Removal of distortion from indicator dilution curves with analog computer.

• In the usual procedure for determination of indicator dilution curves, the sensing or measuring device is placed at some distance from the sampling site. The fluid is transported by a sampling system, usually a cathe-ter, from the withdrawal site to the sensing point. This sampling system always introduces some distortion in the time-concentration curve, so that the recorded curve is not the same shape as the in situ concentration-time relation at the withdrawal site. These sampling errors have been described by several workers. 1-2 ' 3i * In an effort to eliminate the sampling system error, other systems that place the sensing device directly at the point of observation have been described. Hydrogen electrodes, radioactive counting devices, conductivity electrodes, and thermocouples have been used, but each has introduced new problems. The continued use of indicator dilution procedures using photodensitometry has the advantages of offering rapid response and high sensing efficiency of signal. There is widespread experience with the technique, and the indicators have low toxicity. It seems advantageous, then, to devise a system that would produce curves that did not contain the sampling system distortion. If the concentration-time relation at the point being investigated, c(t), is considered the "actual input curve (to the catheter)," then its distorted version, seen by the densi-tometer after passage through the catheter can be considered the "distorted downstream curve," Cti(t), (where c and c d represent concentration and t, time). If the distorting property* of the catheter can be characterized , then by using the analog computer, any distorted curve, c (i(t), can be transformed into the actual input curve, e(t). A convenient method of analyzing the distorting function of an indicator sampling system is to introduce a sudden square step of concentration change. If the downstream distorted curve is compared to the square step input, the distorting function can be evaluated. This characteristic will be applicable to any other concentration input, providing the distorting function of the catheter does not change during the procedure. If all the concentration curves are available in an electrical form (where concentration is analagous to voltage), then the distorting function can be determined directly with an analog computer. Once this is done, correction of distorted curves is feasible. It is necessary, then, to set up a sampling system and put into it a square step change, knowing both the input curve and the downstream distorted curve. From this, …