{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,6]],"date-time":"2026-11-06T05:54:24Z","timestamp":1793944464051,"version":"4.8.0"},"reference-count":29,"publisher":"Wiley","issue":"4","license":[{"start":{"date-parts":[[2024,7,18]],"date-time":"2024-07-18T00:00:00Z","timestamp":1721260800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001691","name":"Japan Society for the Promotion of Science","doi-asserted-by":"publisher","award":["JP21K19462"],"award-info":[{"award-number":["JP21K19462"]}],"id":[{"id":"10.13039\/501100001691","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Br J Haematol"],"published-print":{"date-parts":[[2024,10]]},"abstract":"<jats:title>Summary<\/jats:title><jats:p>Palpebral conjunctival hue alteration is used in non\u2010invasive screening for anaemia, whereas it is a qualitative measure. This study constructed machine\/deep learning models for predicting haemoglobin values using 150 palpebral conjunctival images taken by a smartphone. The median haemoglobin value was 13.1\u2009g\/dL, including 10 patients with &lt;11\u2009g\/dL. A segmentation model using U\u2010net was successfully constructed. The segmented images were subjected to non\u2010convolutional neural network (CNN)\u2010based and CNN\u2010based regression models for predicting haemoglobin values. The correlation coefficients between the actual and predicted haemoglobin values were 0.38 and 0.44 in the non\u2010CNN\u2010based and CNN\u2010based models, respectively. The sensitivity and specificity for anaemia detection were 13% and 98% for the non\u2010CNN\u2010based model and 20% and 99% for the CNN\u2010based model. The performance of the CNN\u2010based model did not improve with a mask layer guiding the model's attention towards the conjunctival regions, however, slightly improved with correction by the aspect ratio and exposure time of input images. The gradient\u2010weighted class activation mapping heatmap indicated that the lower half area of the conjunctiva was crucial for haemoglobin value prediction. In conclusion, the CNN\u2010based model had better results than the non\u2010CNN\u2010based model. The prediction accuracy would improve by using more input data with anaemia.<\/jats:p>","DOI":"10.1111\/bjh.19621","type":"journal-article","created":{"date-parts":[[2024,7,18]],"date-time":"2024-07-18T12:45:02Z","timestamp":1721306702000},"page":"1590-1598","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":22,"title":["Machine\/deep learning\u2010assisted hemoglobin level prediction using palpebral conjunctival images"],"prefix":"10.1111","volume":"205","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3656-2665","authenticated-orcid":false,"given":"Shota","family":"Kato","sequence":"first","affiliation":[{"name":"Department of Pediatrics, Graduate School of Medicine The University of Tokyo  Tokyo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Keita","family":"Chagi","sequence":"additional","affiliation":[{"name":"LPIXEL Inc.  Tokyo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yusuke","family":"Takagi","sequence":"additional","affiliation":[{"name":"LPIXEL Inc.  Tokyo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Moe","family":"Hidaka","sequence":"additional","affiliation":[{"name":"Department of Pediatrics, Graduate School of Medicine The University of Tokyo  Tokyo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shutaro","family":"Inoue","sequence":"additional","affiliation":[{"name":"Department of Pediatrics, Graduate School of Medicine The University of Tokyo  Tokyo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Masahiro","family":"Sekiguchi","sequence":"additional","affiliation":[{"name":"Department of Pediatrics, Graduate School of Medicine The University of Tokyo  Tokyo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Natsuho","family":"Adachi","sequence":"additional","affiliation":[{"name":"Department of Pediatrics, Graduate School of Medicine The University of Tokyo  Tokyo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kaname","family":"Sato","sequence":"additional","affiliation":[{"name":"Department of Pediatrics, Graduate School of Medicine The University of Tokyo  Tokyo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hiroki","family":"Kawai","sequence":"additional","affiliation":[{"name":"LPIXEL Inc.  