Objectives: The present study aimed to develop a clinical decision support tool to assist coronavirus disease 2019 (COVID-19) diagnoses with machine learning (ML) models using routine laboratory test results. Methods: We developed ML models using laboratory data (n = 1,391) composed of six clinical chemistry (CC) results, 14 CBC parameter results, and results of a severe acute respiratory syndrome coronavirus 2 real-time reverse transcription–polymerase chain reaction as a gold standard method. Four ML algorithms, including random forest (RF), gradient boosting (XGBoost), support vector machine (SVM), and logistic regression, were used to build eight ML models using CBC and a combination of CC and CBC parameters. Performance evaluation was conducted on the test data set and external validation data set from Brazil. Results: The accuracy values of all models ranged from 74% to 91%. The RF model trained from CC and CBC analytes showed the best performance on the present study’s data set (accuracy, 85.3%; sensitivity, 79.6%; specificity, 91.2%). The RF model trained from only CBC parameters detected COVID-19 cases with 82.8% accuracy. The best performance on the external validation data set belonged to the SVM model trained from CC and CBC parameters (accuracy, 91.18%; sensitivity, 100%; specificity, 84.21%). Conclusions: ML models presented in this study can be used as clinical decision support tools to contribute to physicians’ clinical judgment for COVID-19 diagnoses.

Background: This article presents the results of a comprehensive national model developed for managing maladaptive behaviors (MBs) in children with mental special needs (CMSNs) during the coronavirus disease 2019 (COVID-19) pandemic that combines telehealth-based Applied Behavioral Analyses, psychiatric interventions, and support from local psychosocial teams. This study aims to determine the effectiveness of a system that combined telehealth applications with local psychosocial support teams, allowing services from video calls to emergency interventions. Materials and Methods: The system combines the telehealth applications with the services from local psychosocial intervention teams. In addition to system records covering sociodemographic variables and initial complaints, a telephone survey questioning the effectiveness and satisfaction of the system was used as the main outcome. Results: In total, 347 individuals used the system with mothers constituting the majority of applicants (88.7%, n = 332). The overall satisfaction of the system was 8.8/10. In terms of effectiveness, 63.3% ( n = 237) of caregivers reported an improvement in the reason of application. Counselors decided on a need for follow-up visits for 36.6% ( n = 137) of applications. A referral to a psychiatrist was asked for 40 patients (10.6%). Discussion: To our best knowledge, this is the first study presenting a model for managing MBs of CMSNs during the COVID-19 outbreak. In general, therefore, it seems that there is a need for unique systems to handle behavioral problems of CMSNs. Conclusions: The findings of this study suggest that it is possible to establish an integrative multistep multidisciplinary telehealth-based approach in a short while.

Objectives: This study aimed to objectively assess the potential severity of harm associated with erroneous results in 195 laboratory tests by surveying 514 specialist physicians and medical biochemistry experts. Methods: The survey obtained participants’ (75 medical biochemists, 439 clinicians) opinions on severity of harm for the erroneous results of 195 tests. The comprehensive list of errors and their effects on test results were obtained from the literature, and then matched with severity of harm scores, from 1 (negligible effect) to 5 (life-threatening injury/death), obtained from the survey responses. Results: Participants perceived tests such as cardiac biomarkers, blood gases, coagulation parameters (activated partial thromboplastin time, prothrombin time, international normalized ratio, and dimerized plasmin fragment D), critical ions (potassium, sodium), toxic trace elements (lead, mercury), and specific serum drug levels (lithium, digoxin) to have a greater potential for patient harm in case of errors. Medical biochemistry specialists assigned higher severity scores to some laboratory tests, including total bilirubin, pseudocholinesterase, platelet indices, and some drug levels (cyclosporine, methotrexate, vancomycin). Conclusions: A substantial agreement (91%) was observed between medical biochemists and clinicians in terms of the most frequently chosen severity of harm score. The study provided objective severity scores and identified high-risk tests for targeted quality improvement.

This comprehensive review explores the integration of artificial intelligence (AI) in the field of social pediatrics, emphasizing its potential to revolutionize child healthcare. Social pediatrics, a specialized branch within the discipline, focuses on the significant influence of societal, environmental, and economic factors on children’s health and development. This field adopts a holistic approach, integrating medical, psychological, and environmental considerations. This review aims to explore the potential of AI in revolutionizing child healthcare from social pediatrics perspective. To achieve that, we explored AI applications in preventive care, growth monitoring, nutritional guidance, environmental risk factor prediction, and early detection of child abuse. The findings highlight AI’s significant contributions in various areas of social pediatrics. Artificial intelligence’s proficiency in handling large datasets is shown to enhance diagnostic processes, personalize treatments, and improve overall healthcare management. Notable advancements are observed in preventive care, growth monitoring, nutritional counseling, predicting environmental risks, and early child abuse detection. We find that integrating AI into social pediatric healthcare aims to enhance the effectiveness, accessibility, and equity of pediatric health services. This integration ensures high-quality care for every child, regardless of their social background. The study elucidates AI’s multifaceted applications in social pediatrics, including natural language processing, machine learning algorithms for health outcome predictions, and AI-driven tools for health and environmental monitoring, collectively fostering a more efficient, informed, and responsive pediatric healthcare system. Cite this article as: Gülşen M, Yalçın SS. Fostering tomorrow: Uniting artificial intelligence and social pediatrics for comprehensive child well-being. Turk Arch Pediatr. 2024;59(4):345-352.