Unlocking the Specialization Capability of PPG Foundation Models for Wearable Signals through Structured Pruning and Adaptive Fine-Tuning

Huaming Du; Zhixian Xie; Gaoyang Zhen; Gang Kou; Guisong Liu; Yanchao Tan; Wendan Tao; Yuhao Xu

cs.AIOct 24, 2026

Unlocking the Specialization Capability of PPG Foundation Models for Wearable Signals through Structured Pruning and Adaptive Fine-Tuning

Authors: Huaming Du, Zhixian Xie, Gaoyang Zhen, Gang Kou, Guisong Liu, Yanchao Tan, Wendan Tao, Yuhao Xu,

Organizations: Southwestern University of Finance and Economics · Jizhi Technology Co., Ltd. · Fuzhou University · West China Hospital ·

Abstract

Photoplethysmography (PPG) plays a central role in wearable health monitoring and clinical decision support. However, existing approaches to universal PPG representation learning are predominantly trained on high-quality large-scale datasets obtained from oximeters in hospital settings and often struggle to adapt to sparse and low-quality real-world datasets, thereby limiting their generalization to clinically relevant tasks on prevalent PPG data such as collected from wearable devices. Surprisingly, even after full fine-tuning, PPG foundation models cannot consistently outperform general-purpose time-series models. Therefore, a key question is how to effectively adapt PPG foundation models to target prediction tasks. Through empirical studies on various PPG foundation models, we find that pretrained models often exhibit inherent sparsity and computational redundancy, suggesting that PPG foundation models have learned to activate task-relevant network substructures for different prediction tasks. To preserve this valuable prior knowledge, we propose a structured pruning-and-fine-tuning framework, SPAFT, which regularizes the adaptive fine-tuning process by constraining it to a more relevant and compact parameter space based on parameter importance and structural redundancy. Extensive experiments across four PPG foundation models, five benchmark datasets, and ten downstream tasks demonstrate that SPAFT achieves win rates of 82.8% and 87.5% on regression and classification tasks, respectively. This prune-then-finetune paradigm often enables PPG foundation models to achieve state-of-the-art performance, laying a promising foundation for the real-world clinical deployment of PPG foundation models and the development of personalized PPG foundation models.

Figures & tables

Figure 1: Empirical analysis. More results and analyses are provided in Appendix B .

Figure 2: Architecture of Our Proposed SPAFT Framework.

Data	Tasks	FM	Linear Probe		Full-FT		LoRA		QLoRA		PFEC		Taylor-FO		FPGM		SRR-GR		Less-is-More		Ours
Data	Tasks	FM	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$
UCI-BP	SBP	AnyPPG	379.531	15.388	221.626	10.804	327.334	13.870	392.230	15.488	460.250	16.821	191.174	9.871	223.532	10.885	338.435	14.133	379.268	15.086	143.083	8.345
		PaPaGei-P	430.231	16.542	371.494	14.974	379.325	15.183	471.292	17.481	482.716	17.735	371.448	14.924	365.637	14.757	482.458	17.775	434.871	16.591	245.788	11.438
		PaPaGei-S	461.856	17.295	383.028	15.299	422.433	16.301	438.100	16.600	492.882	17.987	393.232	15.628	380.783	15.240	445.977	16.893	387.265	15.478	259.401	11.858
		Pulse-PPG	417.647	16.263	256.021	11.825	356.951	14.652	354.624	14.660	335.480	14.175	257.260	11.845	254.844	11.788	312.325	13.553	277.007	12.581	183.581	9.661
	DBP	AnyPPG	87.673	6.881	59.591	5.156	78.304	6.224	92.546	7.243	82.621	6.564	53.857	4.816	78.587	6.081	76.814	6.256	80.273	6.475	41.276	4.111

Table 1: Model performance on regression tasks. We mark the winner in bold , and the runner-up is indicated by an underline . The complete results are provided in Table 17 of Appendix C.1 .

