{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"medRxiv"}],"indexed":{"date-parts":[[2026,1,28]],"date-time":"2026-01-28T19:50:51Z","timestamp":1769629851484,"version":"3.49.0"},"posted":{"date-parts":[[2024,11,20]]},"group-title":"Health Informatics","reference-count":53,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2024,11,20]],"date-time":"2024-11-20T00:00:00Z","timestamp":1732060800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2024,11,20]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>Alzheimer\u2019s disease (AD) is a neurodegenerative disorder that requires early diagnosis for effective management. However, issues with currently available diagnostic biomarkers preclude early diagnosis, necessitating the development of alternative biomarkers and methods, such as blood-based diagnostics. We propose c-Triadem (constrained triple-input Alzheimer\u2019s disease model), a novel deep neural network to identify potential blood-based biomarkers for AD and predict mild cognitive impairment (MCI) and AD with high accuracy. The model utilizes genotyping data, gene expression data, and clinical information to predict the disease status of participants, i.e., cognitively normal (CN), MCI, or AD. The nodes of the neural network represent genes and their related pathways, and the edges represent known relationships among the genes and pathways. We trained the model with blood genotyping data, microarray, and clinical features from the Alzheimer\u2019s Neuroimaging Disease Initiative (ADNI). We demonstrate that our model\u2019s performance is superior to previous models with an AUC of 97% and accuracy of 89%. We then identified the most influential genes and clinical features for prediction using SHapley Additive exPlanations (SHAP). Our SHAP analysis shows that CASP9, LCK, and SDC3 SNPs and PINK1, ATG5, and ubiquitin (UBB, UBC) expression have a higher impact on model performance. Our model has facilitated the identification of potential blood-based genetic markers of DNA damage response and mitophagy in affected regions of the brain. The model can be used for detection and biomarker identification in other related dementias.<\/jats:p>\n                <jats:sec>\n                  <jats:title>Author Summary<\/jats:title>\n                  <jats:p>C-Triadem, our novel developed deep neural network, accurately predicts moderate cognitive impairment (MCI) and Alzheimer\u2019s disease (AD) while identifying potential blood biomarkers for AD. Current diagnostic methods have limitations, emphasizing the critical need for early AD detection. Our model integrates genetic, gene expression, and clinical data to differentiate among cognitively normal individuals, MCI, and AD cases. Training and validation using Alzheimer\u2019s Disease Neuroimaging Initiative (ADNI) data demonstrate superior performance, with a 97% AUC and 89% accuracy, surpassing previous models. SHapley Additive exPlanations (SHAP) analysis highlights key clinical features (e.g., MMSE scores, brain volume) and genes (e.g., CASP9, LCK, SDC3), revealing potential genetic markers and pathways in blood associated with AD. By incorporating Reactome pathways, our approach enhances interpretability, providing insight into the biological context of predictions. In summary, c-Triadem represents a significant advancement in AD diagnostics, enabling earlier and more accurate diagnoses for improved treatment strategies.