An ensemble-based machine learning model for predicting type 2 diabetes and its effect on bone health.
Bone health
Diabetes
Dual Energy X-ray Absorptiometry
Machine learning
Qatar Biobank (QBB)
Journal
BMC medical informatics and decision making
ISSN: 1472-6947
Titre abrégé: BMC Med Inform Decis Mak
Pays: England
ID NLM: 101088682
Informations de publication
Date de publication:
29 May 2024
29 May 2024
Historique:
received:
01
10
2023
accepted:
17
05
2024
medline:
30
5
2024
pubmed:
30
5
2024
entrez:
29
5
2024
Statut:
epublish
Résumé
Diabetes is a chronic condition that can result in many long-term physiological, metabolic, and neurological complications. Therefore, early detection of diabetes would help to determine a proper diagnosis and treatment plan. In this study, we employed machine learning (ML) based case-control study on a diabetic cohort size of 1000 participants form Qatar Biobank to predict diabetes using clinical and bone health indicators from Dual Energy X-ray Absorptiometry (DXA) machines. ML models were utilized to distinguish diabetes groups from non-diabetes controls. Recursive feature elimination (RFE) was leveraged to identify a subset of features to improve the performance of model. SHAP based analysis was used for the importance of features and support the explainability of the proposed model. Ensemble based models XGboost and RF achieved over 84% accuracy for detecting diabetes. After applying RFE, we selected only 20 features which improved the model accuracy to 87.2%. From a clinical standpoint, higher HDL-Cholesterol and Neutrophil levels were observed in the diabetic group, along with lower vitamin B12 and testosterone levels. Lower sodium levels were found in diabetics, potentially stemming from clinical factors including specific medications, hormonal imbalances, unmanaged diabetes. We believe Dapagliflozin prescriptions in Qatar were associated with decreased Gamma Glutamyltransferase and Aspartate Aminotransferase enzyme levels, confirming prior research. We observed that bone area, bone mineral content, and bone mineral density were slightly lower in the Diabetes group across almost all body parts, but the difference against the control group was not statistically significant except in T12, troch and trunk area. No significant negative impact of diabetes progression on bone health was observed over a period of 5-15 yrs in the cohort. This study recommends the inclusion of ML model which combines both DXA and clinical data for the early diagnosis of diabetes.
Sections du résumé
BACKGROUND
BACKGROUND
Diabetes is a chronic condition that can result in many long-term physiological, metabolic, and neurological complications. Therefore, early detection of diabetes would help to determine a proper diagnosis and treatment plan.
METHODS
METHODS
In this study, we employed machine learning (ML) based case-control study on a diabetic cohort size of 1000 participants form Qatar Biobank to predict diabetes using clinical and bone health indicators from Dual Energy X-ray Absorptiometry (DXA) machines. ML models were utilized to distinguish diabetes groups from non-diabetes controls. Recursive feature elimination (RFE) was leveraged to identify a subset of features to improve the performance of model. SHAP based analysis was used for the importance of features and support the explainability of the proposed model.
RESULTS
RESULTS
Ensemble based models XGboost and RF achieved over 84% accuracy for detecting diabetes. After applying RFE, we selected only 20 features which improved the model accuracy to 87.2%. From a clinical standpoint, higher HDL-Cholesterol and Neutrophil levels were observed in the diabetic group, along with lower vitamin B12 and testosterone levels. Lower sodium levels were found in diabetics, potentially stemming from clinical factors including specific medications, hormonal imbalances, unmanaged diabetes. We believe Dapagliflozin prescriptions in Qatar were associated with decreased Gamma Glutamyltransferase and Aspartate Aminotransferase enzyme levels, confirming prior research. We observed that bone area, bone mineral content, and bone mineral density were slightly lower in the Diabetes group across almost all body parts, but the difference against the control group was not statistically significant except in T12, troch and trunk area. No significant negative impact of diabetes progression on bone health was observed over a period of 5-15 yrs in the cohort.
CONCLUSION
CONCLUSIONS
This study recommends the inclusion of ML model which combines both DXA and clinical data for the early diagnosis of diabetes.
Identifiants
pubmed: 38811939
doi: 10.1186/s12911-024-02540-0
pii: 10.1186/s12911-024-02540-0
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
144Informations de copyright
© 2024. The Author(s).
