A study published in Frontiers in Medicine by investigators at The First Hospital of Putian in Fujian, China, developed MyoSpectrumNet, a deep learning framework designed to assess the full pathologic myopia (PM) spectrum from a color fundus photograph. The model combines 3 public fundus datasets—HPMI (4,011 images), PALM (1,200 images) and MMAC—into 1 multi-output framework using a shared ConvNeXt-T encoder with a source-conditioned adapter, separate heads for PM, MM and SE, and a lesion branch joined through gated fusion. On the 2 headline clinical tasks, the joint model kept pace with dedicated models: PM screening AUCs were 0.985 on HPMI and 0.993 on PALM, compared with 0.987 and 0.996 for the dedicated classifiers, and MM grading produced an identical quadratic weighted kappa of 0.907 for both approaches. Trained on HPMI and tested on PALM with no fine-tuning, it reached an AUC of 0.993; transfer in the opposite direction was weaker, at 0.969. The model used about 31 million parameters and roughly 25 ms per single-view inference on a GPU. SE estimation was weaker: the dedicated regressor achieved a mean absolute error of 1.006 D (95% CI, 0.873–1.147), against 1.259 D for My