Data	Tasks	FM	Linear Probe		Full-FT		LoRA		QLoRA		PFEC		Taylor-FO		FPGM		SRR-GR		Less-is-More		Ours
Data	Tasks	FM	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$
BUT PPG	SQI	AnyPPG	0.622	0.799	0.599	0.724	0.645	0.793	0.579	0.626	0.678	0.811	0.632	0.734	0.622	0.755	0.697	0.800	0.695	0.799	0.707	0.806
		PaPaGei-P	0.535	0.736	0.595	0.682	0.475	0.685	0.558	0.579	0.585	0.628	0.599	0.712	0.544	0.690	0.581	0.598	0.599	0.633	0.669	0.766
		PaPaGei-S	0.544	0.714	0.606	0.708	0.491	0.697	0.629	0.686	0.614	0.645	0.612	0.701	0.594	0.695	0.600	0.633	0.616	0.675	0.686	0.801
		Pulse-PPG	0.514	0.718	0.562	0.724	0.515	0.717	0.517	0.575	0.629	0.717	0.572	0.738	0.540	0.764	0.630	0.749	0.566	0.734	0.638	0.766
DeepBeat	AF	AnyPPG	0.544	0.708	0.651	0.838	0.468	0.525	0.609	0.754	0.642	0.799	0.646	0.833	0.653	0.834	0.638	0.800	0.640	0.796	0.684	0.844

Table 2: Model performance on classification tasks and the complete results are provided in Table 17 of Appendix C.1 .

Figure 3: Performance of PaPaGei-S with and without personalized training.

Model	Variant	UCI-BP DBP		BUT PPG SQI
Model	Variant	MSE $\downarrow$	MAE $\downarrow$	F1 $\uparrow$	AUROC $\uparrow$
PAPAGEI-P	w/o IFD	65.089 $\pm$ 0.416	5.441 $\pm$ 0.054	0.595 $\pm$ 0.069	0.726 $\pm$ 0.009
	w/o SRRBS	65.746 $\pm$ 0.147	5.506 $\pm$ 0.034	0.572 $\pm$ 0.100	0.737 $\pm$ 0.007
	w/o SFF	87.165 $\pm$ 0.225	6.746 $\pm$ 0.016	0.512 $\pm$ 0.049	0.627 $\pm$ 0.078
	w/o ABFA	66.072 $\pm$ 3.360	5.483 $\pm$ 0.196	0.607 $\pm$ 0.017	0.664 $\pm$ 0.061
	SPAFT	63.760 $\pm$ 0.730	5.373 $\pm$ 0.052	0.640 $\pm$ 0.057	0.765 $\pm$ 0.021

Table 3: SPAFT component ablation study.

Appendix figures & tables24 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 5: Overview of Ring-PPG collection and temporal alignment during hospitalization. Patients wore a smart ring that recorded green, red, and infrared PPG signals every 10 minutes from enrollment until discharge, with ring removal documented during procedures. Clinical measurements included real-time heart rate, $\mathrm{SpO_{2}}$ , and respiratory rate, hourly blood pressure readings, and laboratory measurements from fasting blood samples collected on hospital day 2. PPG recordings and clinical measurements were aligned by acquisition time to construct the dataset.

Dataset	Task	Task Type	#Subj.	Train / Val / Test	Total Samples
BUT PPG	HR estimation	R	50	2486 / 622 / 732	3840
BUT PPG	Signal quality assessment	B	50	2486 / 622 / 732	3840
PPG-DaLiA	HR estimation	R	15	8368 / 1829 / 2746	12943
DeepBeat	Atrial fibrillation detection	B	N/A	415,025 / 103,757 / 17,617	536399
UCI-BP	SBP estimation	R	N/A	167399 / 41,850 / 52,312	261561
UCI-BP	DBP estimation	R	N/A	167,399 / 41,850 / 52,312	261561

Appendix

Table 4: Datasets for downstream evaluation. Task type: R (regression), B (binary).

Setting	Value or search domain
Cosine threshold $\tau$	$\{0.1,0.2,0.3,0.4,0.5\}$
Global pruning ratio $\rho$	$\{0.05,0.10,0.15,0.20,0.30\}$
Fine-tuning learning rate	$\{10^{-2},10^{-3},10^{-4},10^{-5}\}$
Weight $w_{1}$	$0.35$
Weight $w_{2}$	$0.65$
Effective / micro-batch size	1,024 / 256

Appendix

Table 5: SPAFT optimization settings and common search domains.

