Method for non-invasive monitoring of fluorescent tracer agent with diffuse reflection corrections
A method of monitoring a time-varying fluorescence signal emitted from a fluorescent agent from within a medium with time-varying optical properties is provided that includes providing a measurement data set that includes a plurality of measurement entries that include at least two measurements obtained from a patient before and after administration of the fluorescent agent. The measurements may include one or more of: a DR ex signal detected by an unfiltered light detector during illumination by excitatory-wavelength light from first region adjacent to the diffuse reflecting medium; a Flr signal detected by a filtered light detector during illumination by excitatory-wavelength light; and a DR em signal detected by the unfiltered light detector during illumination by emission-wavelength light. The method further includes identifying a post-agent administration portion of the measurement data set; and transforming each Flr signal to an IF signal representing a detected fluorescence intensity emitted solely by the fluorescent agent.
1. A method of monitoring a time-varying fluorescence signal emitted from a fluorescent agent from within a diffuse reflecting medium with time-varying optical properties, the method comprising:
providing a measurement data set comprising a plurality of measurement entries, each measurement data entry comprising at least two measurements obtained at one data acquisition time from a patient before and after administration of the fluorescent agent, the at least two measurements selected from:
a DR ex signal detected at a second region adjacent to the diffuse reflecting medium by an unfiltered light detector during illumination of the diffuse reflecting medium by excitatory-wavelength light from a first region adjacent to the diffuse reflecting medium;
an Flr signal detected at a third region adjacent to the diffuse reflecting medium by a filtered light detector during illumination of the diffuse reflecting medium by excitatory-wavelength light from the first region; and
a DR em signal detected at the second region by the unfiltered light detector during illumination of the diffuse reflecting medium by emission-wavelength light from the first region;
identifying a post-agent administration portion of the measurement data set; and
transforming each Flr signal of each measurement data entry within the post-agent administration portion of the measurement data set to an IF signal representing a detected fluorescence intensity emitted solely by the fluorescent agent from within the diffuse reflecting medium, wherein transforming comprises combining the at least two measurements according to a transformation relation comprising a mathematical equation converting Flr to IF.
2. The method of claim 1 , wherein the transformation relation consists of Eqn. (20):
IF
=
Flr
DR
ex
,
meas
k
ex
,
meas
DR
em
k
em
DR
em
,
filtered
k
em
,
filtered
;
Eqn
.
(
20
)
wherein IF represents a fluorescence intensity emitted solely by the fluorescent agent, and k ex,meas , k em , and k em,filtered are exponents of terms DR ex , DR em , and DR em,filtered , respectively.
3. The method of claim 2 , wherein k ex,meas , k em , and k em,filtered are previously determined from a prior analysis of a prior measurement data set.
4. The method of claim 2 , wherein k ex,meas , k em , and k em,filtered are determined by a global error map method comprising:
forming three vectors of proposed exponent values, each of three vectors comprising a plurality of proposed values for k ex,meas , k em , and k em,filtered , respectively,
transforming each Fir signal of each measurement data entry within the post-agent administration portion of the measurement data according to the transformation relation using each combination of proposed values from the three vectors to form a plurality of transformed data measurement sets;
performing a single-exponential curve fit over at least a portion of the measurement data entries for each of the plurality of the transformed data measurement sets to obtain a plurality of curve-fit errors, each curve-fit error corresponding to one combination of the proposed exponent values from the three vectors;
assembling an error map comprising at least a portion of the plurality of curve-fit errors mapped to a volume defined by two or more orthogonal axes, each orthogonal axis comprising a range of proposed exponent values from one of the three vectors;
identifying a minimum curve-fit error within the error map; and
selecting the proposed exponent values corresponding to the minimum curve-fit error for use in Eqn. (20).
5. The method of claim 4 , wherein each of the plurality of curve-fit errors consists of a normalized root-mean-square fitting error of the single-exponential curve-fit.
6. The method of claim 1 , wherein the transformation relation consists of a linear regression model formed with predictor variables DR ex , DR em , and DR em,filtered using a low-variability portion of the plurality of measurement data entries characterized by a curve-fit error falling below a threshold value for a single-exponential curve-fit of the measurement data entries of the low variability portion.
7. The method of claim 6 , wherein the linear regression model is extrapolated to measurement data entries outside of the low variability region.
8. The method of claim 1 , further comprising subtracting a baseline value for Flr from each of the Flrvalues from the plurality of the measurement data sets prior to transforming each Flrsignal of each measurement data entry within the post-agent administration portion of the measurement data set to an IF signal.
9. The method of claim 1 , wherein the at least two measurements are further selected from a DR em,filtered signal detected at the third region by the filtered light detector during illumination of the diffuse reflecting medium by emission-wavelength light from the first region.
10. The method of claim 2 , wherein any one or more of k ex,meas , k em , and k em,filtered is equal to 0.