{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T10:30:45Z","timestamp":1787567445881,"version":"build-2736575974"},"reference-count":52,"publisher":"IOP Publishing","issue":"8","license":[{"start":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T00:00:00Z","timestamp":1787529600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T00:00:00Z","timestamp":1787529600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Physiol. Meas."],"published-print":{"date-parts":[[2026,8,31]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    <jats:italic>Objective.<\/jats:italic>\n                    Deep learning models for photoplethysmography (PPG)-based arrhythmia detection in intensive care are often evaluated by average accuracy, while conformal prediction (CP) is used to add marginal coverage guarantees. We ask whether aggregate accuracy and marginal conformal coverage are sufficient evidence of clinical reliability under intensive care unit (ICU) class imbalance and patient heterogeneity.\n                    <jats:italic>Approach.<\/jats:italic>\n                    Using the MIMIC-III-Ext-PPG benchmark with 826 critically ill patients and over 1.1 million 30 s PPG segments, we conduct a conformal reliability audit of PPG-only arrhythmia classification. We deliberately use a stress-case classifier with superficially reassuring aggregate metrics but weak rare-class behavior, and test whether accuracy and marginal CP conceal per-class and per-patient failures. We evaluate two predictors, three global conformal score constructions (LAC, adaptive prediction sets (APS), and Sorted Adaptive Prediction Sets (SAPS)), class-conditional calibration, an empirical RC3P-style class-aware sensitivity analysis, prevalence reweighting, and signal-quality-aware variants, with uncertainty quantified by patient-clustered resampling.\n                    <jats:italic>Main results.<\/jats:italic>\n                    The stress-case classifier attains 65.0% accuracy but only 0.388 balanced accuracy, revealing severe rare-class failure despite superficially reassuring aggregate performance. Under global CP, marginal coverage reaches 90.0%, yet critical-rhythm coverage is only 7.5%, with 267 of 268 critical-rhythm segments not assigned to the correct top class. This collapse persists across predictors and global scores (APS: 96.0% overall but 23.1% critical; SAPS: 90.2% overall but 7.8% critical) and survives full-benchmark prevalence reweighting. Class-conditional calibration raises critical-rhythm coverage to 0.825, but the patient-clustered interval [0.578, 0.985] shows that target-level reliability remains uncertified for the rarest class. More than a third of patients fall below target even among those with many segments. Signal quality is not the dominant failure axis: low-quality segments are not under-covered, and quality-weighted calibration offers no advantage over simple baselines.\n                    <jats:italic>Significance.<\/jats:italic>\n                    In ICU PPG arrhythmia monitoring, marginal coverage is not clinical reliability. Systems should be evaluated by per-class and per-patient coverage with patient-clustered uncertainty before claims of trustworthiness.\n                  <\/jats:p>","DOI":"10.1088\/1361-6579\/ae92e8","type":"journal-article","created":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T22:53:58Z","timestamp":1785452038000},"page":"085029","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Conformal reliability assessment of PPG-based ICU arrhythmia classification under class imbalance and patient heterogeneity"],"prefix":"10.1088","volume":"47","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7445-3510","authenticated-orcid":true,"given":"Ismail Hakki","family":"Kinalioglu","sequence":"first","affiliation":[{"name":"Sel\u00e7uk University"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2026,8,24]]},"reference":[{"key":"pmeaae92e8bib1","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2026.110345","type":"journal-article","article-title":"Real-time embedded detection of bradycardia, tachycardia and arrhythmia using photoplethysmography and on-device convolutional neural networks","volume":"121","author":"Al Fahoum","year":"2026","journal-title":"Biomed. Signal Process. Control"},{"key":"pmeaae92e8bib2","doi-asserted-by":"publisher","first-page":"R1","DOI":"10.1088\/0967-3334\/28\/3\/R01","type":"journal-article","article-title":"Photoplethysmography and its application in clinical physiological measurement","volume":"28","author":"Allen","year":"2007","journal-title":"Physiol. Meas."},{"key":"pmeaae92e8bib3","doi-asserted-by":"publisher","first-page":"494","DOI":"10.1561\/2200000101","type":"journal-article","article-title":"Conformal prediction: a gentle introduction","volume":"16","author":"Angelopoulos","year":"2023","journal-title":"Found. Trends Mach. Learn."},{"key":"pmeaae92e8bib4","doi-asserted-by":"publisher","DOI":"10.1016\/j.rinam.2025.100589","type":"journal-article","article-title":"Conformal prediction across scales: finite-sample coverage with hierarchical efficiency","volume":"26","author":"Baheri","year":"2025","journal-title":"Res. Appl. Math."