X-linked retinoschisis (XLRS) is a type of hereditary retinopathy with notable complications including disrupted outer retinal layers and cystoid fluid collections (CFC). However, the current method of using central subfield thickness (CST) of the retina to evaluate the extent of the progression of XLRS is faulty because it is time consuming and does not fully capture the size, distribution and severity of CFC. This study conducted by Hensman et al. aimed to train and evaluate a deep learning (DL) model that can detect and quantify CFC on SD-OCT scans in an efficient and objective manner. The DL model they used had been previously trained on OCT images from patients with RVO and AMD, but not XLRS. A set of 37 dense B-scans from 20 existing XLRS patients were given to both the DL model and 3 experienced graders (authors JH, YA, and HA) to determine the areas of the image affected by CFC. Both the mean Dice scores and ICC values were high between the model and the grader’s analyses, indicating a high level of agreement between the DL algorithm and clinical professionals’ performance. However, it was found that the model tended to overestimate the area of the retina affected by CFC, which may be attributed to the fact that the model had only previously been trained to recognize RVO and AMD. Additionally, another notable drawback is that this model was only given very high quality B-scans, so it is unclear whether or not it can function with the same accuracy if given lower-quality scans.