Cuffless blood pressure · uci-bp · Mean absolute error (mmHg); cuffless BP
Protocol not structured
Leaderboard row
2026
Results explorer
One row per reported result from wearable PPG, ECG, motion and sleep papers. Filter by task, dataset, metric, sensor and year, or show only SOTA rows. Each value links to its source: a paper table or figure, or the leaderboard row it came from.
Last updated
433
reported results
224
with a table or figure reference
92
papers and models
134
task × dataset × metric pairs
38
benchmarks with a SOTA star
433 results
Cuffless blood pressure · uci-bp · Mean absolute error (mmHg); cuffless BP
uses extra training data
Leaderboard row
2026
11.94↓
Cuffless blood pressure · uci-bp · Mean absolute error (mmHg); cuffless BP
uses extra training data
Leaderboard row
2026
4.29↓
Cuffless blood pressure · uci-bp · Standard deviation of error (mmHg); cuffless BP
Protocol not structured
Leaderboard row
2026
TCN–BiLSTM PPG→ABP
In Arterial blood pressure waveform reconstruction estimation from PPG using TCN–BiLSTM
3.61↓
Cuffless blood pressure · VitalDB (2,226 patients) · MAE (mmHg)
not stated in the abstractwithin-datasetPPG→ABP waveform model, end-to-endSBP
TCN–BiLSTM PPG→ABP
Arterial blood pressure waveform reconstruction estimation from PPG using TCN–BiLSTM
3.61↓
Cuffless blood pressure · VitalDB · Mean absolute error (mmHg); cuffless BP
Protocol not structured
Leaderboard row
2026
Cuffless blood pressure · VitalDB · Mean absolute error (mmHg); cuffless BP
Protocol not structured
Leaderboard row
2026
Individual-level Public-Private Decoupling for C
Individual-level Public-Private Decoupling for Continuous Blood Pressure Waveform Prediction (IPPD)
11.65↓
Cuffless blood pressure · VitalDB · RMSE (mmHg); ABP waveform or BP
Protocol not structured
Leaderboard row
2026
Cuffless blood pressure · windkessel-own-10 · Mean absolute error (mmHg); cuffless BP
Protocol not structured
Leaderboard row
2026
Apple PPG encoder (participant-level SSL)
In Large-scale Training of Foundation Models for Wearable Biosignals
0.993↑
Demographics from PPG · Apple Heart and Movement Study · AUROC
participant-level downstream splitswithin-datasetlinear probe (ridge) on participant-mean embeddingsbiological sex
Same encoder, segment-level positive pairs
Baseline in Large-scale Training of Foundation Models for Wearable Biosignals
0.870↑
Demographics from PPG · Apple Heart and Movement Study · AUROC
participant-level downstream splitswithin-datasetlinear probe (ridge) on participant-mean embeddingsbiological sex
Apple PPG encoder (participant-level SSL)
In Large-scale Training of Foundation Models for Wearable Biosignals
3.19↓
Demographics from PPG · Apple Heart and Movement Study · MAE (years)
participant-level downstream splitswithin-datasetlinear probe (ridge) on participant-mean embeddingsage regression
ECG encoder (participant-level)
Baseline in Large-scale Training of Foundation Models for Wearable Biosignals
6.33↓
Demographics from PPG · Apple Heart and Movement Study · MAE (years)
participant-level downstream splitswithin-datasetlinear probe (ridge) on participant-mean embeddingsage regression
Same encoder, segment-level positive pairs
Baseline in Large-scale Training of Foundation Models for Wearable Biosignals
6.60↓
Demographics from PPG · Apple Heart and Movement Study · MAE (years)
participant-level downstream splitswithin-datasetlinear probe (ridge) on participant-mean embeddingsage regression
ECG diagnosis · CODE-test · AUROC
external test setcross-datasetpretrained classifier, no fine-tuningaverage over 6 rhythm/conduction labels, 12-lead
0.963↑
ECG diagnosis · CODE-test · AUROC
external test setcross-datasetpretrained classifier, no fine-tuningaverage over 6 rhythm/conduction labels, 12-lead
ECG-SE-ResNet
Baseline in ECGFounder: An ECG Foundation Model Built on 10+ Million Recordings
0.963↑
