In the field of artificial intelligence (AI) medical imaging, data annotation is a key factor in all AI development. In the traditional manual annotation process, there are prominent problems such as difficult data acquisition, high manual labor intensity, strong professionalism and low labeling quality. Therefore, an intelligent multimodal medical image annotation system is urgently needed to meet the requirements of labeling. Based on the image cloud, West China Hospital of Sichuan University collected the multimodal image data of hospital and allied hospitals, and designed a multi-modal image annotation system through information technology, which integrated various image processing algorithms and AI models to simplify the image data annotation. With the construction of annotation system, the efficiency of data labeling in the hospitals is improved, which provides necessary data support for the AI image research and related industry construction in the hospital, so as to promote the implementation of artificial intelligence industry related to medical images in the hospital.
At present, the rapid integration and development of internet technology and medical services have made internet diagnosis and treatment an important part of medical services, and it is also an inevitable development trend of future diagnosis and treatment services. In order to meet the needs of patients for more timely, accurate and convenient medical treatment, West China Hospital of Sichuan University has innovated the internet diagnosis and treatment mode, adopted the innovative mode of diversified online services, pre-treatment mode, expert team mode, specialized medical consortium platform and whole process management, optimized medical resources and structure, promoted regional medical association linkage, and improved patients’ medical experience. The West China Internet Hospital of Sichuan University takes the whole process ecological closed-loop of “medical+health” as the goal, has greatly improved the efficiency and quality of diagnosis and treatment, which is of great significance for the positioning and development of internet hospital. This article will share the construction experience of West China Internet Hospital of Sichuan University.
Objective To comparatively analyze the rationality of emergency medication after Wenchuan earthquake with that after Lushan earthquake in West China Hospital of Sichuan University, based on the use of medicine of the victim’s disease spectrum. Method By using Excel, defined daily dose system (DDDs) and drug utilization index (DUI) were used as the evaluation indexes to analyze the rationality of emergency medication in West China Hospital of Sichuan University within one month after Wenchuan earthquake and Lushan earthquake. Results Within one month after Wenchuan and Lushan earthquake, there were 1 839 and 488 victims treated in the hospital, respectively. Within one month after the two earthquakes, the variation tendency of DDDs of drugs and number of victims was consistent, and the consistency was better in Lushan earthquake than that in Wenchuan earthquake. Among the 60 drugs which DDDs were ranked top five in their pharmacological class (top ten for antimicrobials) in Wenchuan earthquake, the majority of them were injections (injections vs. non-injections: 70.0% vs. 30.0%); the results showed that the medication (DUI=1) only accounted for 10.0%, the medication (DUI<1) accounted for 28.3%, which implied that the use of drugs was insufficient, the medication (DUI>1) accounted for 61.7%, which implied that drugs were overused; the average of DUI was 1.61. And in Lushan earthquake, injections also accounted for a larger proportion than non-injections (63.3%vs. 36.7%); the results showed that the medication (DUI=1) accounted for 15.0%, the medication (DUI<1) accounted for 38.3%, the medication (DUI>1) accounted for 46.7%; the average of DUI was 1.30. Conclusions Base on the DUI, we draw the conclusion that the rationality of emergency drug use and the timeliness of emergency drug supply were better in Lushan earthquake than those in Wenchuan earthquake. But the rationality of using the DUI, which is an evaluation index for normal conditions, to evaluate the emergency conditions still needs to be further verified.
Objective To analyze the costs of emergency medication in the West China Hospital within one month after Lushan earthquake based on actual medication of the victims’ disease spectrum. Methods We collected emergency medication data as evaluation index in the West China Hospital within one month after Lushan earthquake, including daily cost, cumulative percentage of pharmacology category, average cost per day/person, average cost per day/person when DUI=1, difference between average cost per day /person, and average cost per day/person when DUI=1, etc. Then, we input data using Excel software for statistically analyzing the costs of emergency medication within one month after the earthquake. Results During one month after the earthquake, the costs changed consistently with the number of victims, which implied the change of costs was rational. Injuries were classified into 6 categories and 12 kinds according to ICD-10. The costs of medication accounted for 71% of the total costs. Six kinds of illness accounted for 21%. Medication for injuries was classified into 3 categories and 18 kinds; average cost per day was 186.87 yuan and average cost per person was 1 702.70 yuan. Medication for illness was classified into 5 categories and 28 kinds; average cost per day was 38.96 yuan and average cost per person was 185.13 yuan. The mean value of average cost per day/person of injection was 14.52/5.08 times more than that of non-injection. Meanwhile, the mean value of average cost per day/person of imported medication was 7.10/5.28 times more than that of domestic medication. Conclusion The factors that impact the medication costs include: a) disease burden and traumatic conditions of the sick and wound; b) administration pathway (injection vs. non-injection); c) imported or domestic medication; and d) the rationality of taking DUI as evaluation index. When DUIgt;1, injection through the vein and imported medication take a larger share which increase the costs of medication. Emergency rescue package should mainly prepare for the injury. In order to ensure the timely, safe and effective medication, injection should be given priority to. We also should take into consideration crash-resistance/anti-quake package, reasonable dosage, convenience-to-use of drugs as well as the needs of the illness.
