{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,4]],"date-time":"2025-11-04T13:38:18Z","timestamp":1762263498889,"version":"build-2065373602"},"reference-count":54,"publisher":"MDPI AG","issue":"4","license":[{"start":{"date-parts":[[2025,11,4]],"date-time":"2025-11-04T00:00:00Z","timestamp":1762214400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Health and Performance Wellbeing Fund"},{"name":"QR Pump Priming Fund"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["MAKE"],"abstract":"<jats:p>Hyperbilirubinemia, commonly known as jaundice, is a prevalent condition in newborns, primarily arising from alterations in red blood cell metabolism during the first week of life. While conventional diagnostic methods, such as serum analysis and transcutaneous bilirubinometry, are effective, there remains a critical need for robust, non-invasive, image-based diagnostic tools. In this study, we propose a custom-designed convolutional neural network for classifying jaundice in neonatal images. Image preprocessing and segmentation techniques were systematically evaluated. The optimal workflow, which incorporated contrast enhancement and the extraction of regular skin patches of 144 \u00d7 144 pixels from regions of interest segmented using the Segment Anything Model, achieved a testing F1-score of 0.80. Beyond performance, this study addresses numerous shortcomings in the existing literature in this area relating to trust, replicability, and transparency. To this end, we employ fair performance metrics that are more robust to class imbalance, a transparent workflow, share source code, and use Gradient-weighted Class Activation Mapping to visualise and quantify the image regions that influence the classifier\u2019s predictions in pursuit of epistemic justification.<\/jats:p>","DOI":"10.3390\/make7040136","type":"journal-article","created":{"date-parts":[[2025,11,4]],"date-time":"2025-11-04T12:13:08Z","timestamp":1762258388000},"page":"136","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Explainable Deep Learning for Neonatal Jaundice Classification Using Uncalibrated Smartphone Images"],"prefix":"10.3390","volume":"7","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1578-5539","authenticated-orcid":false,"given":"Ashim","family":"Chakraborty","sequence":"first","affiliation":[{"name":"School of Computing and Information Science, Anglia Ruskin University, Cambridge CB1 1PT, UK"}]},{"given":"Yeshwanth","family":"Thota","sequence":"additional","affiliation":[{"name":"School of Computing and Information Science, Anglia Ruskin University, Cambridge CB1 1PT, UK"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4706-324X","authenticated-orcid":false,"given":"Cristina","family":"Luca","sequence":"additional","affiliation":[{"name":"School of Computing and Information Science, Anglia Ruskin University, Cambridge CB1 1PT, UK"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8131-5906","authenticated-orcid":false,"given":"Ian","family":"van der Linde","sequence":"additional","affiliation":[{"name":"School of Computing and Information Science, Anglia Ruskin University, Cambridge CB1 1PT, UK"},{"name":"Vision and Eye Research Institute (VERI), School of Medicine, Anglia Ruskin University, East Road, Cambridge CB1 1PT, UK"}]}],"member":"1968","published-online":{"date-parts":[[2025,11,4]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Hazarika, C.J., Borah, A., Gogoi, P., Ramchiary, S.S., Daurai, B., Gogoi, M., and Saikia, M.J. (2024). Development of non-invasive biosensors for neonatal jaundice detection: A review. Biosensors, 14.","DOI":"10.3390\/bios14050254"},{"key":"ref_2","unstructured":"Ansong-Assoku, B., Shah, S., Adnan, M., and Ankola, P. (2024). Neonatal jaundice. StatPearls, StatPearls Publishing."},{"key":"ref_3","unstructured":"(2025, July 02). NHS Jaundice in Newborns, NHS, 2023. Available online: https:\/\/www.nhs.uk\/conditions\/jaundice-newborn\/."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"699","DOI":"10.12968\/hmed.2017.78.12.699","article-title":"Neonatal jaundice: Aetiology, diagnosis and treatment","volume":"78","author":"Mitra","year":"2017","journal-title":"Br. J. Hosp. Med."