OBJECTIVE: To explore the clinical observation of white eye distribution characteristics of hyperlipidemic patients based on artificial intelligence digital eye diagnosis technology.

METHODS: One hundred and fifty subjects were examined in the outpatient and inpatient departments of Guang'anmen Hospital of the China Academy of Traditional Chinese Medicine from 01 February 2022 to 01 February 2023, including 80 cases in the hyperlipidemic patient (HLP) group and 70 cases in the normal lipid level patient (NC) group. The two groups were collected and extracted by the artificial intelligence visual diagnostic instrument and analyzed by the MyEyeD-10 white eye shadowless imaging health intelligence analysis system. Finally, SPSS 26.0 (Version X; IBM Corp., Armonk, NY, USA) was used for statistical processing.

RESULTS: Significant differences were noted in the scores of "spot" and "foggy" features between the two groups. Between groups, the "spot" feature score of the white eye morphology in the HLP group (11.07 ± 3.22) was higher than that in the NC group (7.50 ± 4.11) (P <0.01). Moreover, the "foggy" feature score of the eye morphology in the HLP group (8.37 ± 2.25) was higher than that in the NC group (P <0.01), higher than that of the NC group (5.72 ± 1.21) (P <0.05). There were significant differences in the "A" (stomach), "B", "O" (spleen), and "M" (liver) eye-contact region scores between the two groups, and the "B", "O" (spleen) and "M" (liver) eye-contact region scores were significantly different. The scores of the white eye channel region in the HLP group were significantly higher than those in the control group, with the "A", "B", and "O" regions (P <0.01), "M" region (P <0.01), "A", "B", "O", and "M" region (P <0.01). "M" zone (P <0.05). The scores of "dull red" and "yellow" features were significantly different, and the scores of "dull red" and "yellow" colors of the white eye choroid in the HLP group were significantly higher than those in the HLP group. The scores of "dull red" and "yellow" were significantly higher in the HLP group than in the NC group (P <0.01).

CONCLUSION: The morphological features of the white eye ocular image, the white eye chakra's color, and the bulbar conjunctiva's vascular zoning are closely related to hyperlipidemia. Importantly, these provide a reference for the objectivity and precision of the identification of Chinese medicine by looking at the eyes.