logan_analysis
=========================================================

.. py:function:: petpal.kinetic_modeling.graphical_analysis.logan_analysis(tac_times_in_minutes: numpy.ndarray, input_tac_values: numpy.ndarray, region_tac_values: numpy.ndarray, t_thresh_in_minutes: float) -> numpy.ndarray

   Performs Logan analysis on given input TAC, regional TAC, times and threshold, considering
   non-zero values.

   This function is similar to :func:`logan_analysis_with_rsquared`, but avoids the issue of
   division by zero by only considering non-zero TAC values for the region TAC since it is in
   the denominator. If the number of non-zero indices is less than or equal to 2, or if the number
   of time points after the threshold is less than or equal to 2, the function returns an array of
   NaNs.

   :param tac_times_in_minutes: Array of times in minutes.
   :type tac_times_in_minutes: np.ndarray
   :param input_tac_values: Array of input TAC values.
   :type input_tac_values: np.ndarray
   :param region_tac_values: Array of ROI TAC values.
   :type region_tac_values: np.ndarray
   :param t_thresh_in_minutes: Threshold time in minutes. Line is fit for all values
                               after the threshold.
   :type t_thresh_in_minutes: np.ndarray

   :returns: *np.ndarray* --

             Array of two elements - (slope, intercept) of the best-fit line to the given
                 data.

   .. important::
       * The interpretation of the returned values depends on the underlying kinetic model.
       * We assume that the input TAC and ROI TAC values are sampled at the same times.