Reply to “Comparing Accuracy Across Nonequivalent Protocol Spaces in Artificial Intelligence–Based CT and MRI Protocoling”