic graph convolutional network for traffic speed forecasting","volume":"641","author":"Yin","journal-title":"Inf Sci (Ny)"},{"key":"R17","doi-asserted-by":"publisher","DOI":"10.1016\/j.engappai.2024.109530","article-title":"Time-lagged relation graph neural network for multivariate time series forecasting","volume":"139","author":"Feng","journal-title":"Eng Appl Artif Intell"},{"key":"R18","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2023.122729","article-title":"Dynamic multi-fusion spatio-temporal graph neural network for multivariate time series forecasting","volume":"241","author":"He","journal-title":"Expert Syst Appl"},{"issue":"15","key":"R19","doi-asserted-by":"publisher","first-page":"2396","DOI":"10.3390\/math12152396","article-title":"A time-series feature-extraction methodology based on multiscale overlapping windows, adaptive KDE, and continuous entropic and information functionals","volume":"12","author":"Squicciarini","journal-title":"Mathematics"},{"key":"R20","unstructured":"Zandieh A Han I Daliri M Karbasi A . KDEformer: accelerating transformers via kernel density estimation. Presented at: ICML\u201923: Proceedings of the 40th International Conference on Machine Learning; Jul 23-29, 2023; Honolulu, HI. [doi: 10.5555\/3618408.3620109]"},{"issue":"4","key":"R21","doi-asserted-by":"publisher","first-page":"2645","DOI":"10.1007\/s11063-020-10404-7","article-title":"Semi-supervised classification of graph convolutional networks with laplacian rank constraints","volume":"54","author":"Zhang","journal-title":"Neural Process Lett"},{"key":"R22","unstructured":"Vaswani A Shazeer N Parmar N et al. Attention is all you need. arXiV. Preprint posted online on Jun 12, 2017. [doi: 10.48550\/ARXIV.1706.03762]"},{"issue":"23","key":"R23","doi-asserted-by":"publisher","first-page":"E215","DOI":"10.1161\/01.cir.101.23.e215","article-title":"PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals","volume":"101","author":"Goldberger","journal-title":"Circulation"},{"key":"R24","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2016.35","article-title":"MIMIC-III, a freely accessible critical care database","volume":"3","author":"Johnson","journal-title":"Sci Data"},{"issue":"1","key":"R25","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1038\/s41597-023-01945-2","article-title":"Author correction: MIMIC-IV, a freely accessible electronic health record dataset","volume":"10","author":"Johnson","journal-title":"Sci Data"},{"key":"R26","doi-asserted-by":"crossref","unstructured":"Baytas IM Xiao C Zhang X Wang F Jain AK Zhou J . Patient subtyping via time-aware LSTM networks. Presented at: KDD \u201917: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; Aug 13-17, 2017:65-74; Halifax, Canada. [doi: 10.1145\/3097983.3097997]","DOI":"10.1145\/3097983.3097997"},{"issue":"1","key":"R27","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-018-24271-9","article-title":"Recurrent neural networks for multivariate time series with missing values","volume":"8","author":"Che","journal-title":"Sci Rep"},{"key":"R28","volume":"2021","author":"Shukla","journal-title":"Int Conf Learn Represent"},{"key":"R29","doi-asserted-by":"crossref","unstructured":"Chen Y Fang Y Li D Ren K Sun W Wang Y . ContiFormer: continuous-time transformer for irregular time series modeling. Presented at: Advances in Neural Information Processing Systems 36; Nov 28 to Dec 9, 2022; New Orleans, LA. [doi: 10.52202\/075280-2042]","DOI":"10.52202\/075280-2042"},{"issue":"4","key":"R30","doi-asserted-by":"publisher","first-page":"640","DOI":"10.1109\/TPAMI.2016.2572683","article-title":"Fully convolutional networks for semantic segmentation","volume":"39","author":"Shelhamer","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"R31","unstructured":"Bai S Kolter JZ Koltun V . An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. arXiv. Preprint posted online on Mar 4, 2018. [doi: 10.48550\/arXiv.1803.01271]"},{"key":"R32","unstructured":"Oreshkin BN Carpov D Chapados N Bengio Y . N-BEATS: neural basis expansion analysis for interpretable time series forecasting. arXiv. Preprint posted online on May 24, 2019. [doi: 10.48550\/arXiv.1905.10437]"},{"key":"R33","unstructured":"Zhang Y Yan J . Crossformer: transformer utilizing cross-dimension dependency for multivariate time series forecasting. Presented at: Proceedings of the International Conference on Learning Representations (ICLR); May 1-5, 2023; Kigali, Rwanda. 2023.URL: https:\/\/openreview.net\/forum?id=vSVLM2j9eie [Accessed 02-06-2026]"},{"key":"R34","doi-asserted-by":"crossref","unstructured":"Zhou B Dey N Schlemper J et al. DSFormer: a dual-domain self-supervised transformer for accelerated multi-contrast MRI reconstruction. Presented at: 2023 IEEE\/CVF Winter Conference on Applications of Computer Vision (WACV); Jan 2-7, 2023:4955-4964; Waikoloa, HI. [doi: 10.1109\/WACV56688.2023.00494]","DOI":"10.1109\/WACV56688.2023.00494"},{"key":"R35","doi-asserted-by":"publisher","DOI":"10.1177\/20552076251361603","article-title":"MSPO: a machine learning hyperparameter optimization method for enhanced breast cancer image classification","volume":"11","author":"Li","journal-title":"Digit Health"},{"key":"R36","doi-asserted-by":"publisher","DOI":"10.1177\/20552076241249661","article-title":"Optimizing lung cancer classification through hyperparameter tuning","volume":"10","author":"Nabeel","journal-title":"Digit Health"},{"key":"R37","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2023.105120","article-title":"Thoughts of brain EEG signal-to-text conversion using weighted feature fusion-based Multiscale Dilated Adaptive DenseNet with Attention Mechanism","volume":"86","author":"Yang","journal-title":"Biomed Signal Process Control"},{"key":"R38","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2024.106501","article-title":"Optimizing diabetic retinopathy detection with inception-V4 and dynamic version of snow leopard optimization algorithm","volume":"96","author":"Yang","journal-title":"Biomed Signal Process Control"},{"issue":"9","key":"R39","doi-asserted-by":"publisher","DOI":"10.1016\/j.heliyon.2024.e30625","article-title":"Colon and lung cancer classification from multi-modal images using resilient and efficient neural network architectures","volume":"10","author":"Uddin","journal-title":"Heliyon"}],"container-title":["JMIR Medical Informatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e81145","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T22:26:33Z","timestamp":1785536793000},"score":1,"resource":{"primary":{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e81145"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,3]]},"references-count":39,"URL":"https:\/\/doi.org\/10.2196\/81145","relation":{},"ISSN":["2291-9694"],"issn-type":[{"value":"2291-9694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,3]]},"article-number":"v14i1e81145"}}