[Etiology composition and prognosis of pediatric chronic critical illness in a pediatric intensive care unit]. / å„¿ç«¥é‡ç—‡ç›‘æŠ¤å®¤å„¿ç«¥æ ¢æ€§å±é‡ç—‡çš„ç— å› ç»„æˆåŠé¢„åŽåˆ†æž.

OBJECTIVES: To explore the etiology composition and outcomes of pediatric chronic critical illness (PCCI) in the pediatric intensive care unit (PICU). METHODS: The children who were hospitalized in the PICU of Dongguan Children's Hospital Affiliated to Guangdong Medical University and met the diagnostic criteria for PCCI from January 2017 to December 2022 were included in the study. The etiology of the children was classified based on their medical records and discharge diagnoses. Relevant clinical data during hospitalization were collected and analyzed. RESULTS: Among the 3 955 hospitalized children in the PICU from January 2017 to December 2022, 321 cases (8.12%) met the diagnostic criteria for PCCI. Among the 321 cases, the most common etiology was infection (71.3%, 229 cases), followed by unintentional injury (12.8%, 41 cases), postoperation (5.9%, 19 cases), tumors/immune system diseases (5.0%, 16 cases), and genetic and chromosomal diseases (5.0%, 16 cases). Among the 321 cases, 249 cases (77.6%) were discharged after improvement, 37 cases (11.5%) were discharged at the request of the family, and 35 cases (10.9%) died in the hospital. Among the deaths, infection accounted for 74% (26/35), unintentional injury accounted for 17% (6/35), tumors/immune system diseases accounted for 6% (2/35), and genetic and chromosomal diseases accounted for 3% (1/35). From 2017 to 2022, the proportion of PCCI in PICU diseases showed an increasing trend year by year (P<0.05). Among the 321 children with PCCI, there were 148 infants and young children (46.1%), 57 preschool children (17.8%), 54 school-aged children (16.8%), and 62 adolescents (19.3%), with the highest proportion in the infant and young children group (P<0.05). The in-hospital mortality rates of the four age groups were 14.9% (22/148), 8.8% (5/57), 5.6% (3/54), and 8.1% (5/62), respectively. The infant and young children group had the highest mortality rate, but there was no statistically significant difference among the four groups (P>0.05). CONCLUSIONS: The proportion of PCCI in PICU diseases is increasing, and the main causes are infection and unintentional injury. The most common cause of death in children with PCCI is infection. The PCCI patient population is mainly infants and young children, and the in-hospital mortality rate of infant and young children with PCCI is relatively high.

A Multi-Task Group Bi-LSTM Networks Application on Electrocardiogram Classification.

BACKGROUND: Cardiovascular diseases (CVD) are the leading cause of death globally. Electrocardiogram (ECG) analysis can provide thoroughly assessment for different CVDs efficiently. We propose a multi-task group bidirectional long short-term memory (MTGBi-LSTM) framework to intelligent recognize multiple CVDs based on multi-lead ECG signals. METHODS: This model employs a Group Bi-LSTM (GBi-LSTM) and Residual Group Convolutional Neural Network (Res-GCNN) to learn the dual feature representation of ECG space and time series. GBi-LSTM is divided into Global Bi-LSTM and Intra-Group Bi-LSTM, which can learn the features of each ECG lead and the relationship between leads. Then, through attention mechanism, the different lead information of ECG is integrated to make the model to possess the powerful feature discriminability. Through multi-task learning, the model can fully mine the association information between diseases and obtain more accurate diagnostic results. In addition, we propose a dynamic weighted loss function to better quantify the loss to overcome the imbalance between classes. RESULTS: Based on more than 170,000 clinical 12-lead ECG analysis, the MTGBi-LSTM method achieved accuracy, precision, recall and F1 of 88.86%, 90.67%, 94.19% and 92.39%, respectively. The experimental results show that the proposed MTGBi-LSTM method can reliably realize ECG analysis and provide an effective tool for computer-aided diagnosis of CVD.