The study, published in the Annals of Biomedical Engineering, is the first to demonstrate that daytime sleepiness can be objectively assessed using only photoplethysmography, or PPG—the same optical technique found in ordinary pulse oximeters and consumer smartwatches. A PPG sensor shines light into the skin and measures how blood volume pulses with each heartbeat, producing a waveform that carries subtle fingerprints of the autonomic nervous system. As a person drifts toward sleep, heart rate slows, parasympathetic tone rises, and the shape and timing of the pulse wave change in characteristic ways. The researchers hypothesized that a neural network trained to recognize these signatures could detect sleep onset during daytime tests just as reliably as the electroencephalogram-based methods used today.

To build the model, the team turned to the Multimorbidity Apnea Respiratory Failure Sleep (MARS) database maintained at the University Hospital Grenoble Alpes in France. They harvested 2,423 diagnostic overnight polysomnography recordings collected between 2013 and 2023, each of which included a medical-grade fingertip pulse oximeter channel alongside the full complement of brain, eye, and muscle sensors. Sleep experts had manually scored every recording according to American Academy of Sleep Medicine criteria, providing the ground truth labels. After harmonizing the data and filtering the PPG signals—low-pass filtering at 32 hertz, downsampling from 128 to 64 hertz, and applying z-score normalization—2,297 valid overnight recordings were split into training, validation, and test sets for a U-time-based deep learning architecture, a convolutional network well suited to segmenting long biomedical time series.

The crucial test came when this overnight-trained model was applied, without any retraining or adaptation, to daytime recordings: 143 patients who had undergone the MSLT and 127 who had taken the MWT at the Grenoble sleep center. In the MSLT, patients are given four 20-minute opportunities to fall asleep at 9 am, 11 am, 1 pm, and 3 pm; in the MWT, they are instead asked to resist sleep across four 40-minute sessions. The model processed the fingertip pulse wave second by second, producing a continuous probability of sleep, from which an automatic mean sleep latency could be derived and compared against the latencies scored by experienced human technicians using conventional EEG-based rules.

The results were strikingly encouraging for the MSLT. Second-by-second classification of sleep versus wakefulness from the pulse wave alone achieved 81 percent accuracy with a Cohen’s kappa of 0.60, indicating moderate-to-substantial agreement with manual scoring, and the detection of sleep was well balanced between precision (0.71) and recall (0.80). When subjects were classified as objectively sleepy or non-sleepy using the standard clinical threshold of a mean sleep latency below 8 minutes, the fingertip-based method reached 80 percent accuracy, with 65 percent sensitivity and 84 percent specificity. The automatically derived mean sleep latency correlated significantly with the manually scored value (r = 0.61, p < 0.001), and the typical difference between the two methods fell within a few minutes.

Perhaps most compelling were the group-level sleep probability curves. When the researchers averaged the model’s second-by-second sleep probability across all tests for each subject, the sleepy and non-sleepy groups produced visibly distinct trajectories. In the MSLT, the sleepy group—31 subjects across 123 tests—showed a consistently higher and more steeply rising sleep probability than the 112 non-sleepy subjects across 441 tests, with the separation evident from the very beginning of the recording and persisting across most of the 20-minute session. A global permutation test confirmed the difference in profile shape was statistically significant (p = 0.0002). In the MWT, the sleepy group likewise displayed elevated sleep probability, particularly during the middle portion of the 40-minute test, while the non-sleepy group remained low and stable (p = 0.0012).

The MWT told a more cautionary tale. Because participants in the MWT are actively trying to stay awake, a staggering 98 percent of the second-by-second data consists of wakefulness, creating an extreme class imbalance that renders raw accuracy figures misleading. Although overall accuracy reached 88 percent, precision for detecting sleep collapsed to just 0.10, meaning that when the model labeled a moment as sleep, it was usually wrong. Cohen’s kappa fell to a weak 0.15, and the misclassification of objectively sleepy subjects rose to 38 percent. The authors attribute this to a fundamental physiological ambiguity: during quiet wakefulness in the MWT, transient parasympathetic relaxation or reduced movement can mimic the cardiovascular signature of sleep onset in the fingertip pulse wave, fooling a model trained exclusively on overnight physiology.

