ORIGINAL ARTICLE Year : 2014  Volume : 13  Issue : 1  Page : 4045 Select the Optimized Effective Dose to Reduce Nuclear Radiations in Pediatric Nuclear Medicine Ying Bai^{1}, Dali Wang^{2}, ^{1} Department of Computer Science and Engineering, Johnson C. Smith University, NC 28216, USA ^{2} Department of Physics and Computer Science, Christopher Newport University, Newport News, VA 23606, USA Correspondence Address: Many techniques and research models on calculating and reducing the nuclear radiation dose on pediatric nuclear medicine procedure have been developed and reported in recent years. However, most those models either utilized simple shapes to present the organs or used more realistic models to estimate the nuclear dose applied on pediatric patients. The former are too simple to provide accurate estimation results, and the latter are too complicated to intensively involve complex calculations. In this study, a simple but practical model is developed to enable physicians to easily and quickly calculate and select the average optimal effective nuclear dose for the given age and bodysize of the pediatric patients. This model is built based on one research result reported by Frederic Fahey, et al and it can be easily implemented in most common pediatric nuclear medicine procedures. This is the first research of using fuzzy inference system to calculate the optimal effective dose applied in the nuclear medicine for pediatric patients.
Introduction Nuclear medicine provides important and critical information that assists in the diagnosis, treatment, and followup of a variety of disorders on pediatric patients, including central nervous, endocrine, cardiopulmonary, renal, and gastrointestinal systems, as well as in the fields of oncology, orthopedics, organ transplantation, and surgery. Due to its high sensitivity, nuclear medicines can detect some disease in its earliest stages to enable it to be treated earlier. The noninvasive nature of nuclear medicine makes it an extremely valuable diagnostic tool for the evaluation of children. It provides useful diagnostic information that may not be easily obtained by using other diagnostic methods, some of them may be more invasive or contain some higher nuclear radiations. [1],[2] Pediatric nuclear medicine includes the application of small amounts of radiopharmaceuticals that emit nuclear radiations such as γrays, βparticles, or positrons to patients during the diagnostic process. This emission exposes the pediatric patient to low levels of nuclear radiations that might be the result in harmful health effects on pediatric patients. In most nuclear medicine procedures, the amounts of radiation (dose) applied on pediatric patients are limited to certain low levels, but they are contradictory to the mechanistic biologic observations. It had been difficult for most physicians to effectively assess the magnitude of exposure or potential risk due to implementation of nuclear radiations on pediatric treatments. The challenge job is how to make a tradeoff between the nuclear radiation dose applied on the pediatric patients and the quality of the diagnostic results, and to select or determine an optimal or minimized effective dose to reduce the risk of nuclear radiations. [3] Effective dose provides an approximate indicator of potential detriment from nuclear radiation and should be used as one parameter in evaluating the appropriateness of examinations involving nuclear radiation. In fact, effective dose is a calculated quantity and cannot be measured. Multiplying the average organ equivalent dose by the International Commission of Radiological Protection tissueweighting factor and summing the results over the whole body yields the effective dose. [4] Although effective dose is an average evaluation value, it is still an important parameter in the estimation of average potential risks of nuclear radiation on patients. Because of the popular applications of nuclear medicines on pediatric diagnostics and treatments, remarkable increase in the use of nuclear medical procedures have been shown in the US in recent years. [5] Different techniques and models have been reported and developed to optimize the nuclear radiation dose to reduce the risk of nuclear radiations on patients in last decades. [6],[7],[8],[9],[10],[11],[12],[13],[14],[15] One of the most important reasons for these developments is to reduce the potential risk of cancers that results from the nuclear radiations exposed from the usage of the nuclear medicine procedures. [16],[17],[18],[19],[20],[21],[22],[23],[24],[25],[26],[27],[28],[29],[30],[31],[32] Accorsi et al. provided a method to improve the dose regimen in pediatric PET. [9] Some other research organizations reported different radiation sources used in nuclear medicines in recent years. [10],[11] Fazel et al. developed a procedure to use lowdose ionizing radiation in medical image process. [13] Frederic Fahey et al. provided a survey to review most recent developments in using minimized dose to reduce the risk of inducing cancer. [16] Loevinger and Budinger in their study have reported a method to calculate the absorbed dose to limit the effects of radiations. [17] Stabin and Siegel discussed some popular physical models and dose factors for use in internal dose assessment. [18] Ward et al. developed a method to reduce the effective dose for the pediatric radiation exposure. [22] Preston reported an on linear nonthreshold doseresponse model and implications for diagnostic radiology procedures. [23] Gelfand developed a method to reduce the dose applied in pediatric hybrid and planar imaging process. [25] Hsiao et al. have reported a technique to reduce the radiation dose in MAG3 renography by enhanced planar processing. [27] Other researchers reported different techniques and methods to reduce radiation exposures in nuclear medicine and medicine image processing. [28],[29],[30],[31],[32] However, most of these technologies and developments either utilized simple shapes to present the organs or used more realistic models to estimate the nuclear dose applied on pediatric patients. The former are too simple to provide accurate estimation results, and the latter are too complicated to intensively involve complex calculations. Furthermore, these estimations are averages over a wide range of patients at each age and they are not related to individual differences in anatomy and physiology from the standard models. Application of these pediatric models is problematic because children can vary greatly in body size and habitus. A good model should deal with both the children's age and the