Academic Journal
Deep learning for carotid Doppler spectra classification.
| Τίτλος: | Deep learning for carotid Doppler spectra classification. |
|---|---|
| Συγγραφείς: | Bhikha C; Manufacturing Cluster, Council for Scientific and Industrial Research (CSIR), Pretoria, South Africa. CBhikha@csir.co.za., Dhuness K; Manufacturing Cluster, Council for Scientific and Industrial Research (CSIR), Pretoria, South Africa., Mennen M; Division of Cardiology, Department of Medicine, University of Cape Town and University of Cape Town/South African Medical Research Council Extramural Unit On Intersection of Noncommunicable Diseases and Infectious Diseases, Cape Town, South Africa., Jamieson-Luff N; Division of Cardiology, Department of Medicine, University of Cape Town and University of Cape Town/South African Medical Research Council Extramural Unit On Intersection of Noncommunicable Diseases and Infectious Diseases, Cape Town, South Africa., Ntusi NAB; Division of Cardiology, Department of Medicine, University of Cape Town and University of Cape Town/South African Medical Research Council Extramural Unit On Intersection of Noncommunicable Diseases and Infectious Diseases, Cape Town, South Africa., Wheatley R; Manufacturing Cluster, Council for Scientific and Industrial Research (CSIR), Pretoria, South Africa. |
| Πηγή: | Medical & biological engineering & computing [Med Biol Eng Comput] 2026 Jun; Vol. 64 (6), pp. 2133-2147. Date of Electronic Publication: 2026 Apr 14. |
| Τύπος έκδοσης: | Journal Article |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: Springer Country of Publication: United States NLM ID: 7704869 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1741-0444 (Electronic) Linking ISSN: 01400118 NLM ISO Abbreviation: Med Biol Eng Comput Subsets: MEDLINE |
| Imprint Name(s): | Publication: New York, NY : Springer Original Publication: Stevenage, Eng., Peregrinus. |
| Ιατρικοί όροι (MeSH): | Carotid Arteries*/diagnostic imaging , Ultrasonography, Carotid Arteries*/methods , Ultrasonography, Doppler*/methods , Classification Algorithms* , Deep Learning*, Cardiovascular Diseases/diagnostic imaging ; Adult ; Female ; Humans ; Male ; Middle Aged ; Convolutional Neural Networks |
| Περίληψη: | Cardiovascular disease (CVD) remains a global health challenge, with limited specialist access in low- and middle-income countries hindering early detection. Carotid Doppler ultrasound offers promise for screening in non-specialist settings. However, spectral Doppler lacks the anatomical context provided by duplex ultrasound, making it challenging to determine which carotid vessel is being assessed. This study focuses on accurately identifying signals from the common, internal, and external carotid arteries (CCA, ICA, and ECA) based solely on Doppler spectra. This forms a critical step for subsequent disease classification. A clinical study enrolled 398 participants who underwent bilateral carotid Doppler examination (198 healthy controls, 200 with CVD). Several classifiers were evaluated including i) five deep convolutional neural networks (CNN) utilizing transfer learning on spectral images, and ii) conventional machine learning classifiers applied to the maximum frequency envelope and extracted features. The best performing classifier was a CNN (GoogLeNet) which achieved a mean area under the curve (AUC) of 0.929, effectively distinguishing between carotid artery segments. It exhibited f1-scores of 0.830, 0.803 and 0.764 for the ICA, ECA and CCA, respectively. Explainable AI tools (GradCAM and LIME) provide intuitive visual insights into these predictions. This study addresses a previously unsolved problem of automated carotid artery segment identification using spectral Doppler waveforms alone. When incorporated into automated screening tools, this approach provides a low-cost, specialist-independent pathway for earlier CVD detection, particularly suited to resource-constrained environments. |
| Competing Interests: | Declarations. Ethics approval: All procedures in this study adhered to relevant laws and institutional guidelines, with ethics approval granted by the University of Cape Town and CSIR Ethics boards (UCT HREC REF: 749/2020 and CSIR REC registration number: 383/2021). The privacy rights of participants have been observed and informed consent was obtained. Competing interest: The authors declare that they have no competing or conflict of interests. |
