{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T21:59:34Z","timestamp":1790805574830,"version":"4.1.0"},"reference-count":51,"publisher":"Wiley","issue":"5","license":[{"start":{"date-parts":[[2022,3,3]],"date-time":"2022-03-03T00:00:00Z","timestamp":1646265600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2022,5]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Background<\/jats:title><jats:p>Early diagnosis of lung cancer is a key intervention for the treatment of lung cancer in which computer\u2010aided diagnosis (CAD) can play a crucial role. Most published CAD methods perform lung cancer diagnosis by classifying each lung nodule in isolation. However, this does not reflect clinical practice, where clinicians diagnose a patient based on a set of images of nodules, instead of looking at one nodule at a time. Besides, the low interpretability of the output provided by these methods presents an important barrier for their adoption.<\/jats:p><\/jats:sec><jats:sec><jats:title>Method<\/jats:title><jats:p>In this article, we treat lung cancer diagnosis as a multiple instance learning (MIL) problem, which better reflects the diagnosis process in the clinical setting and provides higher interpretability of the output. We selected radiomics as the source of input features and deep attention\u2010based MIL as the classification algorithm. The attention mechanism provides higher interpretability by estimating the importance of each instance in the set for the final diagnosis. To improve the model's performance in a small imbalanced dataset, we propose a new bag simulation method for MIL.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results and conclusion<\/jats:title><jats:p>The results show that our method can achieve a mean accuracy of 0.807 with a standard error of the mean (SEM) of 0.069, a recall of 0.870 (SEM 0.061), a positive predictive value of 0.928 (SEM 0.078), a negative predictive value of  (SEM 0.155), and an area under the curve (AUC) of 0.842 (SEM 0.074), outperforming other MIL methods. Additional experiments show that the proposed oversampling strategy significantly improves the model's performance. In addition, experiments show that our method provides a good indication of the importance of each nodule in determining the diagnosis, which combined with the well\u2010defined radiomic features, to make the results more interpretable and acceptable for doctors and patients.<\/jats:p><\/jats:sec>","DOI":"10.1002\/mp.15539","type":"journal-article","created":{"date-parts":[[2022,2,21]],"date-time":"2022-02-21T04:07:51Z","timestamp":1645416471000},"page":"3134-3143","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":38,"title":["Lung cancer diagnosis using deep attention\u2010based multiple instance learning and radiomics"],"prefix":"10.1002","volume":"49","author":[{"given":"Junhua","family":"Chen","sequence":"first","affiliation":[{"name":"Department of Radiation Oncology (MAASTRO) GROW School for Oncology and Developmental Biology Maastricht University Medical Centre+  Maastricht The Netherlands"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Haiyan","family":"Zeng","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology (MAASTRO) GROW School for Oncology and Developmental Biology Maastricht University Medical Centre+  Maastricht The Netherlands"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chong","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology (MAASTRO) GROW School for Oncology and Developmental Biology Maastricht University Medical Centre+  Maastricht The Netherlands"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhenwei","family":"Shi","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology (MAASTRO) GROW School for Oncology and Developmental Biology Maastricht University Medical Centre+  Maastricht The Netherlands"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andre","family":"Dekker","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology (MAASTRO) GROW School for Oncology and Developmental Biology Maastricht University Medical Centre+  Maastricht The Netherlands"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Leonard","family":"Wee","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology (MAASTRO) GROW School for Oncology and Developmental Biology Maastricht University Medical Centre+  Maastricht The Netherlands"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Inigo","family":"Bermejo","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology (MAASTRO) GROW School for Oncology and Developmental Biology Maastricht University Medical Centre+  Maastricht The Netherlands"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2022,3,3]]},"reference":[{"key":"e_1_2_8_2_1","unstructured":"Cancer. [WWW document]. URLhttps:\/\/www.who.int\/news\u2010room\/fact\u2010sheets\/detail\/cancer[access on 20 June 2020] 2018."},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.3322\/caac.21590"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1056\/NEJMoa1102873"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0895-6111(98)00017-2"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00521-018-3773-x"},{"key":"e_1_2_8_7_1","doi-asserted-by":"publisher","DOI":"10.24297\/ijct.v3i3a.2944"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.21037\/jtd.2018.02.57"},{"key":"e_1_2_8_9_1","first-page":"1097","volume-title":"Advances in Neural Information Processing Systems","author":"Krizhevsky A","year":"2012"},{"key":"e_1_2_8_10_1","doi-asserted-by":"crossref","unstructured":"ShenW ZhouM YangF YangC TianJ.Multi\u2010scale convolutional neural networks for lung nodule classification. In:International Conference on Information Processing in Medical Imaging.Springer;2015:588\u2010599.","DOI":"10.1007\/978-3-319-19992-4_46"},{"key":"e_1_2_8_11_1","unstructured":"ZhouB KhoslaA LapedrizaA OlivaA TorralbaA.Learning deep features for discriminative localization. In:Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:2921\u20102929."