@marc_alec, very sneaky question. This is a 2nd order ODE in the second order, but /0's method is within the scope and workable if we substitute everything to get rid of the constants.
@Chavi, highly unlikely that you'll have to do this in an exam [in fact, I'd say impossible]. You wouldn't really have apply a kinematics thing to this kind of 'general' case [unless they give you hints]. I will finish the rest of the question from where /0 has left it.
------[1]
--------[2]
Given that when x=0, y = 1 and d
2y/dx
2 = 4, we then have (by [1]) that 4 = k * 1, which implies k=4.
After integration, we have
----------[3]
Given that when x=0, y = 1, dy/dx = 2, thus by [3], 2^2 = 4 * 1^2 + D, which implies D = 0, [3] can thus be simplified to
^2 = 4y^2)

, taking only the positive branch of the square root since both y and dy/dx are positive.

, where C = 0 by substituting x=0, y=1.


, taking only the positive branch since y is positive.