| References: | WHO (2014) Global status report on noncommunicable diseases 2014,” World Health Organization (WHO). https://doi.org/10.5005/jp/books/11660_5 . Statistics South Africa (2023) Non-communicable diseases in South Africa: Findings from death notifications. 2008–2018 (Report No. 03–08–01). 03–08–01. Mantella LE, Johri AM (2021) A call for non-invasive tools to assess cardiovascular risk. J Am Soc Echocardiogr 34(3):A11–A12. https://doi.org/10.1016/j.echo.2021.01.006. (PMID: 10.1016/j.echo.2021.01.00633676611) Criqui MH, Alberts MJ, Fowkes FGR, Hirsch AT, O’Gara PT, Olin JW (2008) Atherosclerotic peripheral vascular disease symposium II: screening for atherosclerotic vascular diseases: should nationwide programs be instituted? Circulation 118(25):2830–2836. https://doi.org/10.1161/CIRCULATIONAHA.108.191172. (PMID: 10.1161/CIRCULATIONAHA.108.19117219106404) Shabani Varaki E, Gargiulo GD, Penkala S, Breen PP (2018) Peripheral vascular disease assessment in the lower limb: a review of current and emerging non-invasive diagnostic methods. Biomed Eng Online 17(1):1–27. https://doi.org/10.1186/s12938-018-0494-4. (PMID: 10.1186/s12938-018-0494-4) Nkosi S, Makin J, Hlongwane T, Pattinson RC (2019) Screening and managing a low-risk pregnant population using continuous-wave Doppler ultrasound in a low-income population: a cohort analytical study. S Afr Med J 109(5):347–352. https://doi.org/10.7196/SAMJ.2019.v109i5.13611. (PMID: 10.7196/SAMJ.2019.v109i5.1361131131803) Feucht U, Hlongwane T, Vannevel V, Mulol H, Botha T, Pattinson R (2022) Identifying the high-risk fetus in the low-risk mother using fetal Doppler screening. Glob Health Sci Pract 10(3):1–9. https://doi.org/10.9745/GHSP-D-21-00692. (PMID: 10.9745/GHSP-D-21-00692) Hlongwane TMAG, Cronje T, Nkosi BSS, Pattinson RC (2021) The prevalence of abnormal Doppler’s of the umbilical artery in a low-risk pregnant population in South Africa. EClinicalMedicine 34:100792. https://doi.org/10.1016/j.eclinm.2021.100792. (PMID: 10.1016/j.eclinm.2021.100792339977268102719) Vannevel V et al (2022) Antenatal Doppler screening for fetuses at risk of adverse outcomes: a multicountry cohort study of the prevalence of abnormal resistance index in low-risk pregnant women. BMJ Open 12(3):e053622. https://doi.org/10.1136/bmjopen-2021-053622. (PMID: 10.1136/bmjopen-2021-053622352964778928296) Lee W (2013) General principles of carotid Doppler ultrasonography. Ultrasonography 33(1):11–17. https://doi.org/10.14366/usg.13018. (PMID: 10.14366/usg.13018249364904058969) Dong J, Inthavong K, Tu J (2013) Image-based computational hemodynamics evaluation of atherosclerotic carotid bifurcation models. Comput Biol Med 43(10):1353–1362. https://doi.org/10.1016/j.compbiomed.2013.06.013. (PMID: 10.1016/j.compbiomed.2013.06.01324034726) De Wilde D et al (2016) Assessment of shear stress related parameters in the carotid bifurcation using mouse-specific FSI simulations. J Biomech 49(11):2135–2142. https://doi.org/10.1016/j.jbiomech.2015.11.048. (PMID: 10.1016/j.jbiomech.2015.11.04826655592) Fang H et al (2016) Compensatory patterns of collateral flow in stroke patients with unilateral and bilateral carotid stenosis. BMC Neurol 16(1):4–9. https://doi.org/10.1186/s12883-016-0560-0. (PMID: 10.1186/s12883-016-0560-0) Özşen S, Kara S, Latifoǧlu F, Güneş S (2007) A new supervised classification algorithm in artificial immune systems with its application to carotid artery Doppler signals to diagnose atherosclerosis. Comput Methods Programs Biomed 88(3):246–255. https://doi.org/10.1016/j.cmpb.2007.09.002. (PMID: 10.1016/j.cmpb.2007.09.00217976855) Polat K, Kara S, Latifoǧlu F, Güneş S (2007) Pattern detection of atherosclerosis from carotid artery doppler signals using fuzzy weighted pre-processing and Least Square Support Vector Machine (LSSVM). Ann Biomed Eng 35(5):724–732. https://doi.org/10.1007/s10439-007-9289-7. (PMID: 10.1007/s10439-007-9289-717387616) Latifoǧlu F, Polat K, Kara S, Güneş S (2008) Medical diagnosis of atherosclerosis from Carotid Artery Doppler Signals using principal component analysis (PCA), k-NN based weighting pre-processing and Artificial Immune Recognition System (AIRS). J Biomed Inform 41(1):15–23. https://doi.org/10.1016/j.jbi.2007.04.001. (PMID: 