This paper describes methods of entering data into an expert system used to study lower urinary tract pressure and flow. Data is usually entered into an expert system directly from the keyboard or on an online basis. We have developed a new method of offline data entry using a photoscanner. For this method, we developed versatile software to convert the graphic data read by a photoscanner to online data format. This software allows us to handle graphic data printed on paper in the same way as online data. The software is also designed to solve the problems caused by blurred lines or ink spots on printed charts. In other words, the software compensates for data reading errors caused by blurred or stained lines on printed charts. Because data can be entered through the photoscanner, the enormous volume of data already recorded in graphic forms can be used by the expert system. As a result, the diagnostic capability of the expert system is significantly improved.