Wendelstein 7-X (W7-X) is a large, superconducting, optimized stellarator being operated in Greifswald, Germany. One of the original optimization criteria (Grieger et al 1992 Phys. Fluids B: Plasma Physics 4 2081) is good fast-ion confinement, mainly at increased plasma beta (Drevlak et al 2014 Nucl. Fusion 54 073002). This optimization applies mostly to deeply trapped collisionless fast ions. However, it has been shown in the past (Lazerson et al 2024 Phys. Plasmas 31 072506) that the current NBI system at W7-X does not populate these deeply trapped particle orbits in most magnetic configurations. Additionally, the maximum NBI injection energy is limited to 55 keV in Hydrogen which makes the particles not sufficiently collisionless. On the other hand, generating fast ions with ion-cyclotron-resonance heating (ICRH) does not suffer from such an energy limitation. In principle, the ICRH system can also be more flexible in other regards: By choosing the ICRH frequency appropriately, the resonance can be shifted away from the usual region of high magnetic field strength and placed deliberately into a region of low field strength which favours the generation of deeply trapped fast ions. In this contribution we explore such ICRH schemes for generating deeply trapped fast ions in W7-X using the SCENIC code package (Jucker et al 2011 Comput. Phys. Commun. 182 912). It is found that ideal target plasmas for proving the success of the optimization experimentally have a high temperature, the magnetic configuration should have a large magnetic mirror and the antenna frequency should be reduced from its nominal value to place the resonance in the triangular plane. Not all of these requirements are compatible with the systems as currently installed at W7-X and also detecting the fast ions may be an experimental challenge.