AI- and IoT-Enabled Solutions for Healthcare.

Patient care and management have entered a new arena, where intelligent technology can assist clinicians in both diagnosis and treatment [...].

Non-Invasive Heart Failure Evaluation Using Machine Learning Algorithms.

Heart failure is a prevalent cardiovascular condition with significant health implications, necessitating effective diagnostic strategies for timely intervention. This study explores the potential of continuous monitoring of non-invasive signals, specifically integrating photoplethysmogram (PPG) and electrocardiogram (ECG), for enhancing early detection and diagnosis of heart failure. Leveraging a dataset from the MIMIC-III database, encompassing 682 heart failure patients and 954 controls, our approach focuses on continuous, non-invasive monitoring. Key features, including the QRS interval, RR interval, augmentation index, heart rate, systolic pressure, diastolic pressure, and peak-to-peak amplitude, were carefully selected for their clinical relevance and ability to capture cardiovascular dynamics. This feature selection not only highlighted important physiological indicators but also helped reduce computational complexity and the risk of overfitting in machine learning models. The use of these features in training machine learning algorithms led to a model with impressive accuracy (98%), sensitivity (97.60%), specificity (96.90%), and precision (97.20%). Our integrated approach, combining PPG and ECG signals, demonstrates superior performance compared to single-signal strategies, emphasizing its potential in early and precise heart failure diagnosis. The study also highlights the importance of continuous monitoring with wearable technology, suggesting a significant stride forward in non-invasive cardiovascular health assessment. The proposed approach holds promise for implementation in hardware systems to enable continuous monitoring, aiding in early detection and prevention of critical health conditions.

Effect of occupational exposure to antineoplastic drugs on DNA damage in nurses: a cross-sectional study.

BACKGROUND: Although the therapeutic effect of antineoplastic drugs is incontestable, these agents can also potentially act as carcinogens, mutagens and/or teratogens in people. The aim of this study was to assess the effect of occupational exposure to antineoplastic drugs on DNA damage, assessed by the comet assay and cytokinesis-block micronucleus (CBMN) assay, in nurses. METHODS: The cross-sectional study enrolled 305 nursing staff members from 7 public hospitals in Shenzhen who handled antineoplastic drugs, and 150 healthy nursing staff members who were not exposed to antineoplastic drugs as the control group. DNA damage was assessed by the comet and CBMN assay. Multiple linear regressions and logistic regressions models were used to analyse the effect of occupational exposure to antineoplastic drugs on DNA damage. RESULTS: After adjustment for confounding factors, compared with non-exposure to antineoplastic drugs, exposure to antineoplastic drugs was positively related to tail moment, olive moment, tail length and tail DNA per cent, and adjusted ß or OR (95% CI) was 0.17 (0.08 to 0.26), 0.18 (0.10 to 0.27), 1.03 (0.47 to 1.60) and 1.16 (1.04 to 1.29) (all p<0.05). Moreover, similar significant relationships were observed for the biomarkers of the CBMN assay. Additionally, other than age, there was no interaction between antineoplastic drug exposure and other variables for the levels of biomarkers of the CBMN assay and the comet assay. CONCLUSIONS: The present results showed that exposure to antineoplastic drugs was positively related to the risk of DNA damage in nurses. The results imply that occupational exposure to antineoplastic agents is an important global public health problem that requires urgent attention.

Gleevec and Rapamycin Synergistically Reduce Cell Viability and Inhibit Proliferation and Angiogenic Function of Mouse Bone Marrow-Derived Endothelial Progenitor Cells.

