Not exactly spectroscopy, but microscopy is common for detecting pre-cancerous tissue as part of the standard Pap Test given to women at regular intervals throughout their life:
https://en.wikipedia.org/wiki/Pap_test
This is typically done by a team of cytologists (with pathologists confirming results, particularly in abnormal cases), but is beginning to have automated screening (with human oversight).
One of the problems with any form of screening is that the potentially problematic tissue needs to be observed, which is typically quite an invasive procedure. A potential melanoma is usually cut out prior to analysis, and I expect that any procedure involving a scanning electron microscope will also require tissue removal prior to analysis.
A more interesting development in the area of cancer screening is in looking for cell-free genomic DNA in blood; blood tests are much better tolerated for screening purposes in comparison to other tissue removal. With high-throughput short-read sequencing, it's possible for some cancers to see DNA changes before their effects are detectable through histological methods.
Here's a review paper on cfDNA screening, which is is generally negative about the whole thing:
http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1002205
And one that is generally positive:
http://cancerres.aacrjournals.org/content/76/14_Supplement/LB-136.short