Magnetic Particle Imaging (MPI), a molecular imaging modality that images biocompatible superparamagnetic iron oxide tracers, is well-suited for clinical angiography, in vivo cell tracking, cancer detection, and inflammation imaging. MPI is sensitive and quantitative to tracer concentration, with a positive contrast that is not attenuated or corrupted by tissue background. Like other clinical imaging techniques, such as computed tomography, magnetic resonance imaging, and nuclear medicine, MPI can be modeled as a linear and shift-invariant system with a well-defined point spread function (PSF) capturing the system blur. The key difference, as we show here, is that the MPI PSF is highly dependent on scanning parameters and is anisotropic using only a single-imaging trajectory. This anisotropic resolution poses a major challenge for clear and accurate clinical diagnosis. In this paper, we generalize a tensor imaging theory for multidimensional x-space MPI to explore the physical source of this anisotropy, present a multi-channel scanning algorithm to enable isotropic resolution, and experimentally demonstrate isotropic MPI resolution through the construction and the use of two orthogonal excitation and detector coil pairs.