Mukkesh Kumar1, Li Chen2, Karen Tan2, Li Ting Ang3, Cindy Ho3, Gerard Wong2, Shu E Soh4, Kok Hian Tan5, Jerry Kok Yen Chan6, Keith M Godfrey7, Shiao-Yng Chan8, Mary Foong Fong Chong9, John E Connolly10, Yap Seng Chong8, Johan G Eriksson11, Mengling Feng12, Neerja Karnani13. 1. Singapore Institute for Clinical Sciences, Agency for Science Technology and Research, Singapore, Republic of Singapore; Bioinformatics Institute, Agency for Science Technology and Research, Singapore, Republic of Singapore; Saw Swee Hock School of Public Health and Institute of Data Science, National University of Singapore and National University Health System, Singapore, Republic of Singapore. 2. Singapore Institute for Clinical Sciences, Agency for Science Technology and Research, Singapore, Republic of Singapore. 3. Singapore Institute for Clinical Sciences, Agency for Science Technology and Research, Singapore, Republic of Singapore; Bioinformatics Institute, Agency for Science Technology and Research, Singapore, Republic of Singapore. 4. Department of Paediatrics, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Republic of Singapore. 5. Division of Obstetrics and Gynecology, KK Women's and Children's Hospital, Republic of Singapore; Obstetrics and Gynecology Academic Clinical Programme, Duke-NUS Graduate Medical School, Singapore, Republic of Singapore. 6. Department of Obstetrics and Gynaecology and Human Potential Translational Research Programme, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Republic of Singapore; Department of Reproductive Medicine, KK Women's and Children's Hospital, Republic of Singapore; Cancer and Stem Cell Biology, Duke-NUS Medical School, Singapore, Republic of Singapore. 7. MRC Lifecourse Epidemiology Unit & NIHR Southampton Biomedical Research Centre, University of Southampton & University Hospital Southampton NHS Foundation Trust, UK. 8. Singapore Institute for Clinical Sciences, Agency for Science Technology and Research, Singapore, Republic of Singapore; Department of Obstetrics and Gynaecology and Human Potential Translational Research Programme, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Republic of Singapore. 9. Singapore Institute for Clinical Sciences, Agency for Science Technology and Research, Singapore, Republic of Singapore; Saw Swee Hock School of Public Health and Institute of Data Science, National University of Singapore and National University Health System, Singapore, Republic of Singapore. 10. Institute of Molecular and Cell Biology, Agency for Science Technology and Research, Singapore, Republic of Singapore. 11. Singapore Institute for Clinical Sciences, Agency for Science Technology and Research, Singapore, Republic of Singapore; Department of Obstetrics and Gynaecology and Human Potential Translational Research Programme, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Republic of Singapore; Department of General Practice and Primary Health Care, University of Helsinki, Finland; Folkhälsan Research Center, Helsinki, Finland. 12. Saw Swee Hock School of Public Health and Institute of Data Science, National University of Singapore and National University Health System, Singapore, Republic of Singapore. Electronic address: ephfm@nus.edu.sg. 13. Singapore Institute for Clinical Sciences, Agency for Science Technology and Research, Singapore, Republic of Singapore; Bioinformatics Institute, Agency for Science Technology and Research, Singapore, Republic of Singapore; Department of Biochemistry, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Republic of Singapore. Electronic address: neerja_karnani@sics.a-star.edu.sg.
Abstract
AIMS: The heterogeneity in Gestational Diabetes Mellitus (GDM) risk factors among different populations impose challenges in developing a generic prediction model. This study evaluates the predictive ability of existing UK NICE guidelines for assessing GDM risk in Singaporean women, and used machine learning to develop a non-invasive predictive model. METHODS: Data from 909 pregnancies in Singapore's most deeply phenotyped mother-offspring cohort study, Growing Up in Singapore Towards healthy Outcomes (GUSTO), was used for predictive modeling. We used a CatBoost gradient boosting algorithm, and the Shapley feature attribution framework for model building and interpretation of GDM risk attributes. RESULTS: UK NICE guidelines showed poor predictability in Singaporean women [AUC:0.60 (95% CI 0.51, 0.70)]. The non-invasive predictive model comprising of 4 non-invasive factors: mean arterial blood pressure in first trimester, age, ethnicity and previous history of GDM, greatly outperformed [AUC:0.82 (95% CI 0.71, 0.93)] the UK NICE guidelines. CONCLUSIONS: The UK NICE guidelines may be insufficient to assess GDM risk in Asian women. Our non-invasive predictive model outperforms the current state-of-the-art machine learning models to predict GDM, is easily accessible and can be an effective approach to minimize the economic burden of universal testing & GDM associated healthcare in Asian populations.
AIMS: The heterogeneity in Gestational Diabetes Mellitus (GDM) risk factors among different populations impose challenges in developing a generic prediction model. This study evaluates the predictive ability of existing UK NICE guidelines for assessing GDM risk in Singaporean women, and used machine learning to develop a non-invasive predictive model. METHODS: Data from 909 pregnancies in Singapore's most deeply phenotyped mother-offspring cohort study, Growing Up in Singapore Towards healthy Outcomes (GUSTO), was used for predictive modeling. We used a CatBoost gradient boosting algorithm, and the Shapley feature attribution framework for model building and interpretation of GDM risk attributes. RESULTS: UK NICE guidelines showed poor predictability in Singaporean women [AUC:0.60 (95% CI 0.51, 0.70)]. The non-invasive predictive model comprising of 4 non-invasive factors: mean arterial blood pressure in first trimester, age, ethnicity and previous history of GDM, greatly outperformed [AUC:0.82 (95% CI 0.71, 0.93)] the UK NICE guidelines. CONCLUSIONS: The UK NICE guidelines may be insufficient to assess GDM risk in Asian women. Our non-invasive predictive model outperforms the current state-of-the-art machine learning models to predict GDM, is easily accessible and can be an effective approach to minimize the economic burden of universal testing & GDM associated healthcare in Asian populations.
Authors: Moshe Hod; Anil Kapur; David A Sacks; Eran Hadar; Mukesh Agarwal; Gian Carlo Di Renzo; Luis Cabero Roura; Harold David McIntyre; Jessica L Morris; Hema Divakar Journal: Int J Gynaecol Obstet Date: 2015-10 Impact factor: 3.561
Authors: Yap-Seng Chong; Shirong Cai; Harvard Lin; Shu E Soh; Yung-Seng Lee; Melvin Khee-Shing Leow; Yiong-Huak Chan; Li Chen; Joanna D Holbrook; Kok-Hian Tan; Victor Samuel Rajadurai; George Seow-Heong Yeo; Michael S Kramer; Seang-Mei Saw; Peter D Gluckman; Keith M Godfrey; Kenneth Kwek Journal: BMC Pregnancy Childbirth Date: 2014-10-02 Impact factor: 3.007