Search

Purpