Application of artificial intelligence in glaucoma diagnosis and treatment / 国际眼科杂志(Guoji Yanke Zazhi)

@#Glaucoma is the first irreversible eye disease leading to blindness of the world. Due to its insidious and progressive nature, early diagnosis and monitoring of glaucoma progression is very important in clinical practice. Artificial intelligence(AI)is developing rapidly in the medical field. The research and application of AI and its subsets of machine learning(ML)and deep learning(DL)in glaucoma are becoming increasingly mature, which promotes human understanding of glaucoma, greatly improves the accuracy and efficiency of glaucoma screening and diagnosis, and greatly reduces the cost of examination. Using AI technology for early screening and diagnosis of glaucoma patients can reduce the risk of visual impairment of patients, and second, it can predict the progression of glaucoma and design personalized treatment plans, so as to improve the prognosis of patients. This paper summarizes the new progress of AI in glaucoma screening, diagnosis, and prognosis, the clinical difficulties and challenges, and prospects the advantages and future development trends of AI in glaucoma.

Analysis of prescription and rationality of anti-hypertensive medication among community health centers in Beijing / 中华心血管病杂志

Objective: Explore the usage of anti-hypertension drugs and the rationality of hypertension prescription among the primary health centers in Dongcheng District, Beijing. Method: This cross-sectional and retrospective study was applied to analyze the hypertension prescriptions from the 8 community health centers in Dongcheng District. The anatomical, therapeutic and chemical classification (ATC) codes were used to determine the drug category. ATC information was used to filter data containing antihypertensive drugs, and group the number and proportion of ATC categories. The type of drug was judged by its generic name. According to the diagnosis information in the prescription, the prescription containing the Western medicine diagnosis of hypertension was screened out. The comorbidities of hypertension in the study included 7 types of diseases including diabetes, chronic kidney disease, coronary heart disease, heart failure, atrial fibrillation, stroke, and dyslipidemia. The analysis of prescription rationality included rationality of combination medication, rationality of drug dosage and rationality of drug price. The agreed daily dose (DDD) method was used to analyze the rationality of drug dosage. The drug utilization index (DUI) was used as a quantitative indicator to estimate the rationality of medication, and overdose was expressed by DUI>1. The reasonableness of the drug price was judged based on the price of the drug and whether it was a drug in the "4+7" plan. Results: A total of 658 140 prescriptions were extracted as the final data set, involving 7 categories and 60 commonly used anti-hypertensive drugs, and the corresponding cost of medication was ￥96.58 million. Drugs were prescribed according to comorbidities, and the choice followed the international guidelines. Calcium channel blockers (CCB) were the most prescribed drugs in the prescriptions of patients with comorbidities, and α-adrenergic receptor antagonists were the least prescribed drugs. The proportion of diuretics prescribed in hypertensive patients complicating with heart failure was 21.17% (505/2 385), which was much higher than that of patients complicating with other comorbidities (P<0.05). The proportion of diuretics prescribed in hypertension patients complicating with dyslipidemia was lower than that of patients with other comorbidities (2 639 (0.94%), P<0.05), and β-blockers (BB) or angiotensin Ⅱreceptor blockers (ARB) were more likely to be selected (BB: 59 348 (21.08%), ARB: 51 356 (18.24%))in these patients. The proportion of BB in prescriptions for hypertension patients with chronic kidney disease was lower than that of patients with other comorbidities (P<0.05). The proportion of BB in prescriptions for hypertension patients with coronary heart disease was higher than that of other comorbidities (P<0.05). Hypertension patients with atrial fibrillation or stroke accounted for a higher proportion of CCB prescriptions (P<0.05). Single antihypertensive drug prescriptions accounted for the highest proportion, 61.19% (402 745/658 140). Two-combination prescriptions accounted for the highest proportion of combination prescriptions, 72.19% (184 392/255 395). CCB based two-combination prescriptions accounted for the highest proportion, 122 350(66.36%). ARB-based tri-combination prescriptions accounted for the highest proportion, 48 915(89.50%),followed by CCB based tri-combination prescriptions (44 732(81.85%)).There were 2 174 (0.33%) prescriptions with unreasonable combination therapies and DUI>1 were found in 48 out of 60 commonly used drugs. In all possible antihypertensive drugs, only 40.92% (109 227/266 993)followed the "4+7" plan. Conclusions: The anti-hypertensive agents from these prescriptions in the primary health centers are diverse, and the choice is generally complied with the guidelines, but some unreasonable situations existed, especially on the combined anti-hypertensive medication, overdose, and"4+7"plan is not followed completely.

Infection Prevention Strategy in Operating Room during Coronavirus Disease 2019 (COVID-19) Outbreak / 中国医学科学杂志(英文版)

A novel coronavirus that emerged in late 2019 rapidly spread around the world. Most severe cases need endotracheal intubation and mechanical ventilation, and some mild cases may need emergent surgery under general anesthesia. The novel coronavirus was reported to transmit via droplets, contact and natural aerosols from human to human. Therefore, aerosol-producing procedures such as endotracheal intubation and airway suction may put the healthcare providers at high risk of nosocomial infection. Based on recently published articles, this review provides detailed feasible recommendations for primary anesthesiologists on infection prevention in operating room during COVID-19 outbreak.