Dataset–task pairs	fine-tuning ratio domain $\eta$	BatchNorm
UCI-BP (SBP, DBP); PPG-DaLiA (HR); BUT PPG (HR)	$\{0.10\}$	Frozen
Ring-PPG (HR, SBP, SpO 2 )	$\{0.10\}$	Frozen
Ring-PPG (DBP)	$\{0.10,0.20,0.30\}$	Frozen
BUT PPG (SQI); DeepBeat (AF)	$\{0.10,0.20,0.30\}$	Updated

Appendix

Table 6: Task-specific fine-tuning ratio domains and BatchNorm policies for SPAFT.

Figure 6: Cumulative distribution of the average relative output norm of filters.

Figure 7: Cumulative distribution of the activation probability of a single filter.

Figure 8: Cumulative distribution of the average relative output norm of a single block.

Data	Tasks	FM	Linear Probe		Full-FT		LoRA		QLoRA
Data	Tasks	FM	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$
UCI-BP	SBP	MOMENT	365.007	14.971	301.557	13.470	255.651	12.765	247.612	12.534
		Chronos-T5-base	421.585	16.597	339.169	14.015	301.742	13.564	294.418	13.296
		GPT-PPG	453.942	17.175	395.618	15.525	362.828	14.711	387.024	15.410
		SIGMA-PPG	442.144	16.835	342.883	14.247	365.191	14.743	353.973	14.367
		Ours	MSE: $\mathbf{143.514}$ MAE: $\mathbf{8.388}$

Appendix

Table 7: Comparison of our SOTA model with time-series foundation models and Transformer-based PPG foundation models on regression tasks.

Data	Tasks	FM	Linear Probe		Full-FT		LoRA		QLoRA
Data	Tasks	FM	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$
BUT PPG	SQI	MOMENT	0.634	0.813	0.650	0.815	$\underline{0.675}$	$\underline{0.825}$	0.669	$\mathbf{0.826}$
		Chronos-T5-base	0.580	0.765	0.573	0.760	0.564	0.757	0.621	0.764
		GPT-PPG	0.523	0.660	0.505	0.657	0.464	0.657	0.498	0.711
		SIGMA-PPG	0.585	0.737	0.631	0.761	0.604	0.742	0.671	0.758
		Ours	F1: $\mathbf{0.716}$ AUROC: $0.813$

Appendix

Table 8: Comparison of our SOTA model with time-series foundation models and Transformer-based PPG FMs on classification tasks.

Data	Tasks	FM	Linear Probe		Full-FT		QLoRA		SRR-GR		Less-is-More		PPG-DISTILL		Ours
Data	Tasks	FM	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$
UCI-BP	SBP	PaPaGei-S	461.959	17.296	396.019	15.678	479.878	17.635	449.427	16.861	442.235	17.005	426.989	16.473	296.102	12.935
UCI-BP	DBP	PaPaGei-S	104.351	7.697	83.100	6.504	111.985	7.770	102.661	7.599	105.346	7.766	92.758	7.054	65.138	5.414
PPG-DaLiA	HR	PaPaGei-S	244.569	12.075	274.732	12.418	296.118	13.759	220.111	11.389	182.748	10.174	215.872	10.979	133.177	8.304
Ring-PPG	HR	PaPaGei-S	82.417	7.474	90.252	7.721	71.257	7.156	91.127	8.159	66.563	6.491	89.961	7.220	64.737	6.631
	SBP	PaPaGei-S	213.565	12.105	258.791	13.101	201.459	12.211	240.100	12.786	191.251	12.000	252.562	13.344	176.200	11.440

Appendix

Table 9: Comparison with PPG-Distill on regression tasks.

Data	Tasks	FM	Linear Probe		Full-FT		QLoRA		SRR-GR		Less-is-More		PPG-DISTILL		Ours
Data	Tasks	FM	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$	F1 $\uparrow$	AUROC $\uparrow$
BUT PPG	SQI	PaPaGei-S	0.584	0.707	0.639	0.727	0.646	0.732	0.570	0.671	0.590	0.706	0.580	0.686	0.666	0.791
Deepbeat	AF	PaPaGei-S	0.444	0.610	0.616	0.786	0.584	0.660	0.636	0.749	0.597	0.723	0.493	0.754	0.659	0.815

Appendix

Table 10: Comparison with PPG-Distill on classification tasks.

Model	Method	Test MSE $\downarrow$	Test MAE $\downarrow$
PaPaGei-P	Linear Probe	184.968	10.172
	Full-FT	160.699	8.988
	LoRA	119.482	7.146
	Less-is-More	102.964	6.637
	Ours	88.019	6.226
PaPaGei-S	Linear Probe	240.210	11.845

Appendix

Table 11: Performance comparison.