Objective To analyze the rationality of emergency medication in the West China Hospital after Lushan earthquake based on the actual drug use of earthquake victims. Methods We applied DDDs and DUI as evaluation index, input data using Excel software, and analyzed if the emergency medication was required for the injury/illness and the rationality of emergency medication. Results Earthquake victims mainly had trauma and wound infection and they were given antibiotics as main treatment one week after the earthquake. Drugs for the respiratory system and digestive system were mainly used in patients who mainly manifested as non-traumatic diseases and internal diseases four weeks after the earthquake. Among 49 kinds of drugs which could be calculated for the value of DUI, injection accounted for a larger proportion than non-injection (59.18% vs. 40.82%). The results showed that, the medication (DUI=1) only accounted for 12.24%; the medication (DUIlt;1, Mean=0.65) accounted for 38.78%, which implied insufficient drug use; the medication (DUIgt;1, Mean=1.77) accounted for 46.94%, which implied drug overuse. Conclusion Medication for every system is basically timely and rational when treating symptoms and causes in the West China Hospital within one month after Lushan earthquake. However, the rationality of emergency medications using DUI=1 under normal conditions still needs to be further verified.
Yushu earthquake bettered a record all round in ability and speed about emergency command, provisioning, medical rescue, epidemic prevention, rebuilding and policy ensuring. The article compared the achievements in medical rescue of Yushu earthquake and Wenchuan earthquake from Ministry of Health and the hospitals who took part in medical rescue within 3 months after Yushu earthquake in order to summarize the experience, form the standard and provide decision-making references.
Objective To analyze retrospectively the 1861 wounded inpatients and deaths in West China Hospital of Sichuan University after the Wenchuan earthquake, so as to provide reference for the improvement of emergency plans for the disaster of earthquake and the establishment of state-level regional medical centers. Methods The analysis was based on the data provided by the Department of Information of the Hospital up until July 23. The software of Microsoft EXCEL was used for data input, and SPSS 11.5 was used for statistical analysis. Results Up to July 23, 2728 cases from the disaster area have been treated in the hospital, of whom 872 were admitted into the emergency department and 1856 into the inpatient department (974 men, median age 43 years; 882 women, median age 46 years). Most patients were sent to the hospital within the first 2 weeks after the quake (82.4%), and the number of inpatients reached its peak on Day 8 after the quake (976 cases). The majority of the inpatients were discharged on Day 9 to Day 18 after the quake (60.2%). The wounded were mainly from Deyang, Aba Prefecture and Dujiangyan. The admission diagnoses were mainly fracture (54.84%), craniocerebral injury (9.81%) and thoracoabdominal injury (7.54%). There were totally 33 deaths, including 5 pre-hospital deaths, 1 emergency death and 27 inpatient deaths. Conclusion The development of an emergency plan for the medical rescue after an earthquake disaster is an essential step to enhance the emergent response capability, improve the scientific process of field triage, transport and transfer, and ensure the rational allocation and application of healthcare resources after any unexpected big disasters in the future.
ObjectiveTo evaluate the effectiveness and safety of tranexamic acid (TXA) combined with intraoperative controlled hypotension (ICH) for reducing perioperative blood loss in primary total hip arthroplasty (THA).MethodsThe clinical data of 832 patients with initial THA due to osteonecrosis of femoral head between January 2017 and July 2020 were retrospectively analyzed. All patients received TXA treatment, and 439 patients (hypotension group) received ICH treatment with an intraoperative mean arterial pressure (MAP) below 80 mm Hg (1 mm Hg=0.133 kPa) while 393 patients (normotension group) received standard general anesthesia with no special invention on blood pressure. There was no significant difference in age, gender, body mass index, American Society of Anesthesiologists (ASA) classification, basic arterial pressure, hip range of motion, internal diseases, preoperative hemoglobin (HB) and hematocrit (HCT), coagulation function, surgical approach, and TXA dosage between the two groups (P>0.05). The perioperative blood loss and blood transfusion, anesthesia and operation time, hospitalization stay, postoperative range of motion, and complications were recorded and compared between the two groups. The patients were further divided into MAP<70 mm Hg group (group A), MAP 70-80 mm Hg group (group B), and normotension group (group C). The perioperative blood loss and postoperative complications were further analyzed to screen the best range of blood pressure.ResultsThe intraoperative MAP, total blood loss, dominant blood loss, recessive blood loss, blood transfusion rate and blood transfusion volume, anesthesia time, operation time, and hospitalizarion stay in the hypotension group were significantly lower than those in the normotension group (P<0.05). The postoperative hip flexion range of motion in the hypotension group was significantly better than that of the normotension group (Z=2.743, P=0.006), but there was no significant difference in the abduction range of motion between the two groups (Z=0.338, P=0.735). In terms of postoperative complications, the incidence of postoperative hypotension in the hypotension group was significantly higher than that in the normotension group (χ2=6.096, P=0.014), and there was no significant difference in the incidence of other complications (P>0.05). There was no stroke, pulmonary embolism, or deep vein thrombosis in the two groups, and no patients died during hospitalization. Subgroup analysis showed that there was no significant difference in total blood loss, dominant blood loss, and recessive blood loss in groups A and B during the perioperative period (P>0.05), which were significantly lower than those in group C (P<0.05). There was no significant difference in blood transfusion rate, blood transfusion volume, and incidence of acute myocardial injury between 3 groups (P>0.05); the incidence of acute kidney injury in group A was significantly higher than that in group B, and the incidence of postoperative hypotension in group A was significantly higher than that in groups B and C (P<0.05), but no significant difference was found between groups B and C (P>0.05).ConclusionThe combination of TXA and ICH has a synergistic effect. Controlling the intraoperative MAP at 70-80 mm Hg can effectively reduce the perioperative blood loss during the initial THA, and it is not accompanied by postoperative complications.