OBJECTIVE: This study investigates the synergistic effects of Gleevec (imatinib) and rapamycin on the proliferative and angiogenic properties of mouse bone marrow-derived endothelial progenitor cells (EPCs). MATERIALS AND METHODS: EPCs were isolated from mouse bone marrow and treated with different concentrations of Gleevec or rapamycin individually or in combination. The cell viability and proliferation were examined using the MTT assay. An analysis of cell cycle and apoptosis was performed using flow cytometry. Formation of capillary-like tubes was examined in vitro, and the protein expression of cell differentiation markers was determined using Western blot analysis. RESULTS: Gleevec significantly reduced cell viability, cell proliferation, and induced cell apoptosis in EPCs. Rapamycin had similar effects on EPCs, but it did not induce cell apoptosis. The combination of Gleevec and rapamycin reduced the cell proliferation but increased cell apoptosis. Although rapamycin had no demonstratable effect on tube formation, the combined therapy of Gleevec and rapamycin significantly reduced tube formation when compared with Gleevec alone. Mechanistically, Gleevec, but not rapamycin, induced a significant elevation in caspase-3 activity in EPCs, and it attenuated the expression of the endothelial protein marker platelet-derived growth factor receptor α. Functionally, rapamycin, but not Gleevec, significantly enhanced the expression of endothelial differentiation marker proteins, while attenuating the expression of mammalian target of rapamycin signaling-related proteins. CONCLUSIONS: Gleevec and rapamycin synergistically suppress cell proliferation and tube formation of EPCs by inducing cell apoptosis and endothelial differentiation. Mechanistically, it is likely that rapamycin enhances the proapoptotic and antiangiogenic effects of Gleevec by promoting the endothelial differentiation of EPCs. Given that EPCs are involved in the pathogenesis of some cardiovascular diseases and critical to angiogenesis, pharmacological inhibition of EPC proliferation by combined Gleevec and rapamycin therapy may be a promising approach for suppressing cardiovascular disease pathologies associated with angiogenesis.

Daily self-care practices influence exit-site condition in patients having peritoneal dialysis: A multicenter cross-sectional survey.

AIMS: To investigate practice patterns in exit-site care and identify the risk factors for exit-site infection. DESIGN: A quantitative cross-sectional design. METHODS: Data were collected in 12 peritoneal dialysis (PD) centres in 2018. Daily exit-site care practice patterns and exit-site status of patients receiving PD were assessed through interviews and questionnaires. RESULTS/FINDINGS: Most of the 1,204 patients adhered with the protocols about main aspects of exit-site care, such as cleansing agents selection, frequency of cleansing, catheter fixation, and following the catheter protective measures. However, their adherence levels on hand hygiene, mask wearing, observing exit site, examining secretion, and communicating with PD staff were rather low. Eighty-four patients' exit sites were evaluated as problematic exit site (PES). And 186 patients had catheter-related infection (CRI) history. After multivariable logistic regression analysis, diabetes (OR = 1.631), traction bleeding history (OR = 2.697), antibiotic agents use (OR = 2.460), compliance on mask wearing (OR = 0.794), and observing exit site (OR = 0.806) were influencing factors of CRI history. Traction bleeding history (OR = 2.436), CRI history (OR = 10.280), and effective communication (OR = 0.808) with PD staff were influencing factors for PES. CONCLUSIONS: The adherence levels on different aspects of exit-site care were varied in patients having PD. Their self-care behaviours did correlate with the exit-site status. IMPACT: The adherence level of patients' exit-site care practice needs attention of medical staff. Further studies about the optimal procedure in exit-site care were warranted.

Examination of Potential of Thermopile-Based Contactless Respiratory Gating.

To control the spread of coronavirus disease 2019 (COVID-19), it is effective to perform a fast screening of the respiratory rate of the subject at the gate before entering a space to assess the potential risks. In this paper, we examine the potential of a novel yet cost-effective solution, called thermopile-based respiratory gating, to contactlessly screen a subject by measuring their respiratory rate in the scenario with an entrance gate. Based on a customized thermopile array system, we investigate different image and signal processing methods that measure respiratory rate from low-resolution thermal videos, where an automatic region-of-interest selection-based approach obtains a mean absolute error (MAE) of 0.8 breaths per minute. We show the feasibility of thermopile-based respiratory gating and quantify its limitations and boundary conditions in a benchmark (e.g., appearance of face mask, measurement distance and screening time). The technical validation provided by this study is helpful for designing and implementing a respiratory gating solution toward the prevention of the spread of COVID-19 during the pandemic.

Wearable Vital Signs Monitoring for Patients With Asthma: A Review.

Worldwide,an estimated 461 000 people die from asthma attacks each year. While there remain treatments to alleviate asthma symptoms and reduce deaths, patient deterioration needs to be identified in sufficient time. To prevent asthma deterioration, patients need to be aware of personal and environmental triggers and monitor their asthma symptoms. The aim of this article is to provide a comprehensive review of the current state-of-the-art wearable sensors and devices that use vital signs for asthma patient monitoring and management. Among all vital signs, breathing rate and airflow sound are key indicators of asthmatic patients' health that can be measured directly using wearable sensors to provide continuous and constant patient monitoring or indirectly by estimations based on proven algorithms using electrocardiogram (ECG), photoplethysmogram (PPG), and chest movements. ECG and PPG signals are widely used in smart watches and chest bands, enabling easy integration of a more extensive body sensor framework for asthmatic exacerbation prediction. Other vital signs used in asthma patient monitoring include blood oxygen saturation, temperature, blood pressure, verbal sound, and pain responses. The use of wearable vital signs enabled a broad range of wearable sensor application scenarios for asthma monitoring and management.

