Rotating default coordinate axes orientation using mplot3d

Question

I am plotting data from 3D imaging where the reference coordinate system is unconventionally oriented. Specifically, for a patient lying supine (face up) the +X direction points to the patient's left, +Y points toward the floor and +Z points in the direction of their head. It's still a right handed coordinate system. If I use ax.view_init() to set a viewpoint that gets close to this orientation I find that interactive manipulation works horribly. My solution so far is to pass my data points through a rotation matrix to map it into this orientation in world space and plot the result. I could also call ax.plot(-z, x, -y) to achieve the same result. This is functional but makes for confusing code and I've found it to be error prone.

Is there a way to embed my transformation matrix into the mplot3d processing chain so this automatically occurs? The documentation is still a bit incomplete for the transforms that it does perform and I haven't been able to determine where to start, although I think it may be possible.

To be a little more concrete, here is code that plots conventionally on the left and in the orientation I would prefer on the right:

from mpl_toolkits.mplot3d import axes3d
import matplotlib.pyplot as plt
import numpy as np

fig = plt.figure(figsize=(8,4))
ax1 = fig.add_subplot(121, projection='3d')
ax2 = fig.add_subplot(122, projection='3d')

X, Y, Z = axes3d.get_test_data(0.05)

pts = np.vstack((X.flatten(), Y.flatten(), Z.flatten(), np.ones_like(X.flatten())))
xf = np.array([
    [0., 0.,-1., 0.],
    [1., 0., 0., 0.],
    [0.,-1., 0., 0.],
    [0., 0., 0., 1.]
])
pts = xf @ pts
xt = pts[0].reshape((X.shape))
yt = pts[1].reshape((Y.shape))
zt = pts[2].reshape((Z.shape))

ax1.plot_surface(X, Y, Z, rstride=10, cstride=10)
ax2.plot_surface(xt, yt, zt, rstride=10, cstride=10)
for ax in (ax1, ax2):
    for ff in (ax.set_xlim, ax.set_ylim, ax.set_zlim):
        ff(-50, 50)