Références
Mohsin S, Baniyas MM, AlDarmaki RS, Tekes K, Kalász H, Adeghate EA. An update on therapies for the treatment of diabetes-induced osteoporosis. Exp Opin Biol Ther. 2019;19(9):937–48.
doi: 10.1080/14712598.2019.1618266
Patel R. Studies on Genotype-Phenotype Correlation in Type II Diabetics and Evaluation of Melatonin and DPP-IV inhibitor on Experimental Diabetic Models. India: Maharaja Sayajirao University of Baroda; 2021.
Egan AM, Dinneen SF. What is diabetes? Medicine. 2019;47(1):1–4.
doi: 10.1016/j.mpmed.2018.10.002
World Health Organization. Global diffusion of eHealth: making universal health coverage achievable: report of the third global survey on eHealth. World Health Organization; 2017.
Sabanayagam C, Yip W, Ting DS, Tan G, Wong TY. Ten emerging trends in the epidemiology of diabetic retinopathy. Ophthalmic Epidemiol. 2016;23(4):209–22.
doi: 10.1080/09286586.2016.1193618
pubmed: 27355693
Hernandez C. The diabetes epidemic prediction in 2045. 2019. https://wordofhealth.com/2018/11/30/the-diabetes-epidemic-prediction-in-2045/ . Accessed 15 Dec 2023.
Ministry of Public Health Q. Ministry of Public Health - Qatar Public Health Strategy 2022 - 2017. 2017. https://moph.gov.qa/english/strategies/supporting-strategies-and-frameworks/qatarpublichealthstrategy/pages/default.aspx . Accessed 03 Feb 2023.
Lefèbvre P, Pierson A. The global challenge of diabetes. World Hosp Health Serv. 2004;40(3):37–9.
pubmed: 15566278
International Diabetes Federation. IDF Diabetes Atlas 10th Edition. 2021. https://www.diabetesatlas.org/data/en/ . Accessed 04 Feb 2023.
Hassan A. Trends in nutrition related chronic diseases in Qatar: a call for action. Emirates J Food Agric. 1994;6:128–40.
Nazeemudeen A, Al-Absi HR, Refaee MA, Househ M, Shah Z, Alam T. Understanding the Food Habits and Physical Activities of Diabetes Cohort in Qatar. In: The Importance of Health Informatics in Public Health during a Pandemic. IOS Press; 2020. pp. 453–6.
Abbas M, Mall R, Errafii K, Lattab A, Ullah E, Bensmail H, et al. Simple risk score to screen for prediabetes: A cross-sectional study from the Qatar Biobank cohort. J Diabetes Investig. 2021;12(6):988–97.
doi: 10.1111/jdi.13445
pubmed: 33075216
Sadek KW, Abdelhafez I, Al-Hashimi I, Al-Shafi W, Tarmizi F, Al-Marri H, et al. Diabetes risk score in Qatar: Model development, validation, and external validation of several models. medRxiv. 2021;2021–04.
Musleh S, Alam T, Bouzerdoum A, Belhaouari SB, Baali H. Identification of potential risk factors of diabetes for the qatari population. In: 2020 IEEE International Conference on Informatics, IoT, and Enabling Technologies (ICIoT). IEEE; 2020. pp. 243–6.
Islam MT, Al-Absi HR, Ruagh EA, Alam T. DiaNet: A deep learning based architecture to diagnose diabetes using retinal images only. IEEE Access. 2021;9:15686–95.
doi: 10.1109/ACCESS.2021.3052477
Al-Absi HR, Pai A, Naeem U, Mohamed FK, Arya S, Sbeit RA, et al. DiaNet v2 deep learning based method for diabetes diagnosis using retinal images. Sci Rep. 2024;14(1):1595.
doi: 10.1038/s41598-023-49677-y
pubmed: 38238377
pmcid: 10796402
Wachinger C, Wolf TN, Pölsterl S. Deep learning for the prediction of type 2 diabetes mellitus from neck-to-knee Dixon MRI in the UK biobank. Heliyon. 2023;9(11).