},{"key":"pmeaae92e8bib5","doi-asserted-by":"publisher","first-page":"816","DOI":"10.1214\/23-AOS2276","type":"journal-article","article-title":"Conformal prediction beyond exchangeability","volume":"51","author":"Barber","year":"2023","journal-title":"Ann. Stat."},{"key":"pmeaae92e8bib6","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1007\/s13755-025-00412-z","type":"journal-article","article-title":"Conformal uncertainty quantification to evaluate predictive fairness of foundation AI model for skin lesion classes across patient demographics","volume":"14","author":"Bhattacharyya","year":"2026","journal-title":"Health Inf. Sci. Syst."},{"key":"pmeaae92e8bib7","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6579\/acead2","type":"journal-article","article-title":"The 2023 wearable photoplethysmography roadmap","volume":"44","author":"Charlton","year":"2023","journal-title":"Physiol. Meas."},{"key":"pmeaae92e8bib8","doi-asserted-by":"publisher","first-page":"719","DOI":"10.1038\/s41551-023-01056-8","type":"journal-article","article-title":"Algorithmic fairness in artificial intelligence for medicine and healthcare","volume":"7","author":"Chen","year":"2023","journal-title":"Nat. Biomed. Eng."},{"key":"pmeaae92e8bib9","first-page":"pp 785","type":"conference-proceedings","article-title":"XGBoost: a scalable tree boosting system","author":"Chen","year":"2016"},{"key":"pmeaae92e8bib10","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2023.102861","type":"journal-article","article-title":"Uncertainty aware training to improve deep learning model calibration for classification of cardiac MR images","volume":"88","author":"Dawood","year":"2023","journal-title":"Med. Image Anal."},{"key":"pmeaae92e8bib11","article-title":"A few observations on sample-conditional coverage in conformal prediction","author":"Duchi","year":"2025","type":"preprint"},{"key":"pmeaae92e8bib12","doi-asserted-by":"publisher","first-page":"21","DOI":"10.3390\/bioengineering3040021","type":"journal-article","article-title":"Optimal signal quality index for photoplethysmogram signals","volume":"3","author":"Elgendi","year":"2016","journal-title":"Bioengineering"},{"key":"pmeaae92e8bib13","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1038\/s44328-024-00002-1","type":"journal-article","article-title":"Optimal signal quality index for remote photoplethysmogram sensing","volume":"1","author":"Elgendi","year":"2024","journal-title":"npj Biosensing"},{"key":"pmeaae92e8bib14","doi-asserted-by":"publisher","first-page":"1100","DOI":"10.1093\/jrsssb\/qkaf008","type":"journal-article","article-title":"Conformal prediction with conditional guarantees","volume":"87","author":"Gibbs","year":"2025","journal-title":"J. R. Stat. Soc. B"},{"key":"pmeaae92e8bib15","first-page":"pp 1321","type":"conference-proceedings","article-title":"On calibration of modern neural networks","author":"Guo","year":"2017"},{"key":"pmeaae92e8bib16","first-page":"pp 20331","type":"conference-proceedings","article-title":"Conformal prediction for deep classifier via label ranking","volume":"vol 235","author":"Huang","year":"2024"},{"key":"pmeaae92e8bib17","doi-asserted-by":"publisher","first-page":"917","DOI":"10.1007\/s10618-019-00619-1","type":"journal-article","article-title":"Deep learning for time series classification: a review","volume":"33","author":"Ismail Fawaz","year":"2019","journal-title":"Data Mining Knowl. Disc."},{"key":"pmeaae92e8bib18","doi-asserted-by":"publisher","first-page":"1936","DOI":"10.1007\/s10618-020-00710-y","type":"journal-article","article-title":"InceptionTime: finding AlexNet for time series classification","volume":"34","author":"Ismail Fawaz","year":"2020","journal-title":"Data Mining Knowl. Disc."},{"key":"pmeaae92e8bib19","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2025.112051","type":"journal-article","article-title":"Verifiably robust conformal prediction for probabilistic guarantees under adversarial attacks","volume":"170","author":"Jeary","year":"2026","journal-title":"Pattern Recognit."},{"key":"pmeaae92e8bib20","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2016.35","type":"journal-article","article-title":"MIMIC-III, a freely accessible critical care database","volume":"3","author":"Johnson","year":"2016","journal-title":"Sci. Data"},{"key":"pmeaae92e8bib21","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2026.103953","type":"journal-article","article-title":"Reliable uncertainty quantification for 2D\/3D anatomical landmark localization using multi-output conformal prediction","volume":"110","author":"Jonkers","year":"2026","journal-title":"Med. Image Anal."},{"key":"pmeaae92e8bib22","doi-asserted-by":"publisher","DOI":"10.1177\/03000605241232519","type":"journal-article","article-title":"Improving accuracy of vascular access quality classification in hemodialysis patients using deep learning with k highest score feature selection","volume":"52","author":"Julkaew","year":"2024","journal-title":"J. Int. Med. Res."