{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,4]],"date-time":"2026-03-04T05:39:58Z","timestamp":1772602798141,"version":"3.50.1"},"reference-count":20,"publisher":"Springer Science and Business Media LLC","issue":"S5","license":[{"start":{"date-parts":[[2021,11,1]],"date-time":"2021-11-01T00:00:00Z","timestamp":1635724800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2021,11,8]],"date-time":"2021-11-08T00:00:00Z","timestamp":1636329600000},"content-version":"vor","delay-in-days":7,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100004663","name":"Ministry of Science and Technology, Taiwan","doi-asserted-by":"publisher","award":["MOST 109-2221-E-037-005"],"award-info":[{"award-number":["MOST 109-2221-E-037-005"]}],"id":[{"id":"10.13039\/501100004663","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004663","name":"Ministry of Science and Technology, Taiwan","doi-asserted-by":"publisher","award":["MOST 109-2221-E-153-005-MY3"],"award-info":[{"award-number":["MOST 109-2221-E-153-005-MY3"]}],"id":[{"id":"10.13039\/501100004663","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004663","name":"Ministry of Science and Technology, Taiwan","doi-asserted-by":"publisher","award":["MOST 109-2222-E-035-002"],"award-info":[{"award-number":["MOST 109-2222-E-035-002"]}],"id":[{"id":"10.13039\/501100004663","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Bioinformatics"],"published-print":{"date-parts":[[2021,11]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec>\n                <jats:title>Background<\/jats:title>\n                <jats:p>Heart sound measurement is crucial for analyzing and diagnosing patients with heart diseases. This study employed phonocardiogram signals as the input signal for heart disease analysis due to the accessibility of the respective method. This study referenced preprocessing techniques proposed by other researchers for the conversion of phonocardiogram signals into characteristic images composed using frequency subband. Image recognition was then conducted through the use of convolutional neural networks (CNNs), in order to classify the predicted of phonocardiogram signals as normal or abnormal. However, CNN requires the tuning of multiple hyperparameters, which entails an optimization problem for the hyperparameters in the model. To maximize CNN robustness, the uniform experiment design method and a science-based methodical experiment design were used to optimize CNN hyperparameters in this study.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Results<\/jats:title>\n                <jats:p>An artificial intelligence prediction model was constructed using CNN, and the uniform experiment design method was proposed to acquire hyperparameters for optimal CNN robustness. The results indicate Filters (<jats:inline-formula><jats:alternatives><jats:tex-math>$${X}_{1}$$<\/jats:tex-math><mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                    <mml:msub>\n                      <mml:mi>X<\/mml:mi>\n                      <mml:mn>1<\/mml:mn>\n                    <\/mml:msub>\n                  <\/mml:math><\/jats:alternatives><\/jats:inline-formula>), Stride (<jats:inline-formula><jats:alternatives><jats:tex-math>$${X}_{3}$$<\/jats:tex-math><mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                    <mml:msub>\n                      <mml:mi>X<\/mml:mi>\n                      <mml:mn>3<\/mml:mn>\n                    <\/mml:msub>\n                  <\/mml:math><\/jats:alternatives><\/jats:inline-formula>), Activation functions (<jats:inline-formula><jats:alternatives><jats:tex-math>$${X}_{6}$$<\/jats:tex-math><mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                    <mml:msub>\n                      <mml:mi>X<\/mml:mi>\n                      <mml:mn>6<\/mml:mn>\n                    <\/mml:msub>\n                  <\/mml:math><\/jats:alternatives><\/jats:inline-formula>), and Dropout (<jats:inline-formula><jats:alternatives><jats:tex-math>$${X}_{7}$$<\/jats:tex-math><mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                    <mml:msub>\n                      <mml:mi>X<\/mml:mi>\n                      <mml:mn>7<\/mml:mn>\n                    <\/mml:msub>\n                  <\/mml:math><\/jats:alternatives><\/jats:inline-formula>) to be significant factors considerably influencing the ability of CNN to distinguish among