<\/jats:p>\n                <\/jats:sec>","DOI":"10.1101\/2024.11.19.24317595","type":"posted-content","created":{"date-parts":[[2024,11,20]],"date-time":"2024-11-20T12:55:20Z","timestamp":1732107320000},"source":"Crossref","is-referenced-by-count":2,"title":["c-Triadem: A constrained, explainable deep learning model to identify novel biomarkers in Alzheimer\u2019s disease"],"prefix":"10.64898","author":[{"given":"Sherlyn","family":"Jemimah","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-5715-795X","authenticated-orcid":false,"given":"Ferial","family":"Abuhantash","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Aamna","family":"AlShehhi","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"54368","reference":[{"key":"2024112219550406000_2024.11.19.24317595v1.1","unstructured":"Gauthier, S. , Rosa-Neto, P. , Morais, J. & Webster, C. World Alzheimer Report 2021: Journey Through the Diagnosis of Dementia. (Alzheimer\u2019s Disease International, 2021)."},{"key":"2024112219550406000_2024.11.19.24317595v1.2","doi-asserted-by":"publisher","DOI":"10.5402\/2012\/984786"},{"key":"2024112219550406000_2024.11.19.24317595v1.3","doi-asserted-by":"publisher","DOI":"10.1586\/14737159.2015.1002469"},{"key":"2024112219550406000_2024.11.19.24317595v1.4","first-page":"243","article-title":"Beta-protein amyloid is widely distributed in the central nervous system of patients with Alzheimer\u2019s disease","volume":"134","year":"1989","journal-title":"Am. J. Pathol"},{"key":"2024112219550406000_2024.11.19.24317595v1.5","doi-asserted-by":"publisher","DOI":"10.1046\/j.1440-1819.2000.00747.x"},{"key":"2024112219550406000_2024.11.19.24317595v1.6","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuron.2020.06.010"},{"key":"2024112219550406000_2024.11.19.24317595v1.7","doi-asserted-by":"publisher","DOI":"10.1212\/WNL.42.3.631"},{"key":"2024112219550406000_2024.11.19.24317595v1.8","doi-asserted-by":"publisher","DOI":"10.1016\/j.jalz.2018.02.018"},{"key":"2024112219550406000_2024.11.19.24317595v1.9","doi-asserted-by":"publisher","DOI":"10.1007\/s11011-022-01114-1"},{"key":"2024112219550406000_2024.11.19.24317595v1.10","doi-asserted-by":"publisher","DOI":"10.1016\/j.jalz.2017.06.2266"},{"key":"2024112219550406000_2024.11.19.24317595v1.11","doi-asserted-by":"publisher","DOI":"10.1038\/nature25456"},{"key":"2024112219550406000_2024.11.19.24317595v1.12","doi-asserted-by":"publisher","DOI":"10.1186\/s13041-019-0448-1"},{"key":"2024112219550406000_2024.11.19.24317595v1.13","doi-asserted-by":"publisher","DOI":"10.1016\/S1474-4422(21)00361-6"},{"key":"2024112219550406000_2024.11.19.24317595v1.14","doi-asserted-by":"publisher","DOI":"10.1111\/joim.12816"},{"key":"2024112219550406000_2024.11.19.24317595v1.15","doi-asserted-by":"publisher","DOI":"10.1242\/dmm.031781"},{"key":"2024112219550406000_2024.11.19.24317595v1.16","doi-asserted-by":"publisher","DOI":"10.3389\/fnagi.2022.895398"},{"key":"2024112219550406000_2024.11.19.24317595v1.17","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2020.2984355"},{"key":"2024112219550406000_2024.11.19.24317595v1.18","doi-asserted-by":"publisher","DOI":"10.3390\/ijms23169387"},{"key":"2024112219550406000_2024.11.19.24317595v1.19","doi-asserted-by":"publisher","DOI":"10.1186\/s13195-020-00754-8"},{"key":"2024112219550406000_2024.11.19.24317595v1.20","doi-asserted-by":"publisher","DOI":"10.1093\/brain\/awac128"},{"key":"2024112219550406000_2024.11.19.24317595v1.21","doi-asserted-by":"publisher","DOI":"10.1016\/j.ebiom.2022.103929"},{"key":"2024112219550406000_2024.11.19.24317595v1.22","doi-asserted-by":"publisher","DOI":"10.1186\/s41065-022-00252-x"},{"key":"2024112219550406000_2024.11.19.24317595v1.23","doi-asserted-by":"publisher","DOI":"10.1016\/j.trci.2019.11.001"},{"key":"2024112219550406000_2024.11.19.24317595v1.24","doi-asserted-by":"publisher","DOI":"10.1016\/j.jalz.2019.03.007"},{"key":"2024112219550406000_2024.11.19.24317595v1.25","doi-asserted-by":"publisher","DOI":"10.1159\/000330750"},{"key":"2024112219550406000_2024.11.19.24317595v1.26","doi-asserted-by":"publisher","DOI":"10.1093\/brain\/awaa137"},{"key":"2024112219550406000_2024.11.19.24317595v1.27","unstructured":"Lundberg, S. & Lee, S.