ECG diagnosis · CODE-test · AUROC
external test setcross-datasetpretrained classifier, no fine-tuningaverage over 6 rhythm/conduction labels, 12-lead
NormWear
In NormWear: Toward Foundation Model for Multivariate Wearable Sensing of Physiological Signals
99.140%↑
ECG diagnosis · ECG-Abnormal · AUROC (%)
held-out test setwithin-datasetlinear probe, full-shot
Chronos
Baseline in NormWear: Toward Foundation Model for Multivariate Wearable Sensing of Physiological Signals
98.585%↑
ECG diagnosis · ECG-Abnormal · AUROC (%)
held-out test setwithin-datasetlinear probe, full-shot
TF-C
Baseline in NormWear: Toward Foundation Model for Multivariate Wearable Sensing of Physiological Signals
98.275%↑
ECG diagnosis · ECG-Abnormal · AUROC (%)
held-out test setwithin-datasetlinear probe, full-shot
CLAP
Baseline in NormWear: Toward Foundation Model for Multivariate Wearable Sensing of Physiological Signals
97.23%↑
ECG diagnosis · ECG-Abnormal · AUROC (%)
held-out test setwithin-datasetlinear probe, full-shot
Statistical
Baseline in NormWear: Toward Foundation Model for Multivariate Wearable Sensing of Physiological Signals
97.092%↑
ECG diagnosis · ECG-Abnormal · AUROC (%)
held-out test setwithin-datasetlinear probe, full-shot
Modality-specific baseline
Baseline in NormWear: Toward Foundation Model for Multivariate Wearable Sensing of Physiological Signals
89.898%↑
ECG diagnosis · ECG-Abnormal · AUROC (%)
held-out test setwithin-datasetlinear probe, full-shot
ECG diagnosis · PTB-XL (5 superclasses) · AUROC
official PTB-XL stratified split (folds 1–8 train, 9 validation, 10 test)within-datasettransformer pretrained on MIMIC-IV-ECG, evaluated on PTB-XL5 diagnostic superclasses, macro AUROC
0.86↑
ECG diagnosis · PTB-XL · F1 score
not stated in the abstractwithin-datasetjoint ECG–PPG foundation modelcardiovascular disease classification
μDPad (PPG + IMU)
In μDPad: A Large-Scale Multimodal PPG and IMU Dataset for Wrist-Worn Microgesture Recognition
83.7↑
Gesture recognition · μDPad microgestures · Accuracy (%)
not stated in the abstractwithin-dataset1D CNN on accelerometer, gyroscope and spatial PPG
XGBoost (Karunarathna)
94.2↑
Glucose estimation · BIG IDEAs Lab Glycemic Variability and Wearable Device Data · Clarke error grid zone A (%)
subjects overlap
Random forest (Karunarathna)
89.1↑
Glucose estimation · BIG IDEAs Lab Glycemic Variability and Wearable Device Data · Clarke error grid zone A (%)
subjects overlap
XGBoost (Karunarathna)
11.9↓
Glucose estimation · BIG IDEAs Lab Glycemic Variability and Wearable Device Data · Glucose RMSE (mg/dL)
subjects overlap
Random forest (Karunarathna)
15.6↓
Glucose estimation · BIG IDEAs Lab Glycemic Variability and Wearable Device Data · Glucose RMSE (mg/dL)
subjects overlap
Bent 2021 personalized
21.1↓
Glucose estimation · BIG IDEAs Lab Glycemic Variability and Wearable Device Data · Glucose RMSE (mg/dL)
subjects overlap
Bent 2021 population
21.22↓
Glucose estimation · BIG IDEAs Lab Glycemic Variability and Wearable Device Data · Glucose RMSE (mg/dL)
subject-disjoint
XGBoost (Karunarathna)
7.1↓
Glucose estimation · BIG IDEAs Lab Glycemic Variability and Wearable Device Data · MARD (%)
subjects overlap
Random forest (Karunarathna)
9.7↓
Glucose estimation · BIG IDEAs Lab Glycemic Variability and Wearable Device Data · MARD (%)
subjects overlap
Bent 2021 personalized
13.26↓
Glucose estimation · BIG IDEAs Lab Glycemic Variability and Wearable Device Data · MARD (%)
subjects overlap
Bent 2021 population
14.33↓
Glucose estimation · BIG IDEAs Lab Glycemic Variability and Wearable Device Data · MARD (%)
subject-disjoint
PLS (Islam)
86.57↑
Glucose estimation · PPG Signals for Blood Glucose Measurement (Kermani / Mendeley) · Clarke error grid zone A (%)
subject-disjointuses extra training data
MFCC + XGBoost (Prabha)
83.58↑
Glucose estimation · PPG Signals for Blood Glucose Measurement (Kermani / Mendeley) · Clarke error grid zone A (%)