Figure 10: subject-level performance comparison in terms of MSE.

Figure 11: subject-level performance comparison in terms of MAE.

Panel of Figure 12 — Figure 12: Comparison of test MSE with and without personalized training across different models.

Model	Method	w/o Personalized Training		w/ Personalized Training		Relative Improvement
Model	Method	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$	MSE Gain	MAE Gain
PaPaGei-P	LoRA	109.455	6.546	75.293	5.562	31.21%	15.03%
	Less-is-More	93.445	6.167	65.593	4.922	29.81%	20.19%
	Ours	79.733	5.750	48.270	4.607	39.46%	19.88%
PaPaGei-S	LoRA	110.504	7.594	73.173	5.742	33.78%	24.39%
	Less-is-More	146.365	8.604	127.635	8.204	12.80%	4.65%

Appendix

Table 12: Impact of personalized training on different PPG FMs.

Source $\rightarrow$ Target	PPG FM	Linear Probe		SPAFT Transfer
Source $\rightarrow$ Target	PPG FM	MSE $\downarrow$	MAE $\downarrow$	MSE $\downarrow$	MAE $\downarrow$
DBP $\rightarrow$ SBP	AnyPPG	379.386	15.384	306.571	13.562
	PaPaGei-P	430.122	16.539	348.212	14.595
	PaPaGei-S	461.959	17.296	377.095	15.327
	Pulse-PPG	417.640	16.264	333.899	14.267
SBP $\rightarrow$ DBP	AnyPPG	87.652	6.880	71.777	6.068

Appendix

Table 13: Transferability Study on UCI-BP.

Dataset	PPG FMs	Original			SPAFT
Dataset	PPG FMs	TMU (GB)	FLOPs (G)	Params (M)	TMU (GB)	FLOPs (G)	Params (M)
Ring-PPG HR	AnyPPG	0.0479	0.2173	4.0448	0.0442	0.1781	3.4372
	PaPaGei-S	0.0565	0.1502	5.7861	0.0533	0.0803	5.6323
	PaPaGei-P	0.0534	0.1502	4.9935	0.0480	0.1135	3.3352
	Pulse-PPG	0.2359	5.6784	28.4984	0.2240	5.5162	26.4704
PPG-DaLiA HR	AnyPPG	0.0479	0.2173	4.0448	0.0440	0.1583	3.1867

Appendix

Table 14: Comparison of computational efficiency.

Figure 13: Visualization of the Overlap Rate.

Figure 14: Distribution of the Top 10% Filters in the PAPAGEI-P.

Figure 15: Hyperparameter sensitivity of SPAFT on the UCI-BP DBP task across different PPG FMs, including the cosine threshold, global pruning ratio, and fine-tuning ratio.

Figure 16: Hyperparameter sensitivity of SPAFT on the BUT PPG HR task across different PPG FMs, including the cosine threshold, global pruning ratio, and fine-tuning ratio.

FM	Method	MSE $\downarrow$	MAE $\downarrow$
AnyPPG	LoRA	78.374	6.224
	Taylor-FO	53.811	4.811
	Top 1% Adaptation	59.508	5.175
PaPaGei-P	LoRA	88.482	6.768
	Taylor-FO	89.920	6.882
	Top 1% Adaptation	74.135	5.942

Appendix

Table 15: Performance of adapting only the top 1% task-sensitive filters.

Time-Series Foundation Models

Time-series foundation models are pretrained sequence models designed to transfer across forecasting, classification, anomaly detection, and other temporal tasks.

