Objective:To investigate the differences in ametropia among school-age children and between Han and Uygur ethnic groups in the main urban area of Kashgar, and to understand the epidemiological characteristics of myopia among children in Kashgar.Methods:A cross-sectional study was conducted.Random stratified sampling method was used to select the first to third grade students from 6 primary schools in the main urban areas of Kashgar, Xinjiang, from April to June 2018.Visual acuity was examined by the international standard visual acuity chart and refraction was detected by the automatic computer refractometer without cycloplegia.The subjects were classified into low vision (uncorrected visual acuity<0.8 in either eye), hyperopia≥2.00 D, myopia≥-1.00 D, astigmatism≥1.00 D and anisometropia≥1.00 D according to examination results.The percentage of different refraction states and percentage of different ethnic groups among myopic students in the three grades were analyzed.The study protocol was approved by an Ethics Committee of The First People's Hospital in Kashgar ([2021] QSY No.81).Results:A total of 6 108 students were enrolled, among which, there were 3 119 males and 2 989 females, 3 395 Uygur and 2 713 Han people, 2 016 first grade, 2 155 second grade and 1 937 third grade students.The detection rate of low vision was 13.5% (825/6 108) in the whole, 21.5% (584/2 713) in the Han people and 7.1% (241/3 395) in the Uygur people, 8.8% (176/2 016) in the first grade, 11.8% (254/2 155) in the second grade and 20.4% (395/1 937) in the third grade.The detection rate of myopia was 9.9% (606/6 108) in the whole, 17.6% (477/2 713) in the Han people and 3.8% (129/3 395) in the Uygur people, 4.4% (89/2 016) in the first grade, 8.1% (174/2 155) in the second grade and 17.7% (343/1 937) in the third grade.There were significant differences in percentage of low vision, myopia, astigmatism and anisometropia among different grades and between the two nationalities (all at P<0.001), and no significant difference in the percentage of hyperopia was found (P>0.05).Conclusions:The percentage of visual abnormalities and myopia in the first to third grade children in Kashgar is lower than the national average, but the percentage is gradually increasing with age.The percentage of visual abnormalities and myopia among Han pupils is close to the national average, and far higher than that of Uygur pupils.

Objective:To explore the value of artificial intelligence(AI)-assisted diagnosis platform combined with optical coherence tomography(OCT)in diagnosing blinding eye diseases,so as to provide effectively scientific basis for favorable prognosis of patients with blinding eye diseases.Methods:A total of 72 patients with visual impairment who admitted to the outpatient of hospital were selected.All patients received the detection of AI-assisted diagnosis platform combined with OCT diagnosis.The final diagnosis result of the detection of doctor combined with the relevant ophthalmic examination was used as the"gold standard"to assess respectively the consistence among single AI-assisted diagnosis platform,single OCT,the combination of them and the"gold standard",as well as the sensitivity,specificity and accuracy of them in diagnosing the blinding eye diseases.Results:For the 72 patients,the detection rate of the detection result of doctor combined with the relevant ophthalmic examination was 27.78%(20/72)for blinding eye diseases,and the detection rate of that was 72.22%(52/72)for non-blinding eye diseases.The consistency between AI-aided platform diagnosis and the"gold standard"was general in diagnosing the blinding eye diseases(kappa=0.530).The consistency between OCT and the"gold standard"was favorable in diagnosing that(kappa=0.611).The consistency between AI-assisted platform combined with OCT and the"gold standard"was better(kappa=0.799).The specificity,sensitivity,positive predictive value,negative predictive value,diagnostic accuracy of AI-assisted platform combined with OCT diagnosis were respectively 92.31%,90.00%,81.82%,96.00%and 91.67,and the diagnostic value of the combination was higher than that of single AI-assisted platform and that of single OCT for all of above these indicators.Conclusion:Both the AI-assisted diagnosis platform and OCT can detect blinding eye diseases,and the combined detection of them has higher diagnostic value.

ObjectiveTo evaluate the Chinese guidelines and consensus of rehabilitation medicine published in the medical journals in 2022 using Scientific, Transparent and Applicable Rankings (STAR). MethodsGuidelines and consensus which were developed by Chinese institutions or led by Chinese scholars were retrieved in databases of CNKI, Wanfang Data, CBM, Chinese Medical Journal Network, PubMed and Web of Science, in 2022, followed by screening for rehabilitation medicine field. The literature were rated with STAR. ResultsSeven guidelines and eleven consensuses were included. The STAR scores ranged from 11.7 to 69.6, with a median score of 25.9 and mean score of 28.3. There was a significant difference in the total score between guidelines and consensus (U = 12.000, P = 0.014). The score ratio was high in the domains of recommendations (73.6%), evidence (39.5%) and others (33.3%), while it was low in the domains of protocol (1.4%), clinical questions (12.5%) and conflicts of interest (13.9%). The score ratio was high in the items of listing the institutional affiliations of all individuals involved in developing the guideline (94.4%), identifying the references for evidence supporting the main recommendations (94.4%), indicating the considerations (e.g., adverse effects) in clinical practice when implementing the recommendations (88.9%), and making the recommendations clearly identifiable, e.g., in a table, or using enlarged or bold fonts (75%); and it was low in the items of describing the role of funder(s) in the guideline development (0), indicating information about the evaluation and management of conflicts of interest (0), providing tailored editions of the guidelines for different groups of target users (0), presenting the guideline or recommendations visually, such as with figures or videos (0), providing details of the guideline protocol (2.8%), assessing the risk of bias or methodological quality of the included studies (2.8%), describing the responsibilities of all individuals or sub-groups involved in developing the guideline (5.6%), indicating how the clinical questions were selected and sorted (5.6%), formating clinical questions in PICO or other formats (5.6%), making the guideline accessible through multiple platforms (5.6%), and declaring that the funder(s) did not influence the guideline's recommendations (8.3%). ConclusionThe quality of current clinical practice guidelines and consensus of rehabilitation medicine is poor, which should be developed in accordance with the relevant standards.