-I. A unified approach to interpreting model predictions. In Proceedings of the 31st International Conference on Neural Information Processing Systems, NIPS\u201917, 4768\u20134777 (Curran Associates Inc., Red Hook, NY, 2017)."},{"key":"2024112219550406000_2024.11.19.24317595v1.28","doi-asserted-by":"publisher","DOI":"10.1007\/s41666-022-00114-1"},{"key":"2024112219550406000_2024.11.19.24317595v1.29","doi-asserted-by":"publisher","DOI":"10.1016\/j.jalz.2015.05.009"},{"key":"2024112219550406000_2024.11.19.24317595v1.30","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-020-74399-w"},{"key":"2024112219550406000_2024.11.19.24317595v1.31","doi-asserted-by":"publisher","DOI":"10.1002\/hbm.24428"},{"key":"2024112219550406000_2024.11.19.24317595v1.32","doi-asserted-by":"publisher","DOI":"10.1016\/j.jneumeth.2022.109582"},{"key":"2024112219550406000_2024.11.19.24317595v1.33","doi-asserted-by":"publisher","DOI":"10.3389\/fphar.2021.701301"},{"key":"2024112219550406000_2024.11.19.24317595v1.34","doi-asserted-by":"publisher","DOI":"10.1006\/nbdi.2002.0549"},{"key":"2024112219550406000_2024.11.19.24317595v1.35","doi-asserted-by":"publisher","DOI":"10.1016\/j.clineuro.2016.02.015"},{"key":"2024112219550406000_2024.11.19.24317595v1.36","doi-asserted-by":"publisher","DOI":"10.1093\/hmg\/ddg007"},{"key":"2024112219550406000_2024.11.19.24317595v1.37","doi-asserted-by":"publisher","DOI":"10.1016\/j.jns.2005.06.017"},{"key":"2024112219550406000_2024.11.19.24317595v1.38","doi-asserted-by":"publisher","DOI":"10.1007\/s00018-012-1168-1"},{"key":"2024112219550406000_2024.11.19.24317595v1.39","doi-asserted-by":"publisher","DOI":"10.3390\/ijms23063407"},{"key":"2024112219550406000_2024.11.19.24317595v1.40","doi-asserted-by":"publisher","DOI":"10.1186\/s40035-022-00305-1"},{"key":"2024112219550406000_2024.11.19.24317595v1.41","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-019-56614-5"},{"key":"2024112219550406000_2024.11.19.24317595v1.42","doi-asserted-by":"publisher","DOI":"10.1016\/j.nicl.2016.12.011"},{"key":"2024112219550406000_2024.11.19.24317595v1.43","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.abd5575"},{"key":"2024112219550406000_2024.11.19.24317595v1.44","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0218036"},{"key":"2024112219550406000_2024.11.19.24317595v1.45","doi-asserted-by":"publisher","DOI":"10.1016\/j.nrleng.2018.10.022"},{"key":"2024112219550406000_2024.11.19.24317595v1.46","doi-asserted-by":"publisher","DOI":"10.1111\/j.1749-6632.2009.05064.x"},{"key":"2024112219550406000_2024.11.19.24317595v1.47","doi-asserted-by":"publisher","DOI":"10.1186\/s13059-015-0750-x"},{"key":"2024112219550406000_2024.11.19.24317595v1.48","doi-asserted-by":"publisher","DOI":"10.1212\/WNL.0b013e3181cb3e25"},{"key":"2024112219550406000_2024.11.19.24317595v1.49","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btx441"},{"key":"2024112219550406000_2024.11.19.24317595v1.50","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-021-03922-4"},{"key":"2024112219550406000_2024.11.19.24317595v1.51","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkaa1074"},{"key":"2024112219550406000_2024.11.19.24317595v1.52","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-023-37569-0"},{"key":"2024112219550406000_2024.11.19.24317595v1.53","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-021-02827-6"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1101\/2024.11.19.24317595","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T20:20:46Z","timestamp":1768508446000},"score":1,"resource":{"primary":{"URL":"http:\/\/medrxiv.org\/lookup\/doi\/10.1101\/2024.11.19.24317595"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,11,20]]},"references-count":53,"URL":"https:\/\/doi.org\/10.1101\/2024.11.19.24317595","relation":{"is-preprint-of":[{"id-type":"doi","id":"10.1371\/journal.pone.0320360","asserted-by":"subject"}]},"subject":[],"published":{"date-parts":[[2024,11,20]]},"subtype":"preprint"}}