Universitat Politècnica de Catalunya. Departament d'Enginyeria de Sistemes, Automàtica i Informàtica Industrial
2022-10-03
Type 2 diabetes mellitus (T2DM) is a highly heterogeneous chronic disease with different pathophysiological and genetic characteristics affecting its progression, associated complications and response to therapies. The advances in deep learning (DL) techniques and the availability of a large amount of healthcare data allow us to investigate T2DM characteristics and evolution with a completely new approach, studying common disease trajectories rather than cross sectional values. We used an Kernelized-AutoEncoder algorithm to map 5 years of data of 11,028 subjects diagnosed with T2DM in a latent space that embedded similarities and differences between patients in terms of the evolution of the disease. Once we obtained the latent space, we used classical clustering algorithms to create longitudinal clusters representing different evolutions of the diabetic disease. Our unsupervised DL clustering algorithm suggested seven different longitudinal clusters. Different mean ages were observed among the clusters (ranging from 65.3±11.6 to 72.8±9.4). Subjects in clusters B (Hypercholesteraemic) and E (Hypertensive) had shorter diabetes duration (9.2±3.9 and 9.5±3.9 years respectively). Subjects in Cluster G (Metabolic) had the poorest glycaemic control (mean glycated hemoglobin 7.99±1.42%), while cluster E had the best one (mean glycated hemoglobin 7.04±1.11%). Obesity was observed mainly in clusters A (Neuropathic), C (Multiple Complications), F (Retinopathy) and G. A dashboard is available at dm2.b2slab.upc.edu to visualize the different trajectories corresponding to the 7 clusters.
This work was supported by the Spanish Ministry of Economy and Competitiveness (www.mineco.gob.es) TEC2014-60337-R, DPI2017-89827-R, Networking Biomedical Research Centre in the subject area of Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN), Spain , initiatives of Instituto de Investigación Carlos III (ISCIII), and Share4Rare project (Grant Agreement 780262). This study was possible thanks to the commitment of physicians and nurses working in the Catalan Health Institute to provide optimal care to patients with diabetes. CIBER of Diabetes and Associated Metabolic Diseases (CIBERDEM) is an initiative from Instituto de Salud Carlos III, Madrid, Spain. This analysis is part of the DiaCare Project of Novo Nordisk and the Fundació TicSalut (Departament de Salut, Generalitat de Catalunya), in collaboration with Grupo Pulso, for the benefit of people with type 2 diabetes.
Peer Reviewed
Postprint (published version)
Article
Anglès
Àrees temàtiques de la UPC::Enginyeria biomèdica::Electrònica biomèdica; Non-insulin-dependent diabetes; Type 2 diabetes; Deep learning; Longitudinal cluster; AutoEncoder; Diabetic complications; Electronic health records; Diabetis no-insulinodependent; Aprenentatge profund
https://www.sciencedirect.com/science/article/pii/S1532046422002234
info:eu-repo/grantAgreement/MINECO//TEC2014-60337-R/ES/IMPACTO DEL ENTRENAMIENTO EN DEPORTISTAS DE ELITE EN LA FUNCION CARDIACA, REGULACION NEURAL Y REGULACION GENETICA ASOCIADA/
info:eu-repo/grantAgreement/AEI/Plan Estatal de Investigación Científica y Técnica y de Innovación 2013-2016/DPI2017-89827-R/ES/RENDIMIENTO EN ALTURA EN ATLETAS DE ELITE: ANALISI DE FACTORES GENETICOS, METABOLICOS Y NERUOCARDIOVASCULARES EN ENTRENAMIENTO Y COMPETICION./
info:eu-repo/grantAgreement/EC/H2020/780262/EU/Social media platform dedicated to rare diseases, using collective intelligence for the generation of awareness and advanced knowledge on this large group of diseases./SHARE4RARE
http://creativecommons.org/licenses/by-nc-nd/4.0/
Open Access
Attribution-NonCommercial-NoDerivatives 4.0 International
E-prints [72987]