},{"key":"pmeaae92e8bib23","author":"Kao","year":"2025","type":"book"},{"key":"pmeaae92e8bib24","doi-asserted-by":"publisher","first-page":"551","DOI":"10.1016\/j.procs.2019.04.074","type":"journal-article","article-title":"A real-time PPG quality assessment approach for healthcare internet-of-things","volume":"151","author":"Kasaeyan Naeini","year":"2019"},{"key":"pmeaae92e8bib25","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.21469011","type":"dataset","article-title":"PPG ICU Arrhythmia Conformal Reliability Audit \u2014 Reproducibility Package (Version 2.1.2)","author":"Kinalioglu","year":"2026","unstructured":"Kinalioglu I H 2026 PPG ICU Arrhythmia Conformal Reliability Audit \u2014 Reproducibility Package (Version 2.1.2) Zenodo 10.5281\/zenodo.21469011"},{"key":"pmeaae92e8bib26","article-title":"A critical perspective on finite sample conformal prediction theory in medical applications","author":"Kladny","year":"2025","type":"preprint"},{"key":"pmeaae92e8bib27","doi-asserted-by":"publisher","first-page":"1094","DOI":"10.1080\/01621459.2017.1307116","type":"journal-article","article-title":"Distribution-free predictive inference for regression","volume":"113","author":"Lei","year":"2018","journal-title":"J. Am. Stat. Assoc."},{"key":"pmeaae92e8bib28","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2025.108798","type":"journal-article","article-title":"Machine-learning guided differentiation between photoplethysmography waveforms of supraventricular and ventricular origin","volume":"267","author":"Manninger","year":"2025","journal-title":"Comput. Methods Programs Biomed."},{"key":"pmeaae92e8bib29","doi-asserted-by":"publisher","DOI":"10.1016\/j.spl.2024.110350","type":"journal-article","article-title":"universal distribution of the empirical coverage in split conformal prediction","volume":"219","author":"Marques","year":"2025","journal-title":"Stat. Prob. Lett."},{"key":"pmeaae92e8bib30","doi-asserted-by":"publisher","first-page":"3868","DOI":"10.1109\/TMI.2020.3006437","type":"journal-article","article-title":"Confidence calibration and predictive uncertainty estimation for deep medical image segmentation","volume":"39","author":"Mehrtash","year":"2020","journal-title":"IEEE Trans. Med. Imaging"},{"key":"pmeaae92e8bib31","article-title":"MIMIC-III waveform database matched subset","author":"Moody","year":"2020","type":"other"},{"key":"pmeaae92e8bib32","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2026.109831","type":"journal-article","article-title":"Machine-learning for photoplethysmography analysis: Benchmarking feature, image and signal-based approaches","volume":"120","author":"Moulaeifard","year":"2026b","journal-title":"Biomed. Signal Process. Control"},{"key":"pmeaae92e8bib33","doi-asserted-by":"publisher","DOI":"10.13026\/r6k1-xt76","type":"other","article-title":"MIMIC-III-Ext-PPG: a PPG benchmark dataset for cardiorespiratory analysis (version 1.1.0)","author":"Moulaeifard","year":"2026a"},{"key":"pmeaae92e8bib34","doi-asserted-by":"publisher","first-page":"668","DOI":"10.1038\/s41597-026-07335-8","type":"journal-article","article-title":"MIMIC-III-Ext-PPG, a PPG-based benchmark dataset for cardiovascular and respiratory signal analysis","volume":"13","author":"Moulaeifard","year":"2026c","journal-title":"Sci. Data"},{"key":"pmeaae92e8bib35","doi-asserted-by":"publisher","first-page":"13715","DOI":"10.1109\/ACCESS.2022.3148256","type":"journal-article","article-title":"Evaluation of objective distortion measures for automatic quality assessment of processed PPG signals for real-time health monitoring devices","volume":"10","author":"Narendra Kumar Reddy","year":"2022","journal-title":"IEEE Access"},{"key":"pmeaae92e8bib36","doi-asserted-by":"publisher","first-page":"832","DOI":"10.1109\/JBHI.2014.2338351","type":"journal-article","article-title":"Signal-quality indices for the electrocardiogram and photoplethysmogram: derivation and applications to wireless monitoring","volume":"19","author":"Orphanidou","year":"2015","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"pmeaae92e8bib37","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0262838","type":"journal-article","article-title":"Deep learning model calibration for improving performance in class-imbalanced medical image classification tasks","volume":"17","author":"Rajaraman","year":"2022","journal-title":"PLoS One"},{"key":"pmeaae92e8bib38","doi-asserted-by":"crossref","DOI":"10.1200\/CCI-24-00284","type":"other","article-title":"Assessing algorithmic fairness with a multimodal artificial intelligence model in men of African and non-African origin on NRG oncology prostate cancer phase III trials","author":"Roach","year":"2025"},{"key":"pmeaae92e8bib39","first-page":"pp 3581","type":"conference-proceedings","article-title":"Classification with valid and adaptive coverage","volume":"vol 33","author":"Romano","year":"2020"},{"key":"pmeaae92e8bib40","doi-asserted-by":"publisher","first-page":"223","DOI":"10.1080\/01621459.2017.1395341","type":"journal-article","article-title":"Least ambiguous set-valued classifiers with bounded error levels","volume":"114","author":"Sadinle","year":"2019","journal-title":"J. Am. Stat. Assoc."