Bit of an aside, but I'd love to know if said Deep Learning could possibly learn to detect whether a pneumonia was merely eosinophilic (no invader, just white c

Bit of an aside, but I'd love to know if said Deep Learning could possibly learn to detect whether a pneumonia was merely eosinophilic (no invader, just white cells gone wild.) I'm hoping that maybe a more general distribution would be characteristic of eosinophilic pneumonia, I suppose.