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1151","DOI":"10.1016\/S0140-6736(17)32152-9","article-title":"Global, regional, and national age-sex specific mortality for 264 causes of death, 1980\u20132016: A systematic analysis for the Global Burden of Disease Study 2016","volume":"390","author":"Naghavi","year":"2017","journal-title":"Lancet"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"610","DOI":"10.1016\/S2352-4642(18)30139-1","article-title":"Neonatal hyperbilirubinaemia: A global perspective","volume":"2","author":"Olusanya","year":"2018","journal-title":"Lancet Child Adolesc. Health"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Olusanya, B.O., Teeple, S., and Kassebaum, N.J. (2018). The contribution of neonatal jaundice to global child mortality: Findings from the GBD 2016 Study. Pediatrics, 141.","DOI":"10.1542\/peds.2017-1471"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Ayalew, T., Molla, A., Kefale, B., Alene, T.D., Abebe, G.K., Ngusie, H.S., and Zemariam, A.B. (2024). Factors associated with neonatal jaundice among neonates admitted at referral hospitals in northeast Ethiopia: A facility-based unmatched case-control study. BMC Pregnancy Childbirth, 24.","DOI":"10.1186\/s12884-024-06352-y"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Shapiro, A., Anderson, J., Bond, M., Schwarz, R., Carns, J., Mtenthaonga, P., Kumwenda, W., Johnston, R., Miros, R., and Dube, Q. (2021). A low-cost bilirubin measurement tool for neonatal jaundice monitoring at the point-of-care: A comparison of BiliDx with a standard laboratory bilirubinometer and transcutaneous bilirubinometer. Lancet Glob. Health, 9.","DOI":"10.1016\/S2214-109X(21)00131-5"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"69","DOI":"10.1016\/j.clinbiochem.2018.03.018","article-title":"The accuracy of transcutaneous bilirubinometer measurements to identify hyperbilirubinemia in an outpatient newborn population","volume":"55","author":"Ercan","year":"2018","journal-title":"Clin. Biochem."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Okwundu, C.I., Olowoyeye, A., Uthman, O.A., Smith, J., Wiysonge, C.S., Bhutani, V.K., Fiander, M., and Gautham, K.S. (2023). Transcutaneous bilirubinometry versus total serum bilirubin measurement for newborns. Cochrane Database Syst. Rev., 5.","DOI":"10.1002\/14651858.CD012660.pub2"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Kirillov, A., Mintun, E., Ravi, N., Mao, H., Silvester, C., Doll\u00e1r, P., and He, K. (2023, January 2\u20133). Segment Anything. Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV), Paris, France.","DOI":"10.1109\/ICCV51070.2023.00371"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Selvaraju, R.R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., and Batra, D. (2017, January 22\u201329). Grad-cam: Visual explanations from deep networks via gradient-based localization. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.","DOI":"10.1109\/ICCV.2017.74"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"2532","DOI":"10.1111\/apa.15287","article-title":"Bilirubin estimates from smartphone images of newborn infants\u2019 skin correlated highly to serum bilirubin levels","volume":"109","author":"Aune","year":"2020","journal-title":"Acta Paediatr."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Munkholm, S.B., Kr\u00f8gholt, T., Ebbesen, F., Szecsi, P.B., and Kristensen, S.R. (2018). The smartphone camera as a potential method for transcutaneous bilirubin measurement. PLoS ONE, 13.","DOI":"10.1371\/journal.pone.0197938"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Outlaw, F., Nixon, M., Odeyemi, O., MacDonald, L.W., Meek, J., and Leung, T.S. (2020). Smartphone screening for neonatal jaundice via ambient-subtracted sclera chromaticity. PLoS ONE, 15.","DOI":"10.1371\/journal.pone.0216970"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"2116","DOI":"10.3906\/elk-2008-76","article-title":"Classification of neonatal jaundice in mobile application with noninvasive imageprocessing methods","volume":"29","author":"Aydin","year":"2021","journal-title":"Turk. J. Electr. Eng. Comput. Sci."},{"key":"ref_18","first-page":"599","article-title":"Hyperbilirubinemia in the term newborn","volume":"65","author":"Porter","year":"2002","journal-title":"Am. Fam. Physician"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"De Greef, L., Goel, M., Seo, M.J., Larson, E.C., Stout, J.W., Taylor, J.A., and Patel, S.N. (2014, January 13\u201317). BiliCam: Using mobile phones to monitor newborn jaundice. Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing, Washington, DC, USA.","DOI":"10.1145\/2632048.2632076"},{"key":"ref_20","first-page":"271","article-title":"A 3 \u00d7 3 isotropic gradient operator for image processing","volume":"1968","author":"Sobel","year":"1968","journal-title":"Talk Stanf. Artif. Proj."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Taylor, J.A., Stout, J.W., de Greef, L., Goel, M., Patel, S., Chung, E.K., Koduri, A., McMahon, S., Dickerson, J., and Simpson, E.A. (2017). Use of a smartphone app to assess neonatal jaundice. Pediatrics, 140.","DOI":"10.1542\/peds.2017-0312"},{"key":"ref_22","first-page":"597","article-title":"Evaluation of an automatic image-based screening technique for neonatal hyperbilirubinemia","volume":"54","author":"Rong","year":"2016","journal-title":"Chin. J. Pediatr."