Accurate differentiation between Left Bundle Branch Block (LBBB) and its strict subtype (sLBBB) is essential for optimizing patient selection for Cardiac Resynchronization Therapy (CRT), yet remains clinically challenging. This study proposes and compares two graph-theory-based pipelines for automated classification of 12-lead electrocardiograms (ECGs) into Healthy, LBBB, and sLBBB categories. Functional connectivity graphs were constructed from inter-lead measures, including Pearson correlation, cross-correlation, and phase difference. The first approach combines Graph Signal Processing (GSP) with machine learning. Graph filtering was performed via spectral decomposition of the Laplacian matrix, selecting dominant eigenmodes and reconstructing signals through the inverse Graph Fourier Transform—integrating spatial and temporal features. The second approach converted connectivity matrices into grayscale images, classified using a Convolutional Neural Network (CNN), and incorporated Explainable AI (XAI) via Grad-CAM to visualize inter-lead interactions and enhance model transparency. The GSP-based method using phase difference and a Support Vector Machine achieved the highest performance (mean balanced accuracy ), while the CNN-based approach with cross-correlation images reached , offering improved interpretability. Both methods distinguished pathological from healthy cases, but precise classification between LBBB and sLBBB remains challenging. These results highlight the complementary value of graph-based ECG analysis and support future hybrid models for CRT stratification.
MoyaABuytaertDPenickaM, et al.State-of-the-art: Noninvasive assessment of left ventricular function through myocardial work. J Am Soc Echocardiogr2023; 36: 1027–1042.
2.
HuizarJFKaszalaKTanA, et al.Abnormal conduction-induced cardiomyopathy: JACC review topic of the week. J Am Coll Cardiol2023; 81: 1192–1200.
PuvrezADuchenneJDonalE, et al.Mechanical dyssynchrony as a selection criterion for cardiac resynchronization therapy: Design of the AMEND-CRT trial. ESC Heart Fail2024; 11: 4390–4399.
5.
DaubertCBeharNMartinsRP, et al.Avoiding non-responders to cardiac resynchronization therapy: a practical guide. Eur Heart J2017; 38: 1463–1472.
6.
OrtegaDFBarjaLDLogarzoE, et al.Non-selective his bundle pacing with a biphasic waveform: enhancing septal resynchronization. EP Europace2017; 20: 816–822.
7.
AnderssonLGWuKCWieslanderB, et al.Left ventricular mechanical dyssynchrony by cardiac magnetic resonance is greater in patients with strict v’s nonstrict electrocardiogram criteria for left bundle-branch block. Am Heart J2013; 165: 956–963.
8.
MacasBCFerrándezJMOrellanaDV, et al.A model of mechanical dyssynchrony based on ECG features. In: Ballina FE, Armentano R, Acevedo RC et al. (eds.) Advances in bioengineering and clinical engineering. SABI 2023, IFMBE Proceedings, Vol. 114, 2024, pp.88–94. Cham: Springer. ISBN 978-3-031-61973-1. DOI: 10.1007/978-3-031-61973-1_9.
9.
StraussDGSelvesterRHWagnerGS. Defining left bundle branch block in the era of cardiac resynchronization therapy. Am J Cardiol2011; 107: 927–934.
10.
ZusterzeelRVicenteJOchoa-JimenezR, et al.The 43rd international society for computerized electrocardiology (ISCE) ECG initiative for the automated detection of strict left bundle branch block (LBBB). J Electrocardiol2018; 51: S25–S30.
11.
StoiculescuFMHadareanuDRHadareanuCD, et al.Refining cardiac resynchronization therapy: a comprehensive review on the role of advanced multimodality imaging. Front Cardiovasc Med2024; 11: 1406899.
12.
MartisRJAcharyaURAdeliH. Current methods in electrocardiogram characterization. Comput Biol Med2014; 48: 133–149.
13.
AnsariYMouradOQaraqeK, et al.Deep learning for ECG arrhythmia detection and classification: an overview of progress for period 2017-2023. Front Physiol2023; 14: 1246746.
14.
Amezquita-SanchezJPValtierra-RodriguezMAdeliH, et al.A novel wavelet transform-homogeneity model for sudden cardiac death prediction using ECG signals. J Med Syst2018; 42: 176.
15.
AyanoYMSchwenkerFDuferaBD, et al.Interpretable machine learning techniques in ECG-based heart disease classification: A systematic review. Diagnostics (Basel)2022; 13: 111.
16.
GórrizJMÁlvarez IllánIÁlvarez MarquinaA, et al.Computational approaches to explainable artificial intelligence: Advances in theory, applications and trends. Inf Fusion2023; 100: 101945.