Expression pattern of microRNAs related with response to trastuzumab in breast cancer.

BACKGROUND: Although an immense effort has been made to develop a novel biomarker for response to trastuzumab, no reliable biomarkers are available to guide management, expect for HER2. The aim of this study was to examine the relationship between microRNA (miRNA) expression and resistance to trastuzumab. METHODS: Differentially expressed miRNAs between trastuzumab-resistant and trastuzumab-sensitive cell lines were analyzed using microarrays. We performed Gene Ontology and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses to determine the functions of differentially expressed miRNA and their targeted genes. Furthermore, the protein-protein interactions (PPI) network was analyzed. Serum samples were collected from patients with HER2-positive breast cancer who were treated with trastuzumab. We validated the miRNAs expression levels by quantitative reverse-transcription polymerase chain reaction (qRT-PCR) in these serums. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the predictive performance of the miRNA. RESULTS: Using miRNA microarrays, 151 miRNAs that significant differentially expressed between the trastuzumab-resistant and sensitive cells were identified, including 46 upregulated and 105 downregulated miRNAs. Results of real-time PCR confirmed seven miRNAs in cell lines. PI3K-Akt signaling pathway was involved in regulating biological function according to KEGG analysis. Compared with the serums of trastuzumab-sensitive patients, three miRNAs, namely miR-200b, miR-135b, and miR-29a, were identified to be upregulated, and miR-224 was downregulated in the trastuzumab-resistant serums. ROC analysis showed that four miRNAs were correlated with trastuzumab resistance. Furthermore, three subnetwork modules of PPI network were obtained. CONCLUSION: The results indicated that miRNAs were reliable predictive biomarkers for response to trastuzumab.

Smart mechanosensing machineries enable migration of vascular smooth muscle cells in atherosclerosis-relevant 3D matrices.

At the early stage of atherosclerosis, neointima is formed due to the migration of vascular smooth muscle cells (VSMCs) from the media to the intima. VSMCs are surrounded by highly adhesive 3D matrices. They take specific strategies to cross various 3D matrices in the media, including heterogeneous collagen and mechanically strong basement membrane. Migration of VSMCs is potentially caused by biomechanical mechanism. Most in vitro studies focus on cell migration on 2D substrates in response to biochemical factors. How the cells move through 3D matrices under the action of mechanosensing machineries remains unexplored. In this review, we propose that several interesting tension-dependent machineries act as "tractor"-posterior myosin II accumulation, and "wrecker"-anterior podosome maintaining, to power VSMCs ahead. VSMCs embedded in 3D matrices may accumulate a minor myosin II isoform, myosin IIB, at the cell rear. Anisotropic myosin IIB distribution creates cell rear, polarizes cell body, pushes the nucleus and reshapes the cell body, and cooperates with a uniformly distributed myosin IIA to propel the cell forward. On the other hand, matrix digestion by podosome further promote the migration when the matrix becomes denser. Actomyosin tension activates Src to induce podosome in soft 3D matrices and retain the podosome integrity to steadily digest the matrix.

Clinical Efficacy of Andrographolide Sulfonate in the Treatment of Severe Hand, Foot, and Mouth Disease (HFMD) is Dependent upon Inhibition of Neutrophil Activation.

Andrographolide sulfonate treatment has been shown to improve clinical severe hand, foot, and mouth disease (HFMD) efficacies when combined with conventional therapy. However, the mechanisms for its therapeutic effects remain elusive. In this study, we aimed to investigate whether andrographolide sulfonate exerts its efficacy by acting on neutrophil activation. We obtained serial plasma samples at two time points (before and after 5 days of therapy) from 28 HFMD patients who received conventional therapy and 18 patients who received combination therapy (andrographolide sulfonate plus conventional therapy). Then, we measured plasma myeloperoxidase (MPO), S100A8/A9, histone, and inflammatory cytokine levels. Furthermore, we examined if andrographolide sulfonate had direct effects on neutrophil activation in vitro. We observed that MPO and S100A8/A9 levels were markedly elevated in the HFMD patients before clinical treatment. At 5 days post-medication, the MPO, S100A8/A9, histone, and interleukin-6 levels were markedly lower in the combination therapy group compared with the conventional therapy group. In vitro studies showed that andrographolide sulfonate inhibited lipopolysaccharide-stimulated neutrophil activation, demonstrated by the decreased production of reactive oxygen species and cytokines. These data indicate that neutrophil activation modulation by andrographolide sulfonate may be a critical determinant for its clinical HFMD treatment efficacy. Copyright © 2015 John Wiley & Sons, Ltd.

