PMID- 36878346 OWN - NLM STAT- MEDLINE DCOM- 20230522 LR - 20240305 IS - 1873-2933 (Electronic) IS - 0009-9120 (Linking) VI - 116 DP - 2023 Jun TI - Machine learning to improve false-positive results in the Dutch newborn screening for congenital hypothyroidism. PG - 7-10 LID - S0009-9120(23)00041-3 [pii] LID - 10.1016/j.clinbiochem.2023.03.001 [doi] AB - OBJECTIVE: The Dutch Congenital hypothyroidism (CH) Newborn Screening (NBS) algorithm for thyroidal and central congenital hypothyroidism (CH-T and CH-C, respectively) is primarily based on determination of thyroxine (T4) concentrations in dried blood spots, followed by thyroid-stimulating hormone (TSH) and thyroxine-binding globulin (TBG) measurements enabling detection of both CH-T and CH-C, with a positive predictive value (PPV) of 21%. A calculated T4/TBG ratio serves as an indirect measure for free T4. The aim of this study is to investigate whether machine learning techniques can help to improve the PPV of the algorithm without missing the positive cases that should have been detected with the current algorithm. DESIGN & METHODS: NBS data and parameters of CH patients and false-positive referrals in the period 2007-2017 and of a healthy reference population were included in the study. A random forest model was trained and tested using a stratified split and improved using synthetic minority oversampling technique (SMOTE). NBS data of 4668 newborns were included, containing 458 CH-T and 82 CH-C patients, 2332 false-positive referrals and 1670 healthy newborns. RESULTS: Variables determining identification of CH were (in order of importance) TSH, T4/TBG ratio, gestational age, TBG, T4 and age at NBS sampling. In a Receiver-Operating Characteristic (ROC) analysis on the test set, current sensitivity could be maintained, while increasing the PPV to 26%. CONCLUSIONS: Machine learning techniques have the potential to improve the PPV of the Dutch CH NBS. However, improved detection of currently missed cases is only possible with new, better predictors of especially CH-C and a better registration and inclusion of these cases in future models. CI - Copyright (c) 2023. Published by Elsevier Inc. FAU - Stroek, Kevin AU - Stroek K AD - Endocrine Laboratory, Department of Clinical Chemistry, Amsterdam Gastroenterology, Endocrinology & Metabolism, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands. FAU - Visser, Allerdien AU - Visser A AD - Department of Clinical Chemistry, Amsterdam Gastroenterology, Endocrinology & Metabolism, Amsterdam UMC, Vrije Universiteit, Amsterdam, The Netherlands. FAU - van der Ploeg, Catharina P B AU - van der Ploeg CPB AD - Netherlands Organization for Applied Scientific Research TNO, Department of Child Health, Leiden, The Netherlands. FAU - Zwaveling-Soonawala, Nitash AU - Zwaveling-Soonawala N AD - Department of Paediatric Endocrinology, Emma Children's Hospital, Amsterdam Gastroenterology, Endocrinology & Metabolism, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands. FAU - Heijboer, Annemieke C AU - Heijboer AC AD - Endocrine Laboratory, Department of Clinical Chemistry, Amsterdam Gastroenterology, Endocrinology & Metabolism, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands; Endocrine Laboratory, Department of Clinical Chemistry, Amsterdam Gastroenterology, Endocrinology & Metabolism, Amsterdam UMC, Vrije Universiteit, Amsterdam, The Netherlands. FAU - Bosch, Annet M AU - Bosch AM AD - Department of Pediatrics, Division of Metabolic Disorders, Emma Children's Hospital, Amsterdam Gastroenterology, Endocrinology & Metabolism, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands. FAU - van Trotsenburg, A S Paul AU - van Trotsenburg ASP AD - Department of Paediatric Endocrinology, Emma Children's Hospital, Amsterdam Gastroenterology, Endocrinology & Metabolism, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands. FAU - Boelen, Anita AU - Boelen A AD - Endocrine Laboratory, Department of Clinical Chemistry, Amsterdam Gastroenterology, Endocrinology & Metabolism, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands. FAU - Hoogendoorn, Mark AU - Hoogendoorn M AD - Department of Computer Science, Vrije Universiteit, Amsterdam, The Netherlands. FAU - de Jonge, Robert AU - de Jonge R AD - Department of Clinical Chemistry, Amsterdam UMC, Vrije Universiteit & University of Amsterdam, Amsterdam Gastroenterology, Endocrinology & Metabolism, Amsterdam, The Netherlands. Electronic address: r.dejonge1@amsterdamumc.nl. LA - eng PT - Journal Article DEP - 20230304 PL - United States TA - Clin Biochem JT - Clinical biochemistry JID - 0133660 RN - Q51BO43MG4 (Thyroxine) RN - 0 (Glycoprotein Hormones, alpha Subunit) RN - 0 (Thyroxine-Binding Globulin) SB - IM MH - Humans MH - *Neonatal Screening MH - *Congenital Hypothyroidism/diagnosis MH - *Machine Learning MH - Thyroxine/analysis MH - Glycoprotein Hormones, alpha Subunit/analysis MH - Thyroxine-Binding Globulin/analysis MH - False Positive Reactions MH - Algorithms MH - Gestational Age MH - Infant, Newborn MH - *Random Forest OTO - NOTNLM OT - Congenital hypothyroidism OT - Machine learning OT - Neonatal screening OT - Random forest COIS- Declaration of Competing Interest Annet M. Bosch has received a speaker's fee from Nutricia and has been a member of advisory boards for Biomarin. EDAT- 2023/03/07 06:00 MHDA- 2023/05/22 06:42 CRDT- 2023/03/06 19:24 PHST- 2022/11/06 00:00 [received] PHST- 2023/02/28 00:00 [revised] PHST- 2023/03/02 00:00 [accepted] PHST- 2023/05/22 06:42 [medline] PHST- 2023/03/07 06:00 [pubmed] PHST- 2023/03/06 19:24 [entrez] AID - S0009-9120(23)00041-3 [pii] AID - 10.1016/j.clinbiochem.2023.03.001 [doi] PST - ppublish SO - Clin Biochem. 2023 Jun;116:7-10. doi: 10.1016/j.clinbiochem.2023.03.001. Epub 2023 Mar 4.