This limitation points to the study’s most important methodological caveat: the model was trained on nocturnal polysomnography and deployed zero-shot onto daytime recordings, without any domain adaptation. Autonomic physiology during nighttime sleep differs meaningfully from that during daytime naps, and even more so from the tense, motionless wakefulness of a patient instructed to resist sleep. The researchers acknowledge that transfer learning or fine-tuning on daytime-specific data will likely be essential before the approach can be trusted in wake-maintenance testing, and they note that some large outliers—differences of 20 to 40 minutes in estimated sleep latency—appeared in the MWT dataset where manual scoring suggested very long latencies that the pulse-wave model dramatically underestimated.

Even so, the broader implications are considerable. The MSLT is embedded in the diagnostic criteria for narcolepsy types 1 and 2 and idiopathic hypersomnia, and both the MSLT and MWT are used to evaluate residual sleepiness in CPAP-treated obstructive sleep apnea patients, to assess fitness for safety-critical professions such as commercial driving, and to measure the effects of wake-promoting medications. Yet excessive sleepiness persists in 9 to 22 percent of sleep apnea patients even when CPAP therapy is well tolerated, and objective testing is rarely performed outside specialized centers because of its cost and labor intensity. Notably, the fingertip-based sleep latency correlated with subjective sleepiness on the Epworth Sleepiness Scale to a degree statistically indistinguishable from the manually scored latency, suggesting the simple signal captures clinically meaningful information about the patient’s experienced state.

The authors are careful to frame this as a first step rather than a finished clinical tool. The analysis was retrospective, the model has not been shown to track sleep onset in real time, and a pulse oximeter-only test conducted without concurrent brain monitoring may not be feasible under current MWT protocols, in which patients must be prevented from sleeping between sessions. Future work, they write, should optimize the MSLT and MWT protocols for pulse-wave analysis, develop daytime-adapted training strategies, and validate the approach prospectively across diverse patient groups. Still, the vision is clear: a cheap, passive, wrist-worn or fingertip-based sensor paired with a neural network could one day bring objective sleepiness assessment out of the sleep laboratory and into routine clinical practice—transforming how narcolepsy is diagnosed, how tired drivers are evaluated, and how millions of chronically sleepy patients are monitored over the course of their treatment.

Subject of Research: Deep learning assessment of daytime sleepiness from fingertip photoplethysmography during MSLT and MWT testing

Article Title: Deep Learning-Based Assessment of Sleepiness from Fingertip Pulse Wave Analysis During Objective Daytime Testing

Article References: Rusanen, M., Kainulainen, S., Myllymaa, S., Leppänen, T., Tamisier, R., Baillieul, S., Bailly, S., & Pepin, J.-L. (2026). Deep Learning-Based Assessment of Sleepiness from Fingertip Pulse Wave Analysis During Objective Daytime Testing. Annals of Biomedical Engineering. https://doi.org/10.1007/s10439-026-04393-2

Image Credits: AI Generated

DOI: 10.1007/s10439-026-04393-2

Keywords: deep learning, photoplethysmography, daytime sleepiness, Multiple Sleep Latency Test, Maintenance of Wakefulness Test, sleep medicine, sleep apnea, narcolepsy, pulse wave analysis, sleep staging, biomedical engineering, Epworth Sleepiness Scale

The research, led by Mathieu Kruska and Volker Liebe of the University Medical Centre Mannheim at Heidelberg University together with colleagues across several German institutions, enrolled eighty-three patients with symptomatic atrial fibrillation who were scheduled for catheter ablation at a single center between October 2020 and March 2022. The cohort had a median age of seventy-two years, and just over one-third of participants were women. Most patients, eighty-two percent, suffered from the paroxysmal form of the arrhythmia, in which episodes come and go rather than persist continuously. Their symptom burden was substantial: the median European Heart Rhythm Association symptom score was three, indicating moderate to severe symptoms, and their average score on a validated quality of life questionnaire was only sixty out of a possible one hundred, underscoring how heavily the condition weighed on daily living.

What sets this study apart is its multimodal approach. Before ablation, each patient underwent pulse wave analysis using an oscillometric device called VascAssist2.0, which measures blood pressure waveforms at the arm and uses a mathematical model of the arterial system to derive a suite of vascular parameters. These include brachial and central blood pressures, pulse wave velocity, augmentation pressure, augmentation index, left ventricular ejection time, and model-based indices of arterial stiffness and vascular resistance. In parallel, patients received transthoracic echocardiography to assess cardiac structure and function, a twelve-lead electrocardiogram, laboratory testing including the heart failure biomarker NT-proBNP, and a detailed quality of life assessment using the AFEQT questionnaire, a disease-specific instrument covering twenty items that captures how atrial fibrillation affects symptoms, daily activities, and treatment satisfaction.

The correlations that emerged were striking. Lower AFEQT scores, indicating worse quality of life, correlated strongly with higher vascular resistance, with a correlation coefficient of minus 0.64, and with increased arterial stiffness, at minus 0.62, both highly statistically significant. Elevated central systolic blood pressure, the pressure actually experienced by the heart and brain rather than the arm, also tracked with poorer quality of life at minus 0.38. On the cardiac side, the strongest association of all was found with reduced left ventricular ejection fraction below fifty percent, which correlated at minus 0.74 with AFEQT scores. Diastolic dysfunction, the inability of the heart’s main pumping chamber to relax and fill properly, correlated at minus 0.34, while a clinical diagnosis of heart failure correlated at minus 0.39 and logarithmically transformed NT-proBNP levels at minus 0.38. Patient-reported quality of life also aligned closely with physician-assessed symptom classification, with EHRA scores correlating at minus 0.91 with AFEQT scores, a reassuring sign that the two instruments are measuring the same underlying phenomenon from different angles.

To understand why stiff arteries should make an abnormal heart rhythm feel worse, the authors turn to the concept of ventricular-arterial and arterial-atrial coupling. When the large arteries lose their elastic cushioning, every heartbeat travels through the vascular tree faster, and reflected pressure waves return to the heart earlier in the cardiac cycle. This raises the central systolic pressure the left ventricle must pump against, increasing afterload and impairing diastolic relaxation. Higher pressures then back up into the left atrium, promoting structural and functional remodeling of that chamber, a process central to atrial cardiomyopathy. The resulting atrial substrate not only facilitates the persistence of atrial fibrillation but may also blunt the atrium’s reservoir function, intensifying symptoms such as fatigue and breathlessness. The same hemodynamic cascade is considered a central driver of heart failure with preserved ejection fraction, tying together several threads of cardiovascular medicine in a single mechanistic framework.

Intriguingly, one conventional measure of arterial health did not follow this pattern. Aortic pulse wave velocity, widely regarded as the reference standard for large-artery stiffness, was within age-adapted reference values in the cohort at a median of 8.6 meters per second and did not correlate significantly with quality of life scores. The authors suggest that pulse wave velocity predominantly reflects the structural properties of the aorta and vascular aging, whereas the model-derived vascular resistance, arterial stiffness index, and central systolic blood pressure may better capture dynamic functional afterload and ventricular-arterial coupling. Those fluctuating hemodynamic loads, they argue, may be more directly connected to the day-to-day symptoms of palpitations, dyspnea, and exercise intolerance than a static measure of aortic structure. Notably, the pulse wave measurements proved robust regardless of rhythm: over ninety percent of patients were in sinus rhythm at the time of testing, and no significant differences were found between measurements taken during sinus rhythm and those taken during atrial fibrillation.