CiGNN: A Causality-Informed and Graph Neural Network Based Framework for Cuffless Continuous Blood Pressure Estimation.

Causalityholds profound potentials to dissipate confusion and improve accuracy in cuffless continuous blood pressure (BP) estimation, an area often neglected in current research. In this study, we propose a two-stage framework, CiGNN, that seamlessly integrates causality and graph neural network (GNN) for cuffless continuous BP estimation. The first stage concentrates on the generation of a causal graph between BP and wearable features from the the perspective of causal inference, so as to identify features that are causally related to BP variations. This stage is pivotal for the identification of novel causal features from the causal graph beyond pulse transit time (PTT). We found these causal features empower better tracking in BP changes compared to PTT. For the second stage, a spatio-temporal GNN (STGNN) is utilized to learn from the causal graph obtained from the first stage. The STGNN can exploit both the spatial information within the causal graph and temporal information from beat-by-beat cardiac signals for refined cuffless continuous BP estimation. We evaluated the proposed method with three datasets that include 305 subjects (102 hypertensive patients) with age ranging from 20-90 and BP at different levels, with the continuous Finapres BP as references. The mean absolute difference (MAD) for estimated systolic blood pressure (SBP) and diastolic blood pressure (DBP) were 3.77 mmHg and 2.52 mmHg, respectively, which outperformed comparison methods. In all cases including subjects with different age groups, while doing various maneuvers that induces BP changes at different levels and with or without hypertension, the proposed CiGNN method demonstrates superior performance for cuffless continuous BP estimation. These findings suggest that the proposed CiGNN is a promising approach in elucidating the causal mechanisms of cuffless BP estimation and can substantially enhance the precision of BP measurement.

Detection of Coronary Artery Disease Based on Clinical Phonocardiogram and Multiscale Attention Convolutional Compression Network.

Heart sound is an important physiological signal that contains rich pathological information related with coronary stenosis. Thus, some machine learning methods are developed to detect coronary artery disease (CAD) based on phonocardiogram (PCG). However, current methods lack sufficient clinical dataset and fail to achieve efficient feature utilization. Besides, the methods require complex processing steps including empirical feature extraction and classifier design. To achieve efficient CAD detection, we propose the multiscale attention convolutional compression network (MACCN) based on clinical PCG dataset. Firstly, PCG dataset including 102 CAD subjects and 82 non-CAD subjects was established. Then, a multiscale convolution structure was developed to catch comprehensive heart sound features and a channel attention module was developed to enhance key features in multiscale attention convolutional block (MACB). Finally, a separate downsampling block was proposed to reduce feature losses. MACCN combining the blocks can automatically extract features without empirical and manual feature selection. It obtains good classification results with accuracy 93.43%, sensitivity 93.44%, precision 93.48%, and F1 score 93.42%. The study implies that MACCN performs effective PCG feature mining aiming for CAD detection. Further, it integrates feature extraction and classification and provides a simplified PCG processing case.

Endocrine disrupting chemical Bisphenol A and its association with cancer mortality: a prospective cohort study of NHANES.

Introduction: There is evidence suggesting that Bisphenol A (BPA) is associated with increased all-cause mortality in adults. However, the specific nature of the relationship between BPA exposure and cancer mortality remains relatively unexplored. Methods: The National Health and Nutrition Examination Survey (NHANES) dataset was used to recruit participants. Urinary BPA was assessed using liquid chromatography-mass spectrum (LC-MS). Through the use of multivariable Cox proportional hazard regressions and constrained cubic splines, the relationships between urine BPA and death from all causes and cancer were investigated. Results: This study has a total of 8,035 participants, and 137 died from cancers after a 7.5-year follow-up. The median level of BPA was 2.0 g/mL. Urinary BPA levels were not independently associated with all-cause mortality. For cancer mortality, the second quartile's multivariable-adjusted hazard ratio was 0.51 (95% confidence interval: 0.30 to 0.86; p = 0.011) compared to the lowest quartile. The restricted cubic splines showed that the association was nonlinear (p for nonlinearity = 0.028) and the inflection point was 1.99 ng/mL. Conclusion: Urinary BPA exposure was U-shaped associated with the risk of cancer mortality, and a lower level of BPA less than 1.99 ng/mL was associated with a higher risk of cancer mortality.

Digital Health and Machine Learning Technologies for Blood Glucose Monitoring and Management of Gestational Diabetes.

Innovations in digital health and machine learning are changing the path of clinical health and care. People from different geographical locations and cultural backgrounds can benefit from the mobility of wearable devices and smartphones to monitor their health ubiquitously. This paper focuses on reviewing the digital health and machine learning technologies used in gestational diabetes - a subtype of diabetes that occurs during pregnancy. This paper reviews sensor technologies used in blood glucose monitoring devices, digital health innovations and machine learning models for gestational diabetes monitoring and management, in clinical and commercial settings, and discusses future directions. Despite one in six mothers having gestational diabetes, digital health applications were underdeveloped, especially the techniques that can be deployed in clinical practice. There is an urgent need to (1) develop clinically interpretable machine learning methods for patients with gestational diabetes, assisting health professionals with treatment, monitoring, and risk stratification before, during and after their pregnancies; (2) adapt and develop clinically-proven devices for patient self-management of health and well-being at home settings ("virtual ward" and virtual consultation), thereby improving clinical outcomes by facilitating timely intervention; and (3) ensure innovations are affordable and sustainable for all women with different socioeconomic backgrounds and clinical resources.

A general way to assay protein by coupling peptide with signal reporter via supermolecule formation.

Protein-binding peptide is recently recognized as an effective artificial affinity reagent for protein assays. However, its application is hampered by the limited choices of available signal readout methods. Herein, we report a general electrochemical signal readout method for protein-binding peptides exploiting the host-guest chemistry of cucurbituril. Via the formation of supermolecules among cucurbituril, electrochemical reporter, and the peptide, a protein-binding peptide can be noncovalently coupled with the electrochemical reporter. To assay the target protein, the protein-binding peptides are first self-assembled in the sensing layer, and after the capturing of the target protein, a portion of the peptides become protein-bound. The protein-free peptides are then coupled with the electrochemical reporter to yield a signal readout inversely proportional to the amount of the captured target proteins. Since the only requirement of supermolecule formation is the incorporation of aromatic amino acids in the peptide sequence, this strategy is universally applicable to many protein-binding peptides. The generality and target specificity of the proposed method are successfully demonstrated in the assays of two kinds of target proteins: tumor necrosis factor-α and amyloid ß 1-42 soluble oligomer, respectively. The feasibility of our method is also tested in the monitoring of tumor necrosis factor-α secretion activity of HL-60 cells. These results indicate that our method can have great use in protein detection in the future.

Causal Inference for Hypertension Prediction.

Hypertension is a leading cause of cardiovascular disease and premature death worldwide and it puts a heavy burden on the healthcare system. It is, therefore, very important to detect and evaluate hypertension and related cardiovascular events as to for efficient diagnosis, treatment and management. Hypertension can be evaluated with noninvasive cardiac signals, such as electrocardiogram (ECG) and photoplethysmogram (PPG) signals. Most of the previous studies predicted hypertension from ECG and PPG signals with extracted features that are correlated with hypertension. However, correlation is sometimes unreliable and may be affected by confounding factors. In this study, we propose a causal inference based approach to identify feature variables from ECG and PPG signals that are potentially causally related with hypertension. The method of greedy equivalence search was employed to construct the causal graph of features and hypertension. With causal features identified from the causal graph, we used machine learning models to diagnose hypertension. The machine learning classification models achieve great classification performance, among which random forest model has the best classification performance, with accuracy being 0.987, precision being 0.990, recall being 0.981, and F1-score being 0.985. The results show that the causal inference based approach can effectively predict hyper-tension.Clinical relevance- This paper proposes a new hypertension risk prediction method, which uses causality instead of correlation as the feature screening criteria to establish a causal graph of hypertension, which can predict the hypertension more reliably.

Continual Learning for Cuffless Blood Pressure Measurement using PPG and ECG Signals.

Although numerous studies have been conducted on cuffless blood pressure (BP) estimation using machine learning methods, most of the data-driven models are static, with model parameters fixed after training is complete. However, BP is dynamic and the performance would degrade for a static model when the to-be predicted BP distribution deviates from the training BP distribution. In this paper, we propose a continual learning (CL) framework in which deep learning models are developed to learn dynamically and continuously for arterial BP (ABP) estimation with photoplethysmography (PPG) and electrocardiogram (ECG) waveforms. The effectiveness of the CL model is validated on UCI Repository and MIMIC-III database with a total of 132 individual samples, and compared with conventional training method. It was found that the CL model improved the ABP estimation accuracy in terms of mean absolute error (MAE) by 17.47% on average compared with conventional training model. Furthermore, the improvement increased with the variability of ABP. These results demonstrate that CL model has potential to estimate dynamic ABP, which has been challenging with conventional training.

Current situation and influencing factors of the nursing practice environment in five tertiary general hospitals in Shenzhen: a cross-sectional study.

Objective: This study investigates the current situation and influencing factors of the nursing practice environment in Shenzhen, China, and provides suggestions for improving it. Background: Nursing shortage is an urgent global problem and also of concern in China. Studies have shown that better work environments are related to high job satisfaction and better patient outcomes. Methods: The 37-item Practice Environment Scale was used to assess the nursing practice environment. Respondents were 1116 nurses from five general tertiary hospitals in Shenzhen. Results: The mean satisfaction score for the nursing practice environment was 3.63 ± 0.72 (where 5 is the highest possible score). Position, being a specialist nurse, choice of nursing major, educational attainment, and night shifts significantly affected nurses' working environment satisfaction. Conclusion: The practice environment of nurses was satisfactory. We recommend reducing the workload and encouraging nurses to complete specialist training, and supporting nurses to expand their roles in hospitals and society to improve the nursing practice environment.

Physical activity and asthma symptom control in children during COVID-19 lockdown: A feasibility study.

Objective: The aim was to evaluate the impact of the COVID-19 lockdown on physical activity (PA) and asthma symptom control in children. Methods: We conducted a single-cohort observational study on 22 children with a median age of 9 (8-11) years with a diagnosis of asthma being included in the study. Participants were asked to wear a PA tracker for 3 months; during the same 3-month period, the Paediatric Asthma Diary (PAD) was administered daily and the Asthma Control (AC) Questionnaire together with the mini-Paediatric Asthma Quality of Life (AQoL) Questionnaire administered at weekly intervals. Results: Compared with the pre-lockdown period, there was a significant reduction in PA levels after the lockdown began. Daily total steps reduced by about 3000 steps (p < 0.001), very active minutes by 9 min (p < 0.001) and fairly active minutes almost halved (p < 0.001); while asthma symptom control marginally improved, with the AC and AQoL score improving by 0.56 (p < 0.005) and 0.47 (p < 0.05), respectively. Further, for those with AC score higher than 1, PA was positively associated with asthma control both before and after the lockdown. Conclusions: This feasibility study suggests that PA engagement of children with asthma is negatively affected during the pandemic, but the beneficial effect of PA on asthma symptom control potentially sustains even during a lockdown period. These findings emphasize the importance of wearable device to monitor longitudinal PA and thus better management of PA for achieving the best outcome of asthma symptom control.

Causal inference based cuffless blood pressure estimation: A pilot study.

Enabled by wearable sensing, e.g., photoplethysmography (PPG) and electrocardiography (ECG), and machine learning techniques, study on cuffless blood pressure (BP) measurement with data-driven methods has become popular in recent years. However, causality has been overlooked in most of current studies. In this study, we aim to examine the feasibility of causal inference for cuffless BP estimation. We first attempt to detect wearable features that are causally related, rather than correlated, to BP changes by identifying causal graphs of interested variables with fast causal inference (FCI) algorithm. With identified causal features, we then employ time-lagged link to integrate the mechanism of causal inference into the BP estimated model. The proposed method was validated on 62 subjects with their continuous ECG, PPG and BP signals being collected. We found new causal features that can better track BP changes than pulse transit time (PTT). Further, the developed causal-based estimation model achieved an estimation error of mean absolute difference (MAD) being 5.10 mmHg and 2.85 mmHg for SBP and DBP, respectively, which outperformed traditional model without consideration of causality. To the best of our knowledge, this work is the first to study the causal inference for cuffless BP estimation, which can shed light on the mechanism, method and application of cuffless BP measurement.