Pain management is a crucial concern in medicine, particularly in the case of children who may struggle to effectively communicate their pain. Despite the longstanding reliance on various assessment scales by medical professionals, these tools have shown limitations and subjectivity. In this paper, we present a pain assessment scheme based on skin potential signals, aiming to convert subjective pain into objective indicators for pain identification using machine learning methods. We have designed and implemented a portable non-invasive measurement device to measure skin potential signals and conducted experiments involving 623 subjects. From the experimental data, we selected 358 valid records, which were then divided into 218 silent samples and 262 pain samples. A total of 38 features were extracted from each sample, with seven features displaying superior performance in pain identification. Employing three classification algorithms, we found that the random forest algorithm achieved the highest accuracy, reaching 70.63%. While this identification rate shows promise for clinical applications, it is important to note that our results differ from state-of-the-art research, which achieved a recognition rate of 81.5%. This discrepancy arises from the fact that our pain stimuli were induced by clinical operations, making it challenging to precisely control the stimulus intensity when compared to electrical or thermal stimuli. Despite this limitation, our pain assessment scheme demonstrates significant potential in providing objective pain identification in clinical settings. Further research and refinement of the proposed approach may lead to even more accurate and reliable pain management techniques in the future.

Background: A good prognosis of congenital heart disease (CHD) depends on early diagnosis and intervention. Under the current screening conditions, a significant proportion still go undetected. Metabolomics, as a phenotype‐correlated research methodology, remains underused in the study of CHD, which could provide the possibility to screen neonatal CHD efficiently. Methods: Data for the analysis are from >22 000 neonates captured in the Network Platform for CHD from April 2020 to November 2021 in 11 cities in China. After data matching and quality control, a total of 22 674 neonates were finally included and divided into the CHD group (n=1823), nonsignificant CHD group (n=17 968), and normal group (n=2748). Demographic and clinical characteristics and tandem mass spectrometry‐based metabolic data for genetic and metabolic disease screening were gathered and compared for all groups. Machine learning models based on metabolic biomarkers were constructed to screen CHD in neonates. Results: After quality control, 22 539 neonates were ultimately included. Among them, 1823 were diagnosed with CHD, 17 968 were nonsignificant CHD, and 2748 were normal. A total of 46 distinguishing metabolic biomarkers were identified, and we found that the CHD group had significantly lower levels of 17‐hydroxyprogesterone (CHD versus nonsignificant CHD, P <0.001, log ₂ fold change=−0.16; CHD versus normal, P <0.001, log ₂ fold change=−0.15). We constructed CHD and ventricular septal disease screening models based on metabolic biomarkers. The best fitting model achieved an area under the receiver operating characteristic curve of 0.745 (95% CI, 0.696–0.791). Conclusions: This study reveals the unique metabolic profile of neonates with CHD. The screening model demonstrates considerable potential in early neonatal CHD screening and reflects significant value from a health economics perspective.