bodysize to determine the optimal effective dose. The advantage of using our model as discussed in this paper is that physicians can easily and quickly calculate and select the optimal or minimized effective dose based on the given age and bodysize of the pediatric patient to significantly reduce the effects of nuclear radiations on patients. This kind of model will be more suitable and appropriate for pediatric examination and diagnoses. Materials and Methods We used the fuzzy inference system (FIS) to build a dynamic model to set a mapping relationship between each age, weight and the desired optimal effective dose. All related data and operational parameters used for this model are based on data provided by. [16] The estimates of critical organ and effective dose for common pediatric nuclear medicine procedures developed by [16] are shown in [Table 1]. This table shows estimated relationships between the pediatric patients' ages, weights, and effective doses for 99m Tcethyl cysteinate dimer (ECD).{Table 1} It can be seen from [Table 1] that this table only provided limited information between certain children ages with selected weights and the minimized nuclear effective dose. In other words, the relationship or mapping between the children ages, weights and the optimal effective dose is incomplete or discrete because it does not provide all optimal effective doses for any given children age and weight. To improve that incomplete and discrete model, in this study, we will use a FIS to build a complete and continuous model to provide all related optimal effective doses for different given children ages and weights in a simple and easy way. In fact, we will use the FIS to interpolate the optimal effective dose based on the specified age and weight of each child group to simplify the calculation process for the effective dose. To make our study simple, we only use the bladder wall with 99m TcECD as an example to illustrate how to use FIS to simplify this effective dose calculation process. This study can be easily extended to cover all other organs and methods shown in [Table 1]. A graphic mapping between the effective dose and given age and weight of each group children with the bladder wall in [Table 1] is shown in [Figure 1].{Figure 1} The basic idea behind this model development is based on the fact, that the optimal effective dose is not a continuous function for all different given ages and weights located between known ages and weights. Furthermore, the relationship between the minimized effective dose and different ageweight is ambiguous, or at least it is not a linear one as shown in [Figure 1]. Therefore, we need to use the fuzzy inference algorithm to derive those optimal effective doses for all those "missed" ageweight pairs. In fact, we use fuzzy inference method to interpolate those optimal effective doses for any specified ageweight pair. Fuzzy inference system We use given actual age and weight of the pediatric patient as inputs, and the optimal effective doses as the output for a FIS. Therefore, this is a multiinput and singleoutput system. Both inputs and output are connected and controlled by the control rules. [Figure 2] shows the block diagram of this FIS.{Figure 2} As for the membership functions for two inputs, pediatric patient age and weight, we utilized Gauss form as the shape for both of them. Similarly, this shape is also used for the output, the optimal effective dose. The membership functions for both inputs (patient's age and weight) are shown in [Figure 3]. The membership function for the output (effective dose) is shown in [Figure 4], respectively. Those membership functions are derived based on the data provided by [16] for common pediatric nuclear medicine procedures.{Figure 3}{Figure 4} The definitions for the membership functions of the pediatric patient's age and weight are shown in [Table 2] and [Table 3], and the membership function for effective dose is shown in [Table 4].{Table 2}{Table 3}{Table 4} For this implementation, fourteen control rules are developed based on the inputoutput member functions. These fourteen control rules are shown in [Table 5]. The surface relationship between the output (Effective Dose) and the inputs, age and weight, is shown in [Figure 5].{Figure 5}{Table 5} Results The optimal effective dose and pediatric age and weight Based on the membership functions of two inputs, patient's age (age) and weight (weight), and the membership function of the output, effective dose, the desired optimal effective dose for the given patient's age and weight can be easily determined and obtained directly from the fuzzy rule relationship. [Figure 6] shows this kind of model for the calculation of optimal effective dose used in pediatric bladder wall inspection using the nuclear medicine procedures.{Figure 6} As shown in [Figure 6], a typical pediatric patient age (12.5 years old) and weight (24.8 kg) are selected. The related optimal effective dose (4.66 mSv) is determined directly from this fuzzy rule relationship. During the implementation process, the vertical bars on both inputs, patient's age and weight, can be moved by the pediatric physician to either left or right to select the specified age and weight group of pediatric patients, and the desired optimal effective dose can be easily determined directly from this fuzzy inputoutput rules relationship map. This model provides great flexibility and simplicity to determine the optimal effective dose for common pediatric nuclear medicine procedures. We can also easily build a similar FIS model using the data provided by [16] to determine the related optimal effective doses for the other kinds of pediatric organs' nuclear medicine procedures. Conclusion and Discussion A flexible and simple model used to set a fuzzy mapping relationship between the pediatric patients' ageweight and the optimal effective dose is developed in this study to enable pediatric physicians to easily and directly determine the optimal effective doses for the common pediatric nuclear medicine procedures. The advantage of using this model is that the pediatric physicians can obtain the desired minimized effective dose based on the given group of pediatric patients' data, such as ages and weights, easily and directly from the fuzzy rule relationship. Acknowledgments Special thanks should be given to Dr. Frederic H. Fahey et al., for their permission to allow us to use their table, [Table 1], in one of their papers, Minimizing and Communicating Radiation Risk in Pediatric Nuclear Medicine published in the Journal of Nuclear Medicine in March 1, 2012. References