subject-disjointuses extra training data
Mean baseline
77.61↑
Glucose estimation · PPG Signals for Blood Glucose Measurement (Kermani / Mendeley) · Clarke error grid zone A (%)
subject-disjointuses extra training data
ResNet34-1s
76.6↑
Glucose estimation · PPG Signals for Blood Glucose Measurement (Kermani / Mendeley) · Clarke error grid zone A (%)
subject-disjointuses extra training data
XGBoost (Gupta)
75.6↑
Glucose estimation · PPG Signals for Blood Glucose Measurement (Kermani / Mendeley) · Clarke error grid zone A (%)
subject-disjointuses extra training data
ResNet34 (Zeynali)
72.17↑
Glucose estimation · PPG Signals for Blood Glucose Measurement (Kermani / Mendeley) · Clarke error grid zone A (%)
subject-disjointuses extra training data
VGG16-1s
71.3↑
Glucose estimation · PPG Signals for Blood Glucose Measurement (Kermani / Mendeley) · Clarke error grid zone A (%)
subject-disjointuses extra training data
CNN-LSTM-Att-1s
69↑
Glucose estimation · PPG Signals for Blood Glucose Measurement (Kermani / Mendeley) · Clarke error grid zone A (%)
subject-disjointuses extra training data
1D-CNN (Alghlayini)
62.69↑
Glucose estimation · PPG Signals for Blood Glucose Measurement (Kermani / Mendeley) · Clarke error grid zone A (%)
subject-disjointuses extra training data
PLS (Islam)
13.37↓
Glucose estimation · PPG Signals for Blood Glucose Measurement (Kermani / Mendeley) · Glucose MAE (mg/dL)
subject-disjointuses extra training data
MFCC + XGBoost (Prabha)
14.23↓
Glucose estimation · PPG Signals for Blood Glucose Measurement (Kermani / Mendeley) · Glucose MAE (mg/dL)
subject-disjointuses extra training data
ResNet34-1s
14.8↓
Glucose estimation · PPG Signals for Blood Glucose Measurement (Kermani / Mendeley) · Glucose MAE (mg/dL)
subject-disjointuses extra training data
Mean baseline
16↓
Glucose estimation · PPG Signals for Blood Glucose Measurement (Kermani / Mendeley) · Glucose MAE (mg/dL)
subject-disjointuses extra training data
ResNet34 (Zeynali)
16.33↓
Glucose estimation · PPG Signals for Blood Glucose Measurement (Kermani / Mendeley) · Glucose MAE (mg/dL)
subject-disjointuses extra training data
Showing 101–150 of 433 results
Recent changes
New entry: A cross-subject stress recognition method…, 0.9158
Stress detection (acute / prolonged / affect): WESAD: accuracy
New entry: PC-AVCT wearable-state model, 7.697
From the hub's dated sync history: new results as they are added, and SOTA stars as they are awarded or move.
How SOTA is decided
- marks the best value for one task, dataset and metric, using the metric's direction (higher or lower is better).
- Results are compared only under the same evaluation protocol: split type, within- or cross-dataset testing, filtering, inputs and label set. Different protocols stay separate and are never ranked together.
- Only results with a table, figure or section reference compete, and a star needs at least two competing methods. Ties go to the earlier publication.
- Stars are recomputed whenever the research hub syncs, and every change is logged.
Downloads of every result (JSON and CSV) are free for members. Sign in to download.
Results as reported by the authors; not independently reproduced. Leaderboard rows without a structured protocol are listed but don't receive stars. See also the leaderboards.
Cite or reuse this
CC BY 4.0ChatPPG compiled this data. You can copy, adapt and republish it, including in articles, slides, papers and AI tools, as long as you credit ChatPPG and link to this page. Please also cite the original studies for any number you use.
Attribution
Source: ChatPPG, Wearable AI results explorer, https://chatppg.com/research/results (CC BY 4.0)
HTML link
Source: <a href="https://chatppg.com/research/results">ChatPPG, Wearable AI results explorer</a> (<a href="https://creativecommons.org/licenses/by/4.0/" rel="license">CC BY 4.0</a>)