},{"key":"pmeaae92e8bib41","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2023.107816","type":"journal-article","article-title":"Understanding calibration of deep neural networks for medical image classification","volume":"242","author":"Sambyal","year":"2023","journal-title":"Comput. Methods Programs Biomed."},{"key":"pmeaae92e8bib42","first-page":"371","type":"journal-article","article-title":"A tutorial on conformal prediction","volume":"9","author":"Shafer","year":"2008","journal-title":"J. Mach. Learn. Res."},{"key":"pmeaae92e8bib43","type":"conference-proceedings","article-title":"Conformal prediction for class-wise coverage via augmented label rank calibration","volume":"vol 37","author":"Shi","year":"2024"},{"key":"pmeaae92e8bib44","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2026.133182","type":"journal-article","article-title":"Conformal prediction-based multi-sensor classification","volume":"678","author":"Siahkali","year":"2026","journal-title":"Neurocomputing"},{"key":"pmeaae92e8bib45","doi-asserted-by":"publisher","DOI":"10.1016\/j.spl.2022.109395","type":"journal-article","article-title":"Multi split conformal prediction","volume":"184","author":"Solari","year":"2022","journal-title":"Stat. Prob. Lett."},{"key":"pmeaae92e8bib46","doi-asserted-by":"publisher","first-page":"88","DOI":"10.1016\/j.bspc.2018.05.020","type":"journal-article","article-title":"PQR signal quality indexes: a method for real-time photoplethysmogram signal quality estimation based on noise interferences","volume":"47","author":"Song","year":"2019","journal-title":"Biomed. Signal Process. Control"},{"key":"pmeaae92e8bib47","type":"conference-proceedings","article-title":"Conformal prediction under covariate shift","volume":"vol 32","author":"Tibshirani","year":"2019"},{"key":"pmeaae92e8bib48","doi-asserted-by":"publisher","first-page":"116","DOI":"10.1038\/s41746-020-00320-4","type":"journal-article","article-title":"Multi-task deep learning for cardiac rhythm detection in wearable devices","volume":"3","author":"Torres-Soto","year":"2020","journal-title":"npj Dig. Med."},{"key":"pmeaae92e8bib49","doi-asserted-by":"publisher","first-page":"349","DOI":"10.1007\/s10994-013-5355-6","type":"journal-article","article-title":"Conditional validity of inductive conformal predictors","volume":"92","author":"Vovk","year":"2013","journal-title":"Mach. Learn."},{"key":"pmeaae92e8bib50","author":"Vovk","year":"2005","type":"book"},{"key":"pmeaae92e8bib51","doi-asserted-by":"publisher","first-page":"1878","DOI":"10.1177\/11297298241304467","type":"journal-article","article-title":"A machine learning approach for identification of vascular access patency in hemodialysis patients using photoplethysmography: a pilot study","volume":"26","author":"Yang","year":"2025","journal-title":"J. Vasc. Access."},{"key":"pmeaae92e8bib52","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2026.110393","type":"journal-article","article-title":"A unified, rhythm-aware diffusion model for denoising PPG signals in atrial fibrillation and sinus rhythm","volume":"122","author":"Zhang","year":"2026","journal-title":"Biomed. Signal Process. Control"}],"container-title":["Physiological Measurement"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6579\/ae92e8","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6579\/ae92e8\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6579\/ae92e8","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6579\/ae92e8\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6579\/ae92e8\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6579\/ae92e8\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6579\/ae92e8\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6579\/ae92e8\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T09:14:33Z","timestamp":1787562873000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6579\/ae92e8"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,24]]},"references-count":52,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2026,8,24]]},"published-print":{"date-parts":[[2026,8,31]]}},"URL":"https:\/\/doi.org\/10.1088\/1361-6579\/ae92e8","relation":{},"ISSN":["0967-3334","1361-6579"],"issn-type":[{"value":"0967-3334","type":"print"},{"value":"1361-6579","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,24]]},"assertion":[{"value":"Conformal reliability assessment of PPG-based ICU arrhythmia classification under class imbalance and patient heterogeneity","name":"article_title","label":"Article Title"},{"value":"Physiological Measurement","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2026 The Author(s). Published on behalf of Institute of Physics and Engineering in Medicine by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2026-06-15","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-07-30","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-08-24","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}