Tokyo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5145-1774","authenticated-orcid":false,"given":"Motohiro","family":"Kato","sequence":"additional","affiliation":[{"name":"Department of Pediatrics, Graduate School of Medicine The University of Tokyo  Tokyo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2024,7,18]]},"reference":[{"key":"e_1_2_14_2_1","doi-asserted-by":"publisher","DOI":"10.1080\/17474086.2017.1354696"},{"key":"e_1_2_14_3_1","unstructured":"Worldwide prevalence of anaemia 1993\u20132005.WHO Global Database on Anaemia."},{"key":"e_1_2_14_4_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.hoc.2015.11.002"},{"key":"e_1_2_14_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jemermed.2007.02.011"},{"key":"e_1_2_14_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2016.08.025"},{"key":"e_1_2_14_7_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0153286"},{"key":"e_1_2_14_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2018.2867110"},{"key":"e_1_2_14_9_1","doi-asserted-by":"publisher","DOI":"10.1002\/ima.22359"},{"key":"e_1_2_14_10_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0253495"},{"key":"e_1_2_14_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/JSEN.2020.3044386"},{"key":"e_1_2_14_12_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.medntd.2023.100269"},{"key":"e_1_2_14_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.medntd.2023.100237"},{"key":"e_1_2_14_14_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0281736"},{"key":"e_1_2_14_15_1","doi-asserted-by":"publisher","DOI":"10.1002\/eng2.12667"},{"key":"e_1_2_14_16_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-017-17204-5"},{"key":"e_1_2_14_17_1","doi-asserted-by":"publisher","DOI":"10.1109\/LGRS.2018.2802944"},{"key":"e_1_2_14_18_1","doi-asserted-by":"crossref","unstructured":"RonnebergerO FischerP BroxT.U\u2010net: convolutional networks for biomedical image segmentation. Medical image computing and computer\u2010assisted intervention 2015: 18th international conference Munich Germany.2015. pp.234\u2013241.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_2_14_19_1","first-page":"6105","article-title":"EfficientNet: rethinking model scaling for convolutional","volume":"97","author":"Tan M","year":"2019","journal-title":"Neural Netw"},{"key":"e_1_2_14_20_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.324"},{"issue":"2","key":"e_1_2_14_21_1","first-page":"125","article-title":"Albumentations: fast and flexible image augmentations","volume":"11","author":"Buslaev A","year":"2020","journal-title":"Inf Dent"},{"key":"e_1_2_14_22_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"e_1_2_14_23_1","unstructured":"LiuL JiangH HeP ChenW LiuX GaoJ et\u00a0al.On the variance of the adaptive learning rate and beyond.arXiv:1908.03265.2019."},{"key":"e_1_2_14_24_1","unstructured":"LoshchilovI HutterF.SGDR: stochastic gradient descent with warm restarts.arXiv:1608.03983.\u00a02016."},{"key":"e_1_2_14_25_1","first-page":"2825","article-title":"Scikit\u2010learn: machine learning in python","volume":"12","author":"Pedregosa F","year":"2011","journal-title":"J Machine Learning Research"},{"key":"e_1_2_14_26_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2018.2846566"},{"key":"e_1_2_14_27_1","unstructured":"LoshchilovI HutterF.Decoupled weight decay regularization.arXiv:1711.05101.2017."},{"key":"e_1_2_14_28_1","first-page":"618","article-title":"Grad\u2010CAM: visual explanations from deep networks via gradient\u2010based localization","volume":"2017","author":"Selvaraju RR","year":"2017","journal-title":"IEEE Int Conf Comput Vis (ICCV)"},{"key":"e_1_2_14_29_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.dsp.2005.08.003"},{"key":"e_1_2_14_30_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2019.12.015"}],"container-title":["British Journal of Haematology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/bjh.19621","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,17]],"date-time":"2025-09-17T08:33:22Z","timestamp":1758098002000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1111\/bjh.19621"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,7,18]]},"references-count":29,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2024,10]]}},"alternative-id":["10.1111\/bjh.19621"],"URL":"https:\/\/doi.org\/10.1111\/bjh.19621","archive":["Portico"],"relation":{},"ISSN":["0007-1048","1365-2141"],"issn-type":[{"value":"0007-1048","type":"print"},{"value":"1365-2141","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,7,18]]},"assertion":[{"value":"2024-03-28","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-06-19","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-07-18","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}