Maternal mortality and infant and child mortality affect many aspects of a country, and can be measured by economic, cultural, health, and other related indicators. Therefore, it is very important for the public health care of pregnant and lying-in women, and the health management of pregnant and lying-in women can also reduce perinatal mortality. Since the full opening of the two-child policy in China, the incidence of diseases of advanced age and high body mass index has also been increasing, and the increase of elderly mothers has also led to an increase in the incidence of cesarean section. Maternal hemorrhage is also a leading cause of maternal mortality, and postpartum hemorrhage is a leading cause of maternal morbidity and mortality worldwide. Accurate assessment of blood loss, identification of risk factors, and timely identification of postpartum hemorrhage remain major challenges in obstetrics. Incomplete facilities in primary hospitals in my country, lack of emergency response capability and lack of clinical experience are also a cause of postpartum hemorrhage. Choosing effective preventive measures is a key to preventing and treating postpartum hemorrhage. Uterine contractions are a factor in postpartum hemorrhage. In this work, the main factors for the emergence of uterine contraction laws were the long labor process, excessive mental progress of the puerperium, excessive consumption of energy, and less food during labor. Polyhydramnios leads to overstretching of uterine muscle fibers. Women with multiple cesarean sections suffer from uterine injury during multiple deliveries, which makes the uterus unable to retract normally, and the uterine wall blood sinuses at the placenta attachment cannot effectively close the placenta, resulting in maternal bleeding. Early application of oxytocin and early implementation of abdominal massage have a significant effect on postpartum hemorrhage caused by contractions. Soft birth canal injury was also found to be a risk factor for postpartum hemorrhage in this work. Soft birth canal injuries are more likely to occur in women undergoing acute labor; the main reasons are excessive force during labor, huge fetus, incomplete suture, incomplete hemostasis, and poor elasticity of the access tract. In the course of treatment, the soft birth should be fully exposed and carefully examined, and the small hematoma should be treated with cold compress and compression. Relevant data show that the cesarean section has twice the blood loss compared with the approach channel, and is more likely to have postpartum hemorrhage (Attali, et al., 2021). Patients with placenta previa are at risk for massive intrapartum and postpartum blood loss, and placenta accreta can lead to serious complications. Risk factors for placenta previa include previous cesarean section, multiple births, advanced maternal age, previous history of placenta previa, previous uterine surgery, and smoking history. The prevalence of pregnancies has increased due to rising cesarean section rates and increasing maternal age. Intraoperative anesthesia management and other interventions have been proposed in studies to control bleeding in patients with anterior prevascular who are expected to experience massive bleeding and require blood transfusion (Park, et al., 2020). In this work, 7 patients were stillborn, 1 patient underwent hysterectomy, and 30 patients had negative urine protein. There were 38 patients who underwent general anesthesia in the lower uterine segment cesarean section, and there were 8 cases of flat birth. The way of parturient delivery is also different. Some women choose the cesarean section of the lower uterine segment under the combined spinal-epidural joint, and there is also the cesarean section of the lower uterine segment under general anesthesia. Different conditions of different mothers can also be personalized for the choice of anesthesia.
D-D is an effective indicator to reflect the efficacy of anticoagulation. PT refers to the time required for the addition of excess tissue thromboplastin and calcium ions to platelet-deficient plasma, the conversion of prothrombin into thrombin, and the coagulation of the plasma. APTT is a more sensitive screening test for the intrinsic coagulation system. Fbg is a protein that can be dissolved in water. It is synthesized by the liver and has coagulation function. It is also an important substance in the process of coagulation and thrombosis. The mean values of prenatal coagulation indexes in this work were PT (12.43 ± 1.15) s, APTT (27.15 ± 2.65) s, Fbg (2.78 ± 1.12) g/L, and D-D (0.37 ± 0.21) ng/mL. RBC, Hct, and Plt are routine blood indicators. This work compared the prenatal and postpartum maternal Hb, RBC, Hct, and Plt after blood transfusion and before blood transfusion, the changes of each index were significantly different (P < 0.05). The coagulation indexes before and after blood transfusion were compared, the changes of PT, APTT, Fbg, and D-D were significantly different after maternal blood transfusion and before blood transfusion (P < 0.05). Risk factors for blood transfusion in the cesarean section include placenta previa, placental abruption, emergency caesarean section, appointment status, multiple pregnancy, and preoperative hematocrit, while previous caesarean sections did not increase the risk of blood transfusion (Iqbal, et al., 2022). In a study by Kvalvik et al, women with SSI were almost three times more likely to be obese before pregnancy and four times more likely to have pre-existing psychiatric disorders (OR 4.4, 95%) compared with women without SSI, receiving Five-fold increased likelihood of blood transfusion (Kvalvik, et al., 2021). Signs of infection during labor are a mildly significant risk factor for surgical site infection. The emergency cesarean section is an important risk factor for surgical site infections. Pre-pregnancy obesity, pre-existing psychiatric disorders, and blood transfusion during or after delivery were independent risk factors for surgical site infections. Signs of infection during labor are a negligible risk factor. Women with one of these risk factors should be carefully monitored and evaluated for signs of postpartum infection.
In clinical practice, the occurrence of postoperative complications of some diseases increases the difficulty of perioperative care of patients, and the length of hospitalization of patients increases, which will also lead to a corresponding substantial increase in hospitalization costs. The ability to identify influencing factors and predict the risk of postoperative complications in patients with liver cancer can help physicians make better clinical decisions. The process of machine learning is also the process of selecting suitable feature variables. According to the characteristics of the features, important features are selected (Zeng, et al., 2021). Many diseases in the clinic use machine learning to facilitate the diagnosis and treatment of diseases. The main trends among different types of supervised machine learning algorithms identified by Uddin et al. show that machine learning has good accuracy for disease prediction (Uddin, et al., 2019). Jhee et al. applied machine learning to preeclampsia, which can effectively predict late-onset preeclampsia (Jhee, et al., 2019). Adler et al. showed that machine learning improved risk prediction for heart failure (Adler, et al., 2020). The machine learning algorithm predicts the prediction and evaluation of blood transfusion in cesarean section and the risk factor analysis of hypothermia during recovery from anesthesia. The five important predictors of blood transfusion in cesarean section are preoperative hemoglobin, expected operation time, uterine weakness, placenta previa, and ASA classification (Ren, et al., 2022). Machine learning models can provide accurate individual predictions for patients, with good prediction performance and promising clinical application prospects. A study by Albright et al showed that regression models including variables available when the cesarean section accurately predicted the need for intraoperative or postoperative blood transfusion (Albright, et al., 2019). In this work, the machine learning model was used to process the data to obtain the best model of blood transfusion, and the test machine was used to evaluate the actual effect of the model. The predicted value of this work reached more than 95%, which can show a good prediction model. There is no fixed equation in machine learning itself. After inputting information into the model for each patient, the individualized intraoperative blood transfusion risk value can be obtained. The machine learning model is adopted to analyze the risk factors of blood transfusion in cesarean section, including placenta accreta, placental abruption, placental adhesion, polyhydramnios, pregnancy-induced hypertension, uterine atony, uterine fibroids, cesarean section, and dangerous placenta previa. It can also be known that with the increase of gestational age, the risk factors for blood transfusion during cesarean section surgery may gradually decrease. With gestational age less than 34 weeks, preterm birth may be a risk factor for intraoperative blood transfusion.