Sadek K, Abdelhafez I, Al-Hashimi I, Al-Shafi W, Tarmizi F, Al-Marri H, et al. Screening for diabetes and impaired glucose metabolism in Qatar: Models’ development and validation. Prim Care Diabetes. 2022;16(1):69–77.
doi: 10.1016/j.pcd.2021.10.002
pubmed: 34716113
Lam B, Catt M, Cassidy S, Bacardit J, Darke P, Butterfield S, et al. Using wearable activity trackers to predict type 2 diabetes: machine learning-based cross-sectional study of the UK Biobank accelerometer cohort. JMIR Diabetes. 2021;6(1):e23364.
doi: 10.2196/23364
pubmed: 33739298
pmcid: 8080299
Jacobs PG, Herrero P, Facchinetti A, Vehi J, Kovatchev B, Breton M, et al. Artificial intelligence and machine learning for improving glycemic control in diabetes: best practices, pitfalls and opportunities. IEEE Rev Biomed Eng. 2023;17:19-41.
D’Antoni F, Petrosino L, Marchetti A, Bacco L, Pieralice S, Vollero L, et al. Layered meta-learning algorithm for predicting adverse events in type 1 diabetes. IEEE Access. 2023;11:9074–94.
doi: 10.1109/ACCESS.2023.3237992
Piepkorn B, Kann P, Forst T, Andreas J, Pfützner A, Beyer J. Bone mineral density and bone metabolism in diabetes mellitus. Horm Metab Res. 1997;29(11):584–91.
doi: 10.1055/s-2007-979106
pubmed: 9479561
Refaee MA, Al-Absi HR, Islam MT, Househ M, Shah Z, Rahman MS, et al. The Linkage Between Bone Densitometry and Cardiovascular Disease. In: Informatics and Technology in Clinical Care and Public Health. IOS Press; 2022. pp. 244–7.
Musleh S, Nazeemudeen A, Islam MT, El Hajj N, Alam T. A machine learning based study to assess bone health in a diabetic cohort. Inform Med Unlocked. 2022;33:101079.
doi: 10.1016/j.imu.2022.101079
Remeseiro B, Bolon-Canedo V. A review of feature selection methods in medical applications. Comput Biol Med. 2019;112:103375.
doi: 10.1016/j.compbiomed.2019.103375
pubmed: 31382212
Lundberg SM, Lee SI. A unified approach to interpreting model predictions. Adv Neural Inf Process Syst. 2017;30:4765-74.
Niemeijer-Kanters S, Banga J, Erkelens D. Lipid-lowering therapy in diabetes mellitus. Neth J Med. 2001;58(5):214–22.
doi: 10.1016/S0300-2977(01)00105-X
pubmed: 11414234
Sparre Hermann L, Nilsson B, Wettre S. Vitamin B12 status of patients treated with metformin: a cross-sectional cohort study. Brit J Diabetes Vasc Dis. 2004;4(6):401–6.
doi: 10.1177/14746514040040060701
Grossmann M. Low testosterone in men with type 2 diabetes: significance and treatment. J Clin Endocrinol Metab. 2011;96(8):2341–53.
doi: 10.1210/jc.2011-0118
pubmed: 21646372
Khan RN, Saba F, Kausar SF, Siddiqui MH. Pattern of electrolyte imbalance in Type 2 diabetes patients: Experience from a tertiary care hospital. Pak J Medical Sci. 2019;35(3):797.
doi: 10.12669/pjms.35.3.844
Kunutsor SK, Abbasi A, Apekey TA. Aspartate aminotransferase-risk marker for type-2 diabetes mellitus or red herring? Front Endocrinol. 2014;5:189.
doi: 10.3389/fendo.2014.00189
Benjamin S, Ramanjaneya M, Butler AE, Janjua I, Paramba F, Palaki J, et al. Dapagliflozin, as Add-on Therapy in Type 2 Diabetes Patients, Is Associated With a Reduction in Albuminuria and Serum Transaminase Levels. Front Clin Diabetes Healthc. 2021;2:733693.
doi: 10.3389/fcdhc.2021.733693
pubmed: 36994321
pmcid: 10012067
Zou Q, Qu K, Luo Y, Yin D, Ju Y, Tang H. Predicting diabetes mellitus with machine learning techniques. Front Genet. 2018;9:515.
doi: 10.3389/fgene.2018.00515
pubmed: 30459809
pmcid: 6232260
Sarwar MA, Kamal N, Hamid W, Shah MA. Prediction of diabetes using machine learning algorithms in healthcare. In: 2018 24th international conference on automation and computing (ICAC). IEEE; 2018. pp. 1–6.
Qiu J, Li C, Dong Z, Wang J. Is diabetes mellitus a risk factor for low bone density: a systematic review and meta-analysis. BMC Endocr Disord. 2021;21(1):1–11.
doi: 10.1186/s12902-021-00728-3