As a major challenge in the global field of critical care medicine, sepsis is associated with high morbidity and mortality, driving intensive research into the development of precise risk prediction models. Among these, machine learning and deep learning-based predictive models have garnered particular attention. This article systematically reviews the evolution of sepsis risk prediction models and the challenges associated with their clinical translation, with a focus on their clinical applicability in common septic complications, including acute kidney injury, encephalopathy, and coagulation disorders. Current evidence indicates that machine learning models, by integrating multidimensional dynamic data from electronic health records, significantly enhance early warning capabilities and individualized predictive performance. However, major bottlenecks limiting clinical translation include insufficient data standardization, lack of model interpretability, and inadequate external validation. Future research should prioritize multicenter collaboration, time-series modeling strategies, and explainable artificial intelligence frameworks to optimize clinical translation pathways and facilitate the paradigm shift from traditional empirical approaches to data-driven precision strategies in sepsis risk stratification.

Objective:To explore the reliability and validity of the Chinese version of Hammersmith Neonatal Neurological Examination (C-HNNE).Methods:A prospective cohort study.One hundred and fourteen neonates born in Longgang District Maternity & Child Healthcare Hospital of Shenzhen City between October 2022 and August 2023, who were hospitalized in the Neonatology or Obstetrics Department after birth and met the inclusion criteria, were enrolled as study subjects.They were divided into an early preterm group(34 cases), a mid-late preterm group(50 cases), and a full-term group(30 cases) based on gestational age.The first C-HNNE assessment was completed within 24 hours after birth, and 20 cases from each group were selected for inter-rater reliability assessment; 48 hours after the first C-HNNE assessment, 20 cases from each group were selected to undergo the C-HNNE assessment again for test-retest reliability assessment.At corrected 4 months of age, short-term neurodevelopmental outcomes were determined by pediatric rehabilitation physicians using clinical examination combined with general movements (GMs) assessment.Inter-rater reliability and test-retest reliability were assessed by calculating intra-class correlation coefficient (ICC). The optimal cutoff scores of the C-HNNE for each group were determined by plotting receiver operating characteristic(ROC) curves.Predictive validity was determined by calculating sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).Results:The highest sensitivity for predicting neonatal corrected neuromotor developmental outcome at 4 months of age was achieved when the optimal C-HNNE scores were taken as 22.25, 25.25, and 29.25 in the early preterm, mid-late preterm, and term groups, respectively.The ICCs for the inter-rater reliability of the total C-HNNE score and subscale scores in all 3 groups were above 0.7, indicating good reliability.The ICCs for test-retest reliability of some individual items were<0.6, indicating moderate reliability.In terms of validity, the correlation coefficients between the total C-HNNE scores and the GMs scores of the three groups were 0.550, 0.483, 0.473 (all P<0.01), and the sensitivity of the C-HNNE for predicting neurodevelopmental developmental outcomes in neonates corrected to 4 months of age was 82.9%, specificity was 70.4%, PPV was 58.0%, and NPV was 89.3%. Conclusions:C-HNNE demonstrates good reliability and validity and can be used as a routine bedside examination program for early neonatal life.

Objective:To explore the reliability and validity of the Chinese version of Hammersmith Neonatal Neurological Examination (C-HNNE).Methods:A prospective cohort study.One hundred and fourteen neonates born in Longgang District Maternity & Child Healthcare Hospital of Shenzhen City between October 2022 and August 2023, who were hospitalized in the Neonatology or Obstetrics Department after birth and met the inclusion criteria, were enrolled as study subjects.They were divided into an early preterm group(34 cases), a mid-late preterm group(50 cases), and a full-term group(30 cases) based on gestational age.The first C-HNNE assessment was completed within 24 hours after birth, and 20 cases from each group were selected for inter-rater reliability assessment; 48 hours after the first C-HNNE assessment, 20 cases from each group were selected to undergo the C-HNNE assessment again for test-retest reliability assessment.At corrected 4 months of age, short-term neurodevelopmental outcomes were determined by pediatric rehabilitation physicians using clinical examination combined with general movements (GMs) assessment.Inter-rater reliability and test-retest reliability were assessed by calculating intra-class correlation coefficient (ICC). The optimal cutoff scores of the C-HNNE for each group were determined by plotting receiver operating characteristic(ROC) curves.Predictive validity was determined by calculating sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).Results:The highest sensitivity for predicting neonatal corrected neuromotor developmental outcome at 4 months of age was achieved when the optimal C-HNNE scores were taken as 22.25, 25.25, and 29.25 in the early preterm, mid-late preterm, and term groups, respectively.The ICCs for the inter-rater reliability of the total C-HNNE score and subscale scores in all 3 groups were above 0.7, indicating good reliability.The ICCs for test-retest reliability of some individual items were<0.6, indicating moderate reliability.In terms of validity, the correlation coefficients between the total C-HNNE scores and the GMs scores of the three groups were 0.550, 0.483, 0.473 (all P<0.01), and the sensitivity of the C-HNNE for predicting neurodevelopmental developmental outcomes in neonates corrected to 4 months of age was 82.9%, specificity was 70.4%, PPV was 58.0%, and NPV was 89.3%. Conclusions:C-HNNE demonstrates good reliability and validity and can be used as a routine bedside examination program for early neonatal life.

