With the popularity and development of artificial intelligence (AI), disease screening systems based on AI algorithms are gradually emerging in the medical field. Such systems can be used for primary screening of diseases to relieve the pressure on primary health care. In recent years, AI algorithms have demonstrated good performance in the analysis and identification of lesion signs in the macular region of fundus color photography, and a screening system for fundus lesion signs applicable to primary screening is bound to emerge in the future. Therefore, to standardize the design and clinical application of the macular region lesion sign screening systems based on AI algorithms, the Ocular Fundus Diseases Group of Chinese Ophthalmological Society, in collaboration with relevant experts, has developed this guideline after investigating issues, discussing production evidence, and holding guideline workshops. This guideline aims to establish uniform standards for the definition of the macular region and lesion signs, AI adoption scenarios, algorithm model construction, datasets establishment and labeling, architecture and functions design, and image data acquisition for the screening system to guide the implementation of the screening work.
ObjectiveTo observe the clinical characteristics of children with leukemia combined with ocular fundus lesions, and analyze the related risk factors. MethodsA retrospective clinical study. A total of 126 children with leukemia who were diagnosed and underwent fundus examination in Hunan Children's Hospital, from February 2022 to June 2025 were included. Posterior pole fundus images of both eyes were obtained using a handheld non-mydriatic fundus camera. Optical coherence tomography and fluorescein fundus angiography were performed when necessary. Outcomes were analyzed on a per-patient basis. Fundus involvement was defined as the presence of abnormalities in either eye, and the main lesion type was classified according to the more severe eye. According to the presence or absence of fundus lesions, the patients were divided into the lesion group (38 cases, 30.15%, 38/126) and non-lesion group (88 cases, 69.85%, 88/126). Differences in age, sex distribution, hematological parameters, and leukemia subtype were compared between the two groups. Binary logistic regression was used to identify independent risk factors. ResultsAmong the 126 children, 66 were male and 60 were female. The mean age was (7.62±3.13) years. Seventy-eight patients had acute lymphoblastic leukemia, and 48 patients had acute myeloid leukemia , including 11 patients with acute promyelocytic leukemia (APL). Compared with the non-lesion group, the lesion group had a significantly higher peripheral blood white blood cell count (WBC) level and significantly lower hemoglobin and platelet levels, with statistically significant differences (t=8.210, ?6.940, ?11.040; P＜0.05). Among the 38 patients with fundus lesions, retinal hemorrhage was observed in 28 cases (73.68%, 28/38), optic disc edema in 6 cases (15.79%, 6/38), cotton-wool spots in 3 cases (7.89%, 3/38), and choroidal lesions in 1 case (2.63%, 1/38). Multivariate logistic regression analysis showed that WBC count, per 10×109/L increase [odds ratio (OR)=2.13, 95% confidence interval (CI) 1.35-3.41, P=0.002], early-stage chemotherapy (OR=1.96, 95%CI 1.12-3.15, P=0.010), and APL subtype (OR=2.74, 95%CI 1.28-5.89, P=0.008) were independent risk factors for fundus lesions. ConclusionsFundus lesions are relatively common in children with leukemia and are predominantly characterized by hemorrhagic changes. Elevated WBC count, per 10×109/L increase, early-stage chemotherapy, and APL subtype indicate a high risk of fundus involvement.