Comparison of MOLES and MelAInoma for Differentiating Small Choroidal Melanomas from Nevi

<b>Background:</b> Early identification of small choroidal melanomas is important, as metastatic risk increases with tumor size. However, distinguishing small melanomas from benign choroidal nevi is challenging and may lead to unnecessary referrals and overtreatment. Both the MOLES scoring system and the deep learning algorithm Mel<i>AI</i>noma have been developed to support assessment of pigmented choroidal lesions in non-expert settings. This study aims to compare the association between MOLES and Mel<i>AI</i>noma scores and to assess their relative association with expert melanoma versus nevus diagnosis. <b>Methods:</b> In this retrospective cohort study, 86 patients with small pigmented choroidal lesions (29 melanomas and 57 nevi) diagnosed at a national ocular oncology referral center were included. MOLES scores were assigned by ocular oncologists based on multimodal examination, whereas Mel<i>AI</i>noma scores were generated solely from color fundus photographs. Associations between scores were assessed using linear regression and the Jonckheere-Terpstra test. Univariable and multivariable binary logistic regression was used to evaluate associations with melanoma diagnosis. <b>Results:</b> Mel<i>AI</i>noma scores increased monotonically with higher MOLES categories (<i>p</i> = 0.0001). Linear regression showed a statistically significant association between MOLES and Mel<i>AI</i>noma scores, but with substantial dispersion (R² = 0.16). In univariable logistic regression, both MOLES and Mel<i>AI</i>noma scores were associated with increased odds of melanoma diagnosis. Mel<i>AI</i>noma showed a stronger association with diagnosis than MOLES (R² = 0.38 vs. 0.27). In multivariable analysis including both scores, each remained independently associated with melanoma diagnosis. <b>Conclusions:</b> Both MOLES and Mel<i>AI</i>noma are effective for differentiating small choroidal melanomas from nevi. Although the scores are statistically associated, they capture partly distinct information. Mel<i>AI</i>noma demonstrates slightly stronger association with melanoma diagnosis and provides fully reproducible output, supporting its role as a complementary aid in lesion triage.