Objective:To explore the genetic basis for a child featuring global developmental delay and epilepsy.Methods:A child who had presented at Guangzhou Women and Children′s Medical Center Liuzhou Hospital on February 19, 2023 was selected as the study subject. Clinical data of the child was collected. The child was subjected to whole exome sequencing, and candidate variant was validated by Sanger sequencing and bioinformatic analysis.Results:The child, an 8-month-old girl, had manifested with global developmental delay, epilepsy, and hyperlactacidemia. Cranial MRI revealed diverse hypomyelinating leukodystrophies. Electroencephalogram showed slow background activities. Genetic testing revealed that she has harbored a homozygous variant of the SLC25A12 gene, namely c. 115T>G (p.Phe39Val), for which both of her parents were heterozygous carriers. Based on the guidelines from the American College of Medical Genetics and Genomics, the variant was predicted to be of uncertain significance (PM2_Supporting+ PM3_Supporting+ PP3_Moderate+ PP4_Moderate). I-Mutant v3.0 software predicted that the variant may affect the stability of protein product. Conclusion:The homozygous c. 115T>G (p.Phe39Val) variant of the SLC25A12 gene probably underlay the pathogenesis of the disease in this child.

Objective To evaluate the value of DWI in detecting active sacroiliitis in patients with axial spondyloarthritis ( SpA) .Methods DWI and conventional MRI examinations were done oblique coronary on bilateral SIJs of 19 participants,involving 8 patients (the case group) underwent clinical and conventional MRI evaluation for axial SpA active sacroiliitis and 11 healthy volunteers(the control group).The conventional MRI,DWI,ADC,EADC images were transmitted to two experienced radiologists respectively to read ,evaluate and measure ( the ADC values and EADC values ) at the workstation under the blind method .The experimental data were inputted and analyzed by statistics software .Results In the control group ,the periarticular bone marrow in the SIJ showed hopyintense signal on the DWI image ,hypointense signal on the ADC image and hyperintense signal on the EADC image .The periarticu-lar bone marrow oedema showed hyperintense signal on the DWI and ADC images , while hopyintense signal on the EADC image.At b values of 300s/mm2 ,the ADC values of the periarticular bone marrow oedema at each side in the case group were significantly higher than those in the control group , while EADC values were lower , the difference were statistically significant(Z=-3.115,-2.814,-3.002,-3.115,-3.115,-2.714,-2.889,-3.115,all P<0.05).Conclusion DWI can directly detect the signal changes ,extent of the periarticular bone marrow oedema , and analyze the severity of oedema quantitively ,which shows the utility in determining the activity and early diagnosis .

Objective To study CT and MRI manifestation of untypical cystic lymphangioma.Methods CT and MRI findings of 8 cases of cystic lymphangioma proved surgically and pathologically were analyzed retrospectively.7 cases were performed by CT and the other by MRI.Results 3 lesions are located at facies cranii,2 lesions at neck,2 lesions are located at mesostenium and 1 lesion is located at chest wall.CT and MRI findings include:(1)dense mass(5 cases);(2)lesion contain solid and water dense(2 cases);(3)lesion contain high dense hematoma(2 cases);(4)calcification in lesion(2 cases);(5)infiltrate the adjacent tissue(3 cases).Conclusion Most majority cystic lymphangioma can be identified by CT and MRI which also helpful to evaluation lymphangioma before surgical treatment.

OBJECTIVE: To analyze the relationship between chronic pain and painprone meutal disorders and investigate the effect of psychological intervention and medication.DATA SOURCES: Related literatures about the relationship between chronic pain and pain-prone mental disorders were searched on computer in Medline and Embase from January 1989 to May 2004 with the retrieval word of "Chronic pain, emotional disorder, treatment", the language was limited to English. The subjects were patients who suffered from chronic pain for above 6 months.STUDY SELECTION: Related literatures about the relationship between chronic pain and pain-prone mental disorders and its therapeutic interventions were carried on primary trial to delete those obsolete and repetitive literatures, and the others were used to search in the full text. The quality was appraised mainly on the basis of the data authenticity, preciseness of the designation, strictness of the implementation and reasonability of the statistical analysis.DATA EXTRACTION: Altogether 20 literatures were obtained and 11literatures were proved accordant with the inclusion standard. 9 literatures were excluded because 7 were repetitive experiments and two were Meta analyses.DATA SYNTHESIS: 11 literatures include experimental group and control group, patients in the experimental group were treated with analgetic combined with psychological intervention and antidepressant while those in the control group were only given analgetic; the outcomes of the two groups were compared.CONCLUSION: Chronic pain may co-exist with emotional disorders. It can be remarkably alleviated by the administration of analgetic combined with psychological intervention and antidepressant, which proves better than single analgetic medication.