heart sound states. Finally, the confirmation experiment was conducted, and the hyperparameter combination for optimal model robustness was Filters (<jats:inline-formula><jats:alternatives><jats:tex-math>$${X}_{1}$$<\/jats:tex-math><mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                    <mml:msub>\n                      <mml:mi>X<\/mml:mi>\n                      <mml:mn>1<\/mml:mn>\n                    <\/mml:msub>\n                  <\/mml:math><\/jats:alternatives><\/jats:inline-formula>)\u2009=\u200932, Kernel Size (<jats:inline-formula><jats:alternatives><jats:tex-math>$${X}_{2})$$<\/jats:tex-math><mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                    <mml:mrow>\n                      <mml:msub>\n                        <mml:mi>X<\/mml:mi>\n                        <mml:mn>2<\/mml:mn>\n                      <\/mml:msub>\n                      <mml:mrow>\n                        <mml:mo>)<\/mml:mo>\n                      <\/mml:mrow>\n                    <\/mml:mrow>\n                  <\/mml:math><\/jats:alternatives><\/jats:inline-formula> = 3\u2009\u00d7\u20093, Stride (<jats:inline-formula><jats:alternatives><jats:tex-math>$${X}_{3}$$<\/jats:tex-math><mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                    <mml:msub>\n                      <mml:mi>X<\/mml:mi>\n                      <mml:mn>3<\/mml:mn>\n                    <\/mml:msub>\n                  <\/mml:math><\/jats:alternatives><\/jats:inline-formula>)\u2009=\u2009(1,1), Padding (<jats:inline-formula><jats:alternatives><jats:tex-math>$${X}_{4})$$<\/jats:tex-math><mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                    <mml:mrow>\n                      <mml:msub>\n                        <mml:mi>X<\/mml:mi>\n                        <mml:mn>4<\/mml:mn>\n                      <\/mml:msub>\n                      <mml:mrow>\n                        <mml:mo>)<\/mml:mo>\n                      <\/mml:mrow>\n                    <\/mml:mrow>\n                  <\/mml:math><\/jats:alternatives><\/jats:inline-formula> as same, Optimizer (<jats:inline-formula><jats:alternatives><jats:tex-math>$${X}_{5})$$<\/jats:tex-math><mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                    <mml:mrow>\n                      <mml:msub>\n                        <mml:mi>X<\/mml:mi>\n                        <mml:mn>5<\/mml:mn>\n                      <\/mml:msub>\n                      <mml:mrow>\n                        <mml:mo>)<\/mml:mo>\n                      <\/mml:mrow>\n                    <\/mml:mrow>\n                  <\/mml:math><\/jats:alternatives><\/jats:inline-formula> as the stochastic gradient descent, Activation functions (<jats:inline-formula><jats:alternatives><jats:tex-math>$${X}_{6}$$<\/jats:tex-math><mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                    <mml:msub>\n                      <mml:mi>X<\/mml:mi>\n                      <mml:mn>6<\/mml:mn>\n                    <\/mml:msub>\n                  <\/mml:math><\/jats:alternatives><\/jats:inline-formula>) as relu, and Dropout (<jats:inline-formula><jats:alternatives><jats:tex-math>$${X}_{7}$$<\/jats:tex-math><mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                    <mml:msub>\n                      <mml:mi>X<\/mml:mi>\n                      <mml:mn>7<\/mml:mn>\n                    <\/mml:msub>\n                  <\/mml:math><\/jats:alternatives><\/jats:inline-formula>)\u2009=\u20090.544. With this combination of parameters, the model had an average prediction accuracy rate of 0.787 and standard deviation of 0.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Conclusion<\/jats:title>\n                <jats:p>In this study, phonocardiogram signals were used for the early prediction of heart diseases. The science-based and methodical uniform experiment design was used for the optimization of CNN hyperparameters to construct a CNN with optimal robustness. The results revealed that the constructed model exhibited robustness and an acceptable accuracy rate. Other literature has failed to address hyperparameter optimization problems in CNN; a method is subsequently proposed for robust CNN optimization, thereby solving this problem.