},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2017.10.005"},{"issue":"8","key":"e_1_2_8_13_1","first-page":"409","article-title":"Lung cancer detection and classification with 3D convolutional neural network (3D\u2010CNN)","volume":"8","author":"Alakwaa W","year":"2017","journal-title":"Lung Cancer"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2017.06.015"},{"key":"e_1_2_8_15_1","doi-asserted-by":"publisher","DOI":"10.1056\/NEJMoa1312547"},{"key":"e_1_2_8_16_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-019-0447-x"},{"key":"e_1_2_8_17_1","unstructured":"SelbstA PowlesJ. \u201cMeaningful information\u201d and the right to explanation. In:Conference on Fairness Accountability and Transparency. PMLR;2018:48\u201048."},{"key":"e_1_2_8_18_1","unstructured":"Data protection in the EU. [WWW document]. URLhttps:\/\/ec.europa.eu\/info\/law\/law\u2010topic\/data\u2010protection\/data\u2010protection\u2010eu_en[access on 5 July 2020] 2020."},{"key":"e_1_2_8_19_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0004-3702(96)00034-3"},{"key":"e_1_2_8_20_1","first-page":"570","volume-title":"Advances in Neural Information Processing Systems","author":"Maron O","year":"1998"},{"key":"e_1_2_8_21_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patrec.2020.02.025"},{"key":"e_1_2_8_22_1","unstructured":"IlseM TomczakJM WellingM.Attention\u2010based deep multiple instance learning. In35th International Conference on Machine Learning ICML 2018. International Machine Learning Society (IMLS);2018:3376\u20103391."},{"key":"e_1_2_8_23_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2017.08.026"},{"key":"e_1_2_8_24_1","first-page":"577","volume-title":"Advances in Neural Information Processing Systems","author":"Andrews S","year":"2003"},{"key":"e_1_2_8_25_1","doi-asserted-by":"crossref","unstructured":"ZhouZH SunYY LiYF.Multi\u2010instance learning by treating instances as non\u2010iid samples. In:Proceedings of the 26th Annual International Conference on Machine Learning;2009:1249\u20101256.","DOI":"10.1145\/1553374.1553534"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2016.2519102"},{"key":"e_1_2_8_27_1","first-page":"3104","volume-title":"Advances in Neural Information Processing Systems","author":"Sutskever I","year":"2014"},{"key":"e_1_2_8_28_1","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2019.2900993"},{"issue":"1","key":"e_1_2_8_29_1","first-page":"1","article-title":"Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach","volume":"5","author":"Aerts HJ","year":"2014","journal-title":"Nat Commun"},{"key":"e_1_2_8_30_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejca.2011.11.036"},{"key":"e_1_2_8_31_1","doi-asserted-by":"publisher","DOI":"10.1109\/TII.2014.2349359"},{"key":"e_1_2_8_32_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.3528204"},{"key":"e_1_2_8_33_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10278-013-9622-7"},{"key":"e_1_2_8_34_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2020.2996256"},{"key":"e_1_2_8_35_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10278-014-9718-8"},{"key":"e_1_2_8_36_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10278-015-9857-6"},{"key":"e_1_2_8_37_1","doi-asserted-by":"publisher","DOI":"10.1158\/0008-5472.CAN-17-0339"},{"key":"e_1_2_8_38_1","doi-asserted-by":"publisher","DOI":"10.1117\/1.JMI.3.4.044504"},{"key":"e_1_2_8_39_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10278-017-0037-8"},{"key":"e_1_2_8_40_1","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"e_1_2_8_41_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2008.239"},{"key":"e_1_2_8_42_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2016.12.035"},{"key":"e_1_2_8_43_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-35289-8_25"},{"key":"e_1_2_8_44_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2005.50"},{"key":"e_1_2_8_45_1","unstructured":"Learning and Mining from DatA. [WWW document]. URLhttp:\/\/210.28.132.67\/Data.ashx. [accesses on 4 Jun 2021] 2021."},{"key":"e_1_2_8_46_1","doi-asserted-by":"crossref","unstructured":"YongH HuangJ MengD HuaX ZhangL.Momentum batch normalization for deep learning with small batch size. In:European Conference on Computer Vision.Springer;2020:224\u2010240.","DOI":"10.1007\/978-3-030-58610-2_14"},{"key":"e_1_2_8_47_1","doi-asserted-by":"crossref","unstructured":"HuangX ShanJ VaidyaV.Lung nodule detection in CT using 3D convolutional neural networks. In:2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017).IEEE; 2017:379\u2010383.","DOI":"10.1109\/ISBI.2017.7950542"},{"key":"e_1_2_8_48_1","first-page":"978532","volume-title":"Medical Imaging 2016: Computer\u2010Aided Diagnosis","author":"Anirudh R","year":"2016"},{"key":"e_1_2_8_49_1","doi-asserted-by":"publisher","DOI":"10.1097\/EDE.0b013e3181c30fb2"},{"key":"e_1_2_8_50_1","doi-asserted-by":"publisher","DOI":"10.1093\/eurheartj\/ehu207"},{"key":"e_1_2_8_51_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12188"},{"key":"e_1_2_8_52_1","doi-asserted-by":"publisher","DOI":"10.7326\/M14-0698"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.15539","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.15539","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.15539","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,19]],"date-time":"2023-08-19T15:01:03Z","timestamp":1692457263000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.15539"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,3,3]]},"references-count":51,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2022,5]]}},"alternative-id":["10.1002\/mp.15539"],"URL":"https:\/\/doi.org\/10.1002\/mp.15539","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,3,3]]},"assertion":[{"value":"2021-06-14","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2022-02-07","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2022-03-03","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}