See all 301 papers

Oct 9, 2026

PPG-Bench: A Unified Benchmark for PPG Foundation Models

Photoplethysmography (PPG) is increasingly recorded by wearable and clinical devices at enormous scale, and a growing number of PPG foundation models (FMs) have been developed to compute reusable representations for various downstream tasks. These models are difficult to compare: different evaluations use different datasets, task definitions, preprocessing, downstream heads, hyperparameter tuning, and baselines. We present PPG-Bench, a unifying open-source benchmark evaluating five public PPG FMs across nine open-access datasets and 26 tasks spanning cardiovascular state estimation, activity, sleep, screening, and identification. We evaluate models under four downstream regimes of increasing adaptability (linear probing, random forests, frozen MLP probes, and end-to-end finetuning), with model-specific hyperparameter optimization and subject-disjoint cross-validation. We also evaluate against three controls: random/constant baseline, a deterministic pulse-rate-variability (PRV) feature set, and a generic time-series FM (MOMENT). Across all 68 task-regime combinations, 35.3% show no meaningful FM advantage over the random/constant baseline, and in 83.8% no PPG-specific FM outperforms both PRV and MOMENT. We also find that added downstream head capacity generally improves performance, but can cause significant degradation if the model is performing at chance/constant level. Current PPG FMs therefore cannot yet claim to provide generalised and transferrable representations of PPG signals, and future evaluation in this field should include checks against simple, non-learned controls. Code for the benchmark and instructions to evaluate more models can be found here - https://github.com/mannas-ai/ppg-bench.

Arvasu Kulkarni, Mahir Jain, Aditya Ray Mishra +2

Mannas AI

Photoplethysmography Time-Series Foundation Models

Ask PDF

Jun 5, 2026

A robust PPG foundation model using multimodal physiological supervision

Photoplethysmography (PPG), a non-invasive measure of changes in blood volume, is widely used in both wearable devices and clinical settings. Recent PPG foundation models either use open-source ICU datasets with pretraining paradigms that require curated data and thus complicate generalization to field-like data, or use closed-source field-like PPG data. In contrast, we propose a PPG foundation model that does not require high-quality or field-like pretraining data, and instead leverages accompanying electrocardiogram and respiratory signals in ICU datasets to select contrastive samples during pretraining. Our approach allows the model to retain and learn from noisy PPG segments, improving robustness at inference. Our model, pretrained on 3x fewer subjects than existing state-of-the-art approaches, achieves performance improvements on 14 out of 15 diverse downstream tasks, including field-like daily activity and heart rate prediction. Our results demonstrate that multimodal supervision can integrate complementary physiological information to improve the robustness of PPG foundation models and enhance their generalization to consumer-grade data.

Eloy Geenjaar, Vince Calhoun, Scott Daly +4

Department of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, USA · Tri-Institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS), Georgia State University, Georgia Institute of Technology, Emory University, Atlanta, USA · Dolby Laboratories, San Francisco, USA

Multimodal Robustness Multimodal Pretraining Photoplethysmography Multimodal Contrastive Learning

Ask PDF

Jun 13, 2026

CAP: Towards PPG Universal Representation Learning with Patient-level Supervision

Photoplethysmography (PPG) plays a central role in wearable health monitoring and clinical decision support. Yet existing approaches to universal PPG representation learning largely focus on signal-level objectives and often overlook patient-level health context, which limits generalization to complex clinical tasks and heterogeneous cohorts. To address this gap, we construct a large-scale paired PPG-EHR multimodal dataset by distilling fragmented medical histories and clinical records into cohesive, patient-level electronic health records (EHR). Building on this resource, we propose Clinical Anchored Pretraining for PPG (CAP). During pretraining, CAP performs cross-modal contrastive alignment that anchors PPG representations to patient-level clinical semantics, guiding the encoder beyond waveform fitting toward modeling consistency in a patient's overall physiological state. During downstream adaptation, the pretrained PPG encoder provides clinically grounded representations that strengthen inductive bias and improve robustness and transferability. Experiments demonstrate that CAP consistently outperforms strong baselines on four diverse downstream tasks. CAP achieves a particularly large gain on respiratory rate prediction (up to +87.6% relative improvement over the state-of-the-art baseline) and delivers an average relative +26.7% across all tasks. We further enhance the interpretability of our approach through comprehensive analyses, including ablations and multiple complementary visualizations of the learned representations. The code for our experiments is available at: https://github.com/gody123gody/CAP .

Chenyang He, Xinyi Shao, Shun Huang +4

Nanjing University of Aeronautics and Astronautics Nanjing, Jiangsu, China · Peking University Beijing, Beijing, China · Independent Researcher +1

Representation Learning Cross-Modal Representation Learning Multimodal Pretraining Photoplethysmography Multimodal Contrastive Learning Wearable Health Monitoring

Ask PDF

Abstract

Time-Series Foundation Models

Explore similar work

PPG-Bench: A Unified Benchmark for PPG Foundation Models

A robust PPG foundation model using multimodal physiological supervision

CAP: Towards PPG Universal Representation Learning with Patient-level Supervision