<\/jats:p>\n              <\/jats:sec>","DOI":"10.1186\/s12859-021-04032-8","type":"journal-article","created":{"date-parts":[[2021,11,8]],"date-time":"2021-11-08T15:05:43Z","timestamp":1636383943000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":6,"title":["Robust optimization of convolutional neural networks with a uniform experiment design method: a case of phonocardiogram testing in patients with heart diseases"],"prefix":"10.1186","volume":"22","author":[{"given":"Wen-Hsien","family":"Ho","sequence":"first","affiliation":[]},{"given":"Tian-Hsiang","family":"Huang","sequence":"additional","affiliation":[]},{"given":"Po-Yuan","family":"Yang","sequence":"additional","affiliation":[]},{"given":"Jyh-Horng","family":"Chou","sequence":"additional","affiliation":[]},{"given":"Jin-Yi","family":"Qu","sequence":"additional","affiliation":[]},{"given":"Po-Chih","family":"Chang","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4106-5084","authenticated-orcid":false,"given":"Fu-I.","family":"Chou","sequence":"additional","affiliation":[]},{"given":"Jinn-Tsong","family":"Tsai","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2021,11,8]]},"reference":[{"key":"4032_CR1","first-page":"345","volume":"35","author":"SE Schmidt","year":"2008","unstructured":"Schmidt SE, Toft E, Holst-Hansen C, Graff C, Struijk JJ. Segmentation of heart sound recordings from an electronic stethoscope by a duration dependent hidden-Markov model. Comput Cardiol. 2008;35:345\u20138.","journal-title":"Comput Cardiol"},{"key":"4032_CR2","first-page":"822","volume":"63","author":"DB Springer","year":"2016","unstructured":"Springer DB, Tarassenko L, Clifford GD. Logistic regression-HSMM-based heart sound segmentation. IEEE Trans Biomed Eng. 2016;63:822\u201332.","journal-title":"IEEE Trans Biomed Eng"},{"key":"4032_CR3","doi-asserted-by":"publisher","first-page":"1730","DOI":"10.1088\/1361-6579\/aa6e9f","volume":"38","author":"C Liu","year":"2017","unstructured":"Liu C, Springer D, Meyer J, Clifford GD. Performance of an open-source heart sound segmentation algorithm on eight independent databases. Physiol Meas. 2017;38:1730\u201345.","journal-title":"Physiol Meas"},{"key":"4032_CR4","doi-asserted-by":"publisher","first-page":"931","DOI":"10.1016\/S0026-2692(01)00095-7","volume":"32","author":"SR Messer","year":"2001","unstructured":"Messer SR, Agzarian J, Abbott D. Optimal wavelet denoising for phonocardiograms. Microelectron J. 2001;32:931\u201341.","journal-title":"Microelectron J"},{"key":"4032_CR5","doi-asserted-by":"publisher","first-page":"303","DOI":"10.1109\/TBME.2011.2163157","volume":"59","author":"CG Scully","year":"2012","unstructured":"Scully CG, Lee J, Meyer J, Gorbach AM, Granquist-Fraser D, Mendelson Y, Chon KH. Physiological parameter monitoring from optical recordings with a mobile phone. IEEE Trans Biomed Eng. 2012;59:303\u20136.","journal-title":"IEEE Trans Biomed Eng"},{"key":"4032_CR6","first-page":"1027","volume":"2","author":"J Joy","year":"2013","unstructured":"Joy J, Peter S, John N. Denoising using soft thresholding. Int J Adv Res Electr Electron Instrum Eng. 2013;2:1027\u201332.","journal-title":"Int J Adv Res Electr Electron Instrum Eng"},{"key":"4032_CR7","doi-asserted-by":"publisher","first-page":"1550009","DOI":"10.1142\/S0219519415500098","volume":"15","author":"K Zeng","year":"2015","unstructured":"Zeng K, Tan Z, Dong M. Using short-time Fourier transform and wavelet packet transform to attenuate noise from heart sound signal for wearable e-healthcare device. J Mech Med Biol. 2015;15:1550009.","journal-title":"J Mech Med Biol"},{"key":"4032_CR8","doi-asserted-by":"crossref","unstructured":"Potes C, Parvaneh S, Rahman A, Conroy B. Ensemble of feature-based and deep learning-based classifiers for detection of abnormal heart sounds. In: Computing in cardiology conference (CinC); 2016. p. 621\u20134.","DOI":"10.22489\/CinC.2016.182-399"},{"key":"4032_CR9","doi-asserted-by":"publisher","first-page":"132","DOI":"10.1016\/j.compbiomed.2018.06.026","volume":"100","author":"B Bozkurt","year":"2018","unstructured":"Bozkurt B, Germanakis I, Stylianou Y. A study of time-frequency features for CNN-based automatic heart sound classification for pathology detection. Comput Biol Med. 2018;100:132\u201343.","journal-title":"Comput Biol Med"},{"key":"4032_CR10","doi-asserted-by":"crossref","unstructured":"Zabihi M, Rad AB, Kiranyaz S, Gabbouj M, Katsaggelos AK. Heart sound anomaly and quality detection using ensemble of neural networks without segmentation. In: Computing in cardiology conference (CinC); 2016. p. 613\u20136.","DOI":"10.22489\/CinC.2016.180-213"},{"key":"4032_CR11","unstructured":"Gammatone Filterbank Toolkit. https:\/\/github.com\/detly\/gammatone. Accessed 15 Oct 2020."