Low-redshift strong-lensing galaxies can provide robust measurements of the stellar mass-to-light ratios in early-type galaxies (ETG), and hence constrain variations in the stellar initial mass function (IMF). At present, only a few such systems are known. Here, we report the first results from a blind search for gravitationally-lensed emission line sources behind 52 massive $z$ $<$ 0.07 ETGs with MUSE integral field spectroscopy. For 16 galaxies, new observations were acquired, whilst the other 36 were analysed from archival data. This project has previously yielded one confirmed galaxy-scale strong lens (J0403-0239) which we report in an earlier paper. J0403-0239 has since received follow-up observations, presented here, which indicate support for our earlier IMF results. Three cluster-scale, and hence dark-matter-dominated, lensing systems were also discovered (central galaxies of A4059, A2052 and AS555). For nine further galaxies, we detect a singly-imaged but closely-projected source within 6 arcsec (including one candidate with sources at three different redshifts); such cases can be exploited to derive upper limits on the IMF mass-excess factor, $alpha$. Combining the new lens and new upper limits, with the previously-discovered systems, we infer an average $langle alpha rangle$ = 1.06 $pm$ 0.08 (marginalised over the intrinsic scatter), which is inconsistent with a Salpeter-like IMF ($alpha$ = 1.55) at the 6$sigma$ level. We test the detection threshold in these short-exposure MUSE observations with the injection and recovery of simulated sources, and predict that one in twenty-five observations is expected to yield a new strong-lens system. Our observational results are consistent with this expected yield.