10.1016/j.jbi.2007.04.00117512260) Ceylan M, Ceylan R, Fatma D, Dirgenali F, Kara S, Özbay Y (2007) Classification of carotid artery Doppler signals in the early phase of atherosclerosis using complex-valued artificial neural network. Comput Biol Med 37(1):28–36. https://doi.org/10.1016/j.compbiomed.2005.08.005. (PMID: 10.1016/j.compbiomed.2005.08.00516343473) Übeyli ED, Güler I (2003) Neural network analysis of internal carotid arterial Doppler signals: predictions of stenosis and occlusion. Expert Syst Appl 25(1):1–13. https://doi.org/10.1016/S0957-4174(03)00002-2. (PMID: 10.1016/S0957-4174(03)00002-2) Luo X, Ara L, Ding H, Rollins D, Motaganahalli R, Sawchuk AP (2021) Computational methods to automate the initial interpretation of lower extremity arterial Doppler and duplex carotid ultrasound studies. J Vasc Surg 74(3):988-996.e1. https://doi.org/10.1016/j.jvs.2021.02.050. (PMID: 10.1016/j.jvs.2021.02.05033813023) Elwazir MY, Akkus Z, Oguz D, Ye Z, Oh JK (2020) Fully automated mitral inflow Doppler analysis using deep learning. In: Proceedings - IEEE 20th International Conference on Bioinformatics and Bioengineering, BIBE 2020, pp 691–696. https://doi.org/10.1109/BIBE50027.2020.00116. Frauchiger B, Schmid HP, Roedel C, Moosmann P, Staub D (2001) Comparison of carotid arterial resistive indices with intima-media thickness as sonographic markers of atherosclerosis. Stroke 32(4):836–841. https://doi.org/10.1161/01.STR.32.4.836. (PMID: 10.1161/01.STR.32.4.83611283379) D’Alessio T (1985) Objective algorithm for maximum frequency estimation in Doppler spectral analysers. Med Biol Eng Comput 23(January):63–68. https://doi.org/10.1007/BF02444029. (PMID: 10.1007/BF024440293974321) Li J, Zhang Y, Liu X, Liu P, Yin H, Liu DC (2019) A robust automatic ultrasound spectral envelope estimation. Information (Switzerland) 10(6). https://doi.org/10.3390/info10060199. Szegedy C et al (2015) Going deeper with convolutions. In: Proceedings of the IEEE Computer society conference on computer vision and pattern recognition, vol. 07–12-June, pp 1–9. https://doi.org/10.1109/CVPR.2015.7298594. Sanagala SS et al (2021) Ten fast transfer learning models for carotid ultrasound plaque tissue characterization in augmentation framework embedded with heatmaps for stroke risk stratification. Diagnostics 11(11). https://doi.org/10.3390/diagnostics11112109. Nayak T et al (2023) Deep learning based detection of monkeypox virus using skin lesion images. Med Nov Technol Devices 18(May):100243. https://doi.org/10.1016/j.medntd.2023.100243. (PMID: 10.1016/j.medntd.2023.1002433729313410236906) Gifani P, Shalbaf A, Vafaeezadeh M (2021) Automated detection of COVID-19 using ensemble of transfer learning with deep convolutional neural network based on CT scans. Int J Comput Assist Radiol Surg 16(1):115–123. https://doi.org/10.1007/s11548-020-02286-w. (PMID: 10.1007/s11548-020-02286-w33191476) Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D (2020) Grad-CAM: visual explanations from deep networks via gradient-based localization. Int J Comput Vis 128(2):336–359. https://doi.org/10.1007/s11263-019-01228-7. (PMID: 10.1007/s11263-019-01228-7) Ribeiro MT, Singh S, Guestrin C (2016) Why should i trust you?’ Explaining the predictions of any classifier. In: Proceedings of the ACM SIGKDD international conference on knowledge discovery and data mining, vol 13–17-Augu, pp 1135–1144. https://doi.org/10.1145/2939672.2939778. Blanco P, Abdo-Cuza A (2018) Transcranial Doppler ultrasound in neurocritical care. J Ultrasound 21(1):1–16. https://doi.org/10.1007/s40477-018-0282-9. (PMID: 10.1007/s40477-018-0282-9294290155845939) Ottakath N et al (2023) Ultrasound-based image analysis for predicting carotid artery stenosis risk: a comprehensive review of the problem, techniques, datasets, and future directions. Diagnostics 13(15):33. https://doi.org/10.3390/diagnostics13152614. (PMID: 10.3390/diagnostics13152614) |
| Contributed Indexing: | Keywords: Artery identification; Carotid Doppler ultrasound; Explainable AI; Spectral Doppler; Transfer learning |
| Entry Date(s): | Date Created: 20260414 Date Completed: 20260627 Latest Revision: 20260726 |
| Update Code: | 20260726 |
| PubMed Central ID: | PMC13269549 |
| DOI: | 10.1007/s11517-026-03569-1 |
| PMID: | 41979720 |
| Βάση Δεδομένων: | MEDLINE |
καταχωρήστε σχόλιο πρώτοι!