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"93","DOI":"10.3126\/jnps.v40i3.29412","article-title":"Validity of bilirubin measured by Biliscan (smartphone application) in neonatal jaundice: An observational study","volume":"41","author":"Lingaldinna","year":"2021","journal-title":"J. Nepal Paediatr. Soc."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Ngeow, A.J.H., Moosa, A.S., Tan, M.G., Zou, L., Goh, M.M.R., Lim, G.H., Tagamolila, V., Ereno, I., Durnford, J.R., and Cheung, S.K.H. (2024). Development and Validation of a Smartphone Application for Neonatal Jaundice Screening. JAMA Netw. Open, 7.","DOI":"10.1001\/jamanetworkopen.2024.50260"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Ayd\u0131n, M., Hardala\u00e7, F., Ural, B., and Karap, S. (2016). Neonatal jaundice detection system. J. Med. Syst., 40.","DOI":"10.1007\/s10916-016-0523-4"},{"key":"ref_26","unstructured":"Juliastuti, E., Nadhira, V., Satwika, Y.W., Aziz, N.A., and Zahra, N. (August, January 31). Risk zone estimation of newborn jaundice based on skin color image analysis. Proceedings of the 6th International Conference on Instrumentation, Control, and Automation (ICA), Bandung, Indonesia."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Hakimi Abu Bakar, A., Hassan, N.M., Zakaria, A., Halim, K.A.A., and Halim, A.A.A. (2017, January 10\u201312). Jaundice (hyperbilirubinemia) detection and prediction system using color card technique. Proceedings of the IEEE 13th International Colloquium on Signal Processing & Its Applications (CSPA), Penang, Malaysia.","DOI":"10.1109\/CSPA.2017.8064952"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"869","DOI":"10.1001\/archderm.1988.01670060015008","article-title":"The validity and practicality of sun-reactive skin types I through VI","volume":"124","author":"Fitzpatrick","year":"1988","journal-title":"Arch. Dermatol."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Hashim, W., Al-Naji, A., Al-Rayahi, I.A., and Oudah, M. (2021). Computer vision for jaundice detection in neonates using graphic user interface. IOP Conf. Ser. Mater. Sci. Eng., 1105.","DOI":"10.1088\/1757-899X\/1105\/1\/012076"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Hashim, W., Al-Naji, A., Al-Rayahi, I.A., Alkhaled, M., and Chahl, J. (2021). Neonatal jaundice detection using a computer vision system. Designs, 5.","DOI":"10.3390\/designs5040063"},{"key":"ref_31","first-page":"10","article-title":"A mobile computer-aided diagnosis of neonatal hyperbilirubinemia using digital image processing and machine learning techniques","volume":"5","author":"Dissaneevate","year":"2022","journal-title":"Int. J. Innov. Res. Sci. Stud."},{"key":"ref_32","first-page":"25","article-title":"PCA-based feature extraction and k-NN algorithm for early jaundice detection","volume":"1","author":"Mansor","year":"2011","journal-title":"Int. J. Soft Comput. Softw. Eng."},{"key":"ref_33","first-page":"431","article-title":"Detection of jaundice in neonates using artificial intelligence","volume":"Volume 2","author":"Nihila","year":"2021","journal-title":"Soft Computing: Theories and Applications: Proceedings of SoCTA 2020"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"258","DOI":"10.1016\/j.neucom.2012.10.034","article-title":"New newborn jaundice monitoring scheme based on combination of pre-processing and color detection method","volume":"120","author":"Mansor","year":"2013","journal-title":"Neurocomputing"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Nayagi, B.S., and Angel, S.T. (2023). Diagnosis of neonatal hyperbilirubinemia using CNN model along with color card techniques. Biomed. Signal Process. Control, 85.","DOI":"10.1016\/j.bspc.2023.104746"},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"387","DOI":"10.1007\/978-981-19-3679-1_30","article-title":"Deep learning approach for early diagnosis of jaundice","volume":"Volume 492","author":"Gupta","year":"2023","journal-title":"International Conference on Innovative Computing and Communications"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27\u201330). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"He, K., Gkioxari, G., Doll\u00e1r, P., and Girshick, R. (2017, January 22\u201329). Mask R-CNN. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.","DOI":"10.1109\/ICCV.2017.322"},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Althnian, A., Almanea, N., and Aloboud, N. (2021). Neonatal jaundice diagnosis using a smartphone camera based on eye, skin, and fused features with transfer learning. Sensors, 21.","DOI":"10.3390\/s21217038"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"217","DOI":"10.1016\/j.patrec.2020.03.004","article-title":"Adjusting the imbalance ratio by the dimensionality of imbalanced data","volume":"133","author":"Zhu","year":"2020","journal-title":"Pattern Recognit. Lett."},{"key":"ref_41","first-page":"197","article-title":"Neonatal jaundice detection using colour detection method","volume":"4","author":"Chowdhary","year":"2017","journal-title":"Int. Adv. Res. J. Sci. Eng. Technol."},{"key":"ref_42","first-page":"248","article-title":"Neonatal jaundice detection system using CNN algorithm and image processing","volume":"11","author":"Chakraborty","year":"2020","journal-title":"Int. J. Electr. Eng. Technol."},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Singla, R., and Singh, S. (2016, January 26\u201327). A framework for detection of jaundice in newborn babies using homomorphic filtering-based image processing. Proceedings of the 2016 IEEE International Conference on Inventive Computation Technologies (ICICT), Coimbatore, India.","DOI":"10.1109\/INVENTIVE.2016.7830209"},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"92","DOI":"10.35882\/jeeemi.v7i1.576","article-title":"Neonatal jaundice severity detection from skin images using deep transfer learning techniques","volume":"7","author":"Aldabbagh","year":"2024","journal-title":"J. Electron. Electromed. Eng. Med. Inform."},{"key":"ref_45","unstructured":"Simonyan, K., and Zisserman, A. (2015, January 7\u20139). Very deep convolutional networks for large-scale image recognition. Proceedings of the 3rd International Conference on Learning Representations (ICLR 2015), San Diego, CA, USA."},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"623","DOI":"10.3390\/biomedinformatics4010034","article-title":"Real-time jaundice detection in neonates based on machine learning models","volume":"4","author":"Abdulrazzak","year":"2024","journal-title":"BioMedInformatics"},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"543","DOI":"10.3390\/biomedinformatics3030037","article-title":"NJN: A Dataset for the Normal and Jaundiced Newborns","volume":"3","author":"Abdulrazzak","year":"2023","journal-title":"BioMedInformatics"},{"key":"ref_48","first-page":"35","article-title":"Smart screening: Non-invasive detection of severe neonatal jaundice using computer vision and deep learning","volume":"186","author":"Gupta","year":"2024","journal-title":"Int. J. Comput. Appl."},{"key":"ref_49","doi-asserted-by":"crossref","unstructured":"Makhloughi, F. (2025). Artificial intelligence-based non-invasive bilirubin prediction for neonatal jaundice using 1D convolutional neural network. Sci. Rep., 15.","DOI":"10.1038\/s41598-025-96100-9"},{"key":"ref_50","doi-asserted-by":"crossref","unstructured":"Salami, F.O., Muzammel, M., Mourchid, Y., and Othmani, A. (2025). Artificial intelligence non-invasive methods for neonatal jaundice detection: A review. Artif. Intell. Med., 162.","DOI":"10.1016\/j.artmed.2025.103088"},{"key":"ref_51","doi-asserted-by":"crossref","unstructured":"Zuiderveld, K. (1994). Contrast limited adaptive histogram equalization. Graphics Gems IV, Academic Press.","DOI":"10.1016\/B978-0-12-336156-1.50061-6"},{"key":"ref_52","unstructured":"Glorot, X., Bordes, A., and Bengio, Y. (2011, January 11\u201313). Deep sparse rectifier neural networks. Proceedings of the 14th International Conference on Artificial Intelligence and Statistics (AISTATS), Fort Lauderdale, FL, USA."},{"key":"ref_53","doi-asserted-by":"crossref","unstructured":"van der Linde, I. (2025). Why the confusion matrix fails as a model of knowledge. AI Soc.","DOI":"10.1007\/s00146-025-02456-x"},{"key":"ref_54","doi-asserted-by":"crossref","unstructured":"Abebe, M.A., Hardeberg, J.Y., and Vartdal, G. (2021). Smartphones\u2019 skin colour reproduction analysis for neonatal jaundice detection. J. Imaging Sci. Technol., 34.","DOI":"10.2352\/J.ImagingSci.Technol.2021.65.6.060407"}],"container-title":["Machine Learning and Knowledge Extraction"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2504-4990\/7\/4\/136\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,4]],"date-time":"2025-11-04T13:32:29Z","timestamp":1762263149000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2504-4990\/7\/4\/136"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,4]]},"references-count":54,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["make7040136"],"URL":"https:\/\/doi.org\/10.3390\/make7040136","relation":{},"ISSN":["2504-4990"],"issn-type":[{"type":"electronic","value":"2504-4990"}],"subject":[],"published":{"date-parts":[[2025,11,4]]}}}