ترکیب استدلال مبتنی بر بهینه سازی کلونی زنبور عسل برای برنامه ریزی دوز ومقایسه آن در درمان سرطان تیروئید
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|7575||2013||9 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Expert Systems with Applications, Volume 40, Issue 6, May 2013, Pages 2147–2155
Thyroid cancers are the most common endocrine carcinomas. Case-based reasoning (CBR) is used in this paper to describe a physician’s expertise, intuition and experience when treating patients with well differentiated thyroid cancer. Various clinical parameters (the patient’s diagnosis, the patient’s age, the tumor size, the existence of metastases in the lymph nodes and the existence of distant metastases) influence a physician’s decision-making in dose planning. The weights (importance) of these parameters are determined here with the Bee Colony Optimization (BCO) meta-heuristic. The proposed CBR–BCO model suggests the I-131 iodine dose in radioactive iodine therapy. This approach is tested on real data from patients treated in the Department of Nuclear Medicine, Clinical Center Kragujevac, Serbia. By comparing the results that are obtained through the developed CBR–BCO model with those resulting from the physician’s decision, it has been found that the developed model is highly reflective of reality.
Thyroid cancers are the most common endocrine carcinomas. They are also among the ten most common cancers in women. Among thyroid malignancies, more than 95% are well differentiated thyroid cancers (WDTCs) of follicular cell origin, papillary and follicular carcinomas. Malignant transformation of thyroid epithelial cells (follicular and parafollicular), followed by their division and accumulation, leads to the development of malignant tumors of the thyroid gland. Well differentiated thyroid cancers are more common in younger patients; papillary carcinoma is usually diagnosed in the third decade of life, and follicular carcinoma is usually diagnosed in the fourth decade. All types of thyroid cancer occur 1.5–4 times more often in women than in men. The incidence of WDTC has increased in several European countries since the Chernobyl nuclear disaster. The most important risk factors that increase the probability of WDTC are: age, female sex, radiation exposure of the neck region (especially in childhood) and the positive familiar anamnesis of other malignancies. The disproportion between the number of patients with carcinoma of the thyroid gland and the number of deaths (95% vs. 65.9%, respectively) indicates that the malignant disease has a relatively indolent nature, and thus applied therapies are likely to be successful. The aim of the application of radioactive iodine in patients with well differentiated thyroid carcinoma is the ablation of normal thyroid tissue or cancer cell therapy after thyroidectomy. The dosing decision that is made by an experienced physician is based on the clinical stage of the disease, its risk factors and the TNM classification (T – Tumor size, N – metastases in the lymph nodes and M – distant metastases). Although WDTC has a very good prognosis in the vast majority of patients, there are still patients who have high risk factors and who require, even after a total or near total thyroidectomy, radioactive iodine therapy (Riesco-Eizaguirre and Santisteban, 2007, Savin et al., 2010 and Vrndic et al., 2011). The ability of the thyroid follicular cells to take up iodine via a sodium iodide symporter at the basolateral cell membrane enables the use of radioiodine for the therapy of WDTC. The choice of dose proposed by a physician cannot be easily described by precise rules and/or mathematical algorithms. There is a trade-off between the benefit and the risk of radioiodine (I-131) therapy. In this paper, we developed a case-based reasoning (CBR) model Aamondt and Plaza, 1994, Burkhard and Wess, 1998 and Kolodner, 1993 for the prescription of the I-131 iodine dose in WDTC treatment. The case-based reasoning is used in this paper to describe a physician’s expertise, intuition, and experience. Various clinical parameters (patient’s diagnosis, age, and TNM classification) influence a physician’s decision-making in dose planning. These parameters do not have the same importance and influence on proposed doses in the physician’s decision-making process. The weights (importance) of these parameters are determined in this paper by the Bee Colony Optimization (BCO) meta-heuristic (Lučić and Teodorović, 2001, Lučić and Teodorović, 2002 and Lučić and Teodorović, 2003). The main objective of this paper is to research the possibility of developing a new model that could improve the quality of the decisions made by young physicians who study WDTC. In other words, our intention is to use the CBR–BCO model mainly for educational purposes. Our key interest is to develop a model that helps students to test their acquired knowledge and to learn how to reach an appropriate decision. This paper is organized as follows. Section 2 presents a brief literature survey. CBR for a radioiodine (I-131) dose decision is given in Section 3. The BCO meta-heuristic is described in Section 4. BCO application on the weight determination process is given in Section 5. The results and discussion are presented in Section 6. Section 7 contains the conclusions.
نتیجه گیری انگلیسی
In this paper, we studied the radioiodine (I-131) dose decision problem in the case of patients with well differentiated thyroid cancer. This real-life problem presents a very difficult decision-making task for the physician, who must be very knowledgeable, experienced and trained properly to adequately perform the task. We used case-based reasoning (CBR) to describe the physician’s expertise, intuition, and experience. Various clinical parameters (patient’s diagnosis, patient’s age, tumor size, existence of metastases in the lymph nodes and existence of distant metastases) influence the physician’s decision-making process for dose planning. The weights (importance) of these parameters are determined by the Bee Colony Optimization (BCO) meta-heuristic. The proposed CBR–BCO represents a judgment system that suggests an I-131 iodine dose in radioactive iodine therapy. The proposed approach was tested on the real data sets collected from the University Clinical Center in Kragujevac, Serbia. However, because of the nature of the information and the difficulty of obtaining the data, the number of available data points was limited. By comparing the results obtained through the model with those resulting from the physician’s decision, it was found that the developed model is highly representative of reality. The developed CBR–BCO model could be used for educational purposes, and with further improvements, it could assist and guide young physicians.