},{"key":"4032_CR12","first-page":"485","volume":"26","author":"Y Wang","year":"1981","unstructured":"Wang Y, Fang KT. A note on uniform distribution and experimental design. Chin Sci Bull. 1981;26:485\u20139.","journal-title":"Chin Sci Bull"},{"key":"4032_CR13","volume-title":"Uniform design and uniform layout","author":"KT Fang","year":"1994","unstructured":"Fang KT. Uniform design and uniform layout. Beijing: Science Press; 1994."},{"key":"4032_CR14","doi-asserted-by":"publisher","first-page":"299","DOI":"10.1090\/S0025-5718-98-00894-1","volume":"67","author":"FJ Hickernell","year":"1998","unstructured":"Hickernell FJ. A generalized discrepancy and quadrature error bound. Math Comput. 1998;67:299\u2013322.","journal-title":"Math Comput"},{"key":"4032_CR15","first-page":"144","volume":"6","author":"H Tsao","year":"2008","unstructured":"Tsao H, Lee L. Uniform layout implement using Matlab. Stat Decis. 2008;6:144\u20136.","journal-title":"Stat Decis"},{"key":"4032_CR16","doi-asserted-by":"publisher","first-page":"731","DOI":"10.1016\/j.apm.2017.09.030","volume":"53","author":"WH Ho","year":"2018","unstructured":"Ho WH, Chiu YH, Chen YJ. Multi-objective Pareto adaptive algorithm for capacitated lot-sizing problems in glass lens production. Appl Math Model. 2018;53:731\u20138.","journal-title":"Appl Math Model"},{"key":"4032_CR17","doi-asserted-by":"publisher","first-page":"52","DOI":"10.1016\/j.compeleceng.2018.03.019","volume":"67","author":"WH Tang","year":"2018","unstructured":"Tang WH, Chang YJ, Chen YJ, Ho WH. Genetic algorithm with Gaussian function for optimal P-wave morphology in electrocardiography for atrial fibrillation patients. Comput Electr Eng. 2018;67:52\u20137.","journal-title":"Comput Electr Eng"},{"key":"4032_CR18","doi-asserted-by":"publisher","first-page":"1033","DOI":"10.3233\/JIFS-169878","volume":"36","author":"YJ Chen","year":"2019","unstructured":"Chen YJ, Ho WH. Evolutionary algorithm in adaptive neuro-fuzzy inference system for modeling the growth of foodborne fungi. J Intell Fuzzy Syst. 2019;36:1033\u20139.","journal-title":"J Intell Fuzzy Syst"},{"key":"4032_CR19","doi-asserted-by":"publisher","DOI":"10.1002\/9781118950951","volume-title":"Data mining algorithms: explained using R","author":"P Cichosz","year":"2015","unstructured":"Cichosz P. Data mining algorithms: explained using R. New York: Wiley; 2015."},{"issue":"12","key":"4032_CR20","doi-asserted-by":"publisher","first-page":"2181","DOI":"10.1088\/0967-3334\/37\/12\/2181","volume":"37","author":"C Liu","year":"2016","unstructured":"Liu C, Springer D, Li Q, Moody B, Juan RA, Chorro FJ, Castells F, Roig JM, Silva I, Johnson AE, Syed Z, Schmidt SE, Papadaniil CD, Hadjileontiadis L, Naseri H, Moukadem A, Dieterlen A, Brandt C, Tang H, Samieinasab M, Samieinasab MR, Sameni R, Mark RG, Clifford GD. An open access database for the evaluation of heart sound algorithms. Physiol Meas. 2016;37(12):2181\u2013213.","journal-title":"Physiol Meas"}],"container-title":["BMC Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12859-021-04032-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12859-021-04032-8\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12859-021-04032-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,11,8]],"date-time":"2021-11-08T15:07:11Z","timestamp":1636384031000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmcbioinformatics.biomedcentral.com\/articles\/10.1186\/s12859-021-04032-8"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,11]]},"references-count":20,"journal-issue":{"issue":"S5","published-print":{"date-parts":[[2021,11]]}},"alternative-id":["4032"],"URL":"https:\/\/doi.org\/10.1186\/s12859-021-04032-8","relation":{},"ISSN":["1471-2105"],"issn-type":[{"value":"1471-2105","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,11]]},"assertion":[{"value":"15 February 2021","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 February 2021","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"8 November 2021","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"Not applicable.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare that they have no competing interests.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"92"}}