RESUMO
Objective To explore the effect of the orthokeratology lenses on the control of different types of aniso-metropia in myopic children.Methods A total of 99 myopic children aged 8 to 16 years who got the orthokeratology len-ses at the Department of Ophthalmology,Heping Hospital Affiliated to Changzhi Medical College from September 2020 to November 2022 with complete data were included.These children were divided into the simple myopic anisometropia group(monocular myopia,binocular diopter difference ≥ 1.00 D,n=39)and the compound myopic anisometropia group(binoc-ular myopia,binocular diopter difference ≥ 1.00 D,n=60).The children with higher anisometropia(binocular diopter difference ≥ 2.50 D)in the two groups were set as the high anisometropia subgroup(n=18 and 29,respectively),and chil-dren with lower anisometropia(1.00 D≤ binocular diopter difference＜2.50 D)were set as the low anisometropia subgroup(n=21 and 31,respectively).In each group,eyes with a higher diopter were set as the high diopter eyes,and the contra-lateral eyes with a lower diopter were set as the low diopter eyes.Diopter,corneal topography,intraocular pressure,cor-neal endothelium and axial length of children in the two groups were examined and recorded.The changes in axial length before and after wearing orthokeratology lenses for 1 year were compared between the two groups,analyzing the correla-tion between the degree of anisometropia and changes in the binocular axial length.Results After wearing orthokeratolo-gy lenses for 1 year,children in both groups had an increase in the axial length with a lower increase in the axial length of the high diopter eyes compared to the low diopter eyes;before and 1 year after wearing orthokeratology lenses,the axial length of high diopter eyes was greater than that of the low diopter eyes in both groups,and the differences were statistical-ly significant(all P＜0.05).Both groups of children showed a decrease in the binocular axial length difference after wear-ing the orthokeratology lenses for 1 year;before and 1 year after wearing orthokeratology lenses,the binocular axial length difference of children in the simple myopic anisometropia group was greater than that in the compound myopic anisometro-pia group,and the differences were statistically significant(t=4.903 and 2.670;both P＜0.05).The changes in binocular axial length difference before and after wearing the orthokeratology lenses of children in the high anisometropia subgroup and low anisometropia subgroup of the simple myopic anisometropia group were greater than those in the high anisometro-pia subgroup and low anisometropia subgroup of the compound myopic anisometropia group,respectively,and the differ-ences were statistically significant(both P＜0.05).In the simple myopic anisometropia and compound myopic anisometro-pia groups,the degree of anisometropia was positively correlated with the binocular axial length changes before and 1 year after wearing the orthokeratology lenses(r=0.423 and 0.510,both P＜0.05).Conclusion Orthokeratology lenses can effectively reduce the difference in binocular axial length of children with myopic anisometropia,and their control effect on simple myopic anisometropia is better than that of compound myopic anisometropia.
RESUMO
Artificial intelligence (AI) is an emerging science and technology that studies and develops theories, methods, technologies, and application systems for simulating and expanding human intelligence. AI has made great breakthroughs in the field of intelligent medicine, and has shown great potential in the diagnosis and treatment of diabetic retinopathy (DR), retinopathy of prematurity, and other fundus diseases. A number of clinical trials on the application of AI technologies to DR screening have been carried out in the domestic and overseas, which not only have a high accuracy rate, but also save doctors' reading time and reduce the burden of society, medical work and patients. However, due to the lack of evaluation system for DR intelligent diagnosis technology, the accuracy of AI system still lacks of big data verification. Secondly, most of the color fundus photographs are taken in the posterior 45°, which only show the most vulnerable areas, making some lesions undetectable. In addition, the current DR screening system has not yet been applied to the clinic, most of which are in the stage of prospective research and trials. There are still many obstacles from the environment to the hospital or the clinic. Doctors cannot use real patient data to evaluate the AI system, so it is not popular in clinical practice. In the future, DR screening algorithms and diagnostic models can be further improved and established to make DR AI screening more accurate.