The study also highlights the tangled relationship between atrial fibrillation and heart failure, two conditions that each fuel the other’s progression. Thirty percent of the cohort had heart failure, forty-two percent showed echocardiographic evidence of diastolic dysfunction, and elevated NT-proBNP levels were strongly associated with poorer quality of life. Disentangling which symptoms stem from the arrhythmia and which from the failing heart is notoriously difficult, since dyspnea and fatigue dominate both. Interestingly, heart failure with reduced ejection fraction was associated with impaired quality of life in this analysis, whereas the preserved-ejection-fraction phenotype did not reach statistical significance. That observation echoes earlier findings suggesting that the symptomatic benefits of ablation may be attenuated in patients with heart failure with preserved ejection fraction, possibly because their symptoms are driven more by the stiff, non-compliant cardiovascular system than by the arrhythmia itself.

Beyond the vascular and cardiac measurements, broader comorbidity burden left its mark. Coronary artery disease, older age, arterial hypertension, higher total and LDL cholesterol, and reduced kidney function all correlated inversely with quality of life scores, as did higher CHA2DS2-VASc and HAS-BLED risk scores. Taken together, these associations paint quality of life in atrial fibrillation as a barometer of overall cardiovascular health rather than a simple readout of arrhythmia burden. This aligns with large registry data linking cardiovascular comorbidities to worse patient-reported outcomes, and it reinforces current European Society of Cardiology guidelines that emphasize comprehensive management of risk factors alongside rhythm control strategies.

The authors are careful to frame their findings appropriately. The study was exploratory and hypothesis-generating, conducted at a single center with a modest sample size and no formal a priori power calculation. Because many univariate correlations were performed without adjustment for multiple testing, the reported associations should be interpreted descriptively, and the absence of multivariable modeling means the independent contribution of each vascular parameter cannot be isolated. The single-time-point, observational design precludes any causal inference, and recruitment during the COVID-19 pandemic added logistical strain to elective procedural volumes. Whether pulse wave analysis-derived vascular phenotyping genuinely adds predictive value beyond established clinical evaluation will require prospective validation in larger cohorts.

Even with those caveats, the implications are compelling. If stiff arteries, elevated central pressures, and weakened or stiffened heart muscle account for a substantial share of the suffering attributed to atrial fibrillation, then measuring them before ablation could help clinicians identify patients whose symptoms reflect more than the arrhythmia alone, and tailor treatment accordingly, with intensified blood pressure control, vascular risk management, and heart failure therapy running alongside rhythm control. For patients, the message is equally resonant: the health of the arteries is inseparable from the experience of the arrhythmia. As the authors conclude, reduced quality of life in symptomatic atrial fibrillation reflects a complex interplay between vascular function, myocardial performance, and the rhythm disorder itself, and understanding that interplay may ultimately determine who truly benefits from a procedure that millions pin their hopes on.

Subject of Research: Associations between vascular and cardiac functional parameters and quality of life in atrial fibrillation patients scheduled for catheter ablation

Article Title: Impact of vascular and cardiac parameters on quality of life in patients undergoing catheter ablation for atrial fibrillation

Article References: Kruska, M., Liebe, V., Fastner, C., Kranert, M., Jehle, M., Derda, A., Schumacher, G., Akin, I., Duerschmied, D., & Hohneck, A. (2026). Impact of vascular and cardiac parameters on quality of life in patients undergoing catheter ablation for atrial fibrillation. Clinical Research in Cardiology. https://doi.org/10.1007/s00392-026-03005-2

Image Credits: AI Generated

DOI: 10.1007/s00392-026-03005-2

Keywords: atrial fibrillation, catheter ablation, quality of life, arterial stiffness, vascular resistance, pulse wave analysis, echocardiography, NT-proBNP, heart failure, central blood pressure, AFEQT, diastolic dysfunction