17.
MadanPSinghVSinghDP, et al.A hybrid deep learning approach for ECG-based arrhythmia classification. Bioengineering (Basel)2022; 9: 152.
18.
MichailidisPMichailidisITGkeliosS, et al.Neuro-distributed cognitive adaptive optimization for training neural networks in a parallel and asynchronous manner. Integr Comput Aided Eng2023; 31: 19–41.
19.
AlonsoSMoránAPérezD, et al.Gap imputation in related multivariate time series through recurrent neural network-based denoising autoencoder. Integr Comput Aided Eng2023; 31: 157–172.
20.
FanLHeFSongY, et al.Look inside 3D point cloud deep neural network by patch-wise saliency map. Integr Comput Aided Eng2023; 31: 197–212.
21.
MalkovaAAminiMRDenisB, et al.Neural architecture search for radio map reconstruction with partially labeled data. Integr Comput Aided Eng2024; 31: 285–305.
22.
ZhouCFanLNeriF. A spatio-temporal fusion deep learning network with application to lightning nowcasting. Integr Comput Aided Eng2024; 31: 233–247.
23.
SuccettiFRosatoAPanellaM. Multi-label classification with imbalanced classes by fuzzy deep neural networks. Integr Comput Aided Eng2024; 32: 25–38.
24.
RafieiMHAdeliH. A new neural dynamic classification algorithm. IEEE Trans Neural Netw Learn Syst2017; 28: 3074–3083.
25.
AlamKMRSiddiqueNAdeliH. A dynamic ensemble learning algorithm for neural networks. Neural Comput Appl2020; 32: 8675–8690.
26.
PereiraDRPiteriMASouzaAN, et al.FEMa: a finite element machine for fast learning. Neural Comput Appl2020; 32: 6393–6404.
27.
RafieiMHGauthierLVAdeliH, et al.Self-supervised learning for electroencephalography. IEEE Trans Neural Netw Learn Syst2024; 35: 1457–1471.
28.
RafieiMHGauthierLVAdeliH, et al.Self-supervised learning for near-wild cognitive workload estimation. J Med Syst2024; 48: 107.
29.
YangTGreggREBabaeizadehS. Detection of strict left bundle branch block by neural network and a method to test detection consistency. Physiol Meas2020; 41: 025005.
30.
SadeghiARezaeeAHajatiF. Deep conv-attention model for diagnosing left bundle branch block from 12-lead electrocardiograms. arXiv preprint arXiv:221204936 2023.
31.
CalazansMAAFerreiraFABSSantosFAN, et al.Machine learning and graph signal processing applied to healthcare: A review. Bioengineering (Basel)2024; 11: 671.
32.
AhmadlouMAdeliH. Complexity of weighted graph: A new technique to investigate structural complexity of brain activities with applications to aging and autism. Neurosci Lett2017; 650: 103–108.
33.
QiangYDongXLiuX, et al.Conv-RGNN: An efficient convolutional residual graph neural network for ECG classification. Comput Methods Programs Biomed2024; 257: 108406.
34.
HanXZhangALiR, et al.An arrhythmia intelligent recognition method based on a multimodal information and spatio-temporal hybrid neural network model. Comput Mater Continua2025; 82: 3443–3465.
35.
AndayeshgarBAbdali-MohammadiFSepahvandM, et al.Developing graph convolutional networks and mutual information for arrhythmic diagnosis based on multi-channel ECG signals. Int J Environ Res Public Health2022; 19: 10511.
36.
OrdoñezBCMVillavicencioDVOOchoaMAS, et al.Application of Graph Fourier Transform (GFT) in the Diagnosis of Left Bundle Branch Block (LBBB) from Electrocardiographic (ECG) Signals, Lecture Notes in Computer Science, volume 14675, 2024, pp.495–503. Cham: Springer. ISBN 978-3-031-61136-0. DOI: 10.1007/978-3-031-61137-7_46.
37.
SelvarajuRRCogswellMDasA, et al.Grad-CAM: Visual explanations from deep networks via gradient-based localization. Int J Comput Vis2019; 128: 336–359.
38.
LundbergSMLeeSI. A unified approach to interpreting model predictions. In: Guyon I, Von Luxburg U, Bengio S et al. (eds.) Advances in neural information processing systems, Volume 30, 2017, Curran Associates, Inc.
39.
HicksSAIsaksenJLThambawitaV, et al.Explaining deep neural networks for knowledge discovery in electrocardiogram analysis. Sci Rep2021; 11: 10949.
HughesJWOlginJEAvramR, et al.Performance of a convolutional neural network and explainability technique for 12-lead electrocardiogram interpretation. JAMA Cardiol2021; 6: 1285–1295.
42.
ReznichenkoSWhitakerJNiZ, et al.Comparing ECG lead subsets for heart arrhythmia/ECG pattern classification: Convolutional neural networks and random forest. CJC Open2024; 7: 176–186.
43.
KaratziaLAungNAksentijevicD. Artificial intelligence in cardiology: Hope for the future and power for the present. Front Cardiovasc Med2022; 9: 945726. DOI: 10.3389/fcvm.2022.945726
44.
Macas OrdoñezBCGarrigósJMartínezJJ, et al.An explainable machine learning system for left bundle branch block detection and classification. Integr Comput Aided Eng2024; 31: 43–58.
45.
Pérez-RieraARBarbosa-BarrosRde Rezende BarbosaMPC, et al.Left bundle branch block: Epidemiology, etiology, anatomic features, electrovectorcardiography, and classification proposal. Ann Noninvasive Electrocardiol2019; 24: e12572.
46.
ZhengJZhangJDaniokoS, et al.A 12-lead electrocardiogram database for arrhythmia research covering more than 10,000 patients. Sci Data2020; 7: 48.
47.
MossAJBrownMWCannomDS, et al.Multicenter automatic defibrillator implantation trial-cardiac resynchronization therapy (MADIT-CRT): design and clinical protocol. Ann Noninvasive Electrocardiol2005; 10: 34–43.
WoodyCD. Characterization of an adaptive filter for the analysis of variable latency neuroelectric signals. Med Biol Eng1967; 5: 539–554.
50.
KeGMengQFinleyT, et al.LightGBM: A highly efficient gradient boosting decision tree. In: Advances in neural information processing systems, volume 30, 2017, pp.3146–3154.
51.
KligfieldPGettesLSBaileyJJ, et al.Recommendations for the standardization and interpretation of the electrocardiogram: Part I: The electrocardiogram and its technology. A scientific statement from the American heart association electrocardiography and arrhythmias committee, council on clinical cardiology; the American college of cardiology foundation; and the heart rhythm society. endorsed by the international society for computerized electrocardiology. Circulation2007; 115: 1306–1324.
52.
HarrisCRMillmanKJvan der WaltSJ, et al.Array programming with numPy. Nature2020; 585: 357–362.
53.
BradskiGKaehlerA. Learning OpenCV: Computer vision with the OpenCV library. O’Reilly Media Inc., 2008.
54.
SrivastavaNHintonGKrizhevskyA, et al.Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res2014; 15: 1929–1958.
55.
AydoreSPantazisDLeahyRM. A note on the phase locking value and its properties. Neuroimage2013; 74: 231–244.
56.
ShumanDINarangSKFrossardP, et al.The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains. IEEE Signal Process Mag2013; 30: 83–98.
57.
VirtanenPGommersROliphantTE, et al.SciPy 1.0: Fundamental algorithms for scientific computing in python. Nat Methods2020; 17: 261–272.
58.
PedregosaFVaroquauxGGramfortA, et al.Scikit-learn: Machine learning in python. J Mach Learn Res2011; 12: 2825–2830.
59.
LemaîtreGNogueiraFAridasCK. Imbalanced-learn: A python toolbox to tackle the curse of imbalanced datasets in machine learning. J Mach Learn Res2017; 18: 1–5.
60.
JolliffeITCadimaJ. Principal component analysis: a review and recent developments. Philos Trans A Math Phys Eng Sci2016; 374: 20150202.
61.
BrodersenKHGallusserFKoechlinJM, et al.The balanced accuracy and its posterior distribution. In: Proceedings of the 20th international conference on pattern recognition (ICPR), 2010, pp.3121–3124. IEEE. DOI: 10.1109/ICPR.2010.764.
62.
PonnusamySSVijayaramanPEllenbogenK. Left bundle branch block-associated cardiomyopathy: A new approach. Arrhythm Electrophysiol Rev2024; 13: e15.
63.
SankariZAdeliH. HeartSaver: a mobile cardiac monitoring system for auto-detection of atrial fibrillation, myocardial infarction, and atrio-ventricular block. Comput Biol Med2011; 41: 211–220.
64.
WuZGuoC. Deep learning and electrocardiography: systematic review of current techniques in cardiovascular disease diagnosis and management. Biomed Eng Online2025; 24: 23.
65.
AnJGreggRBaileyB, et al.Left bundle branch block detection in 12-Lead ECG using End-to-End deep learning with explainability. In: Proc. Computing in Cardiology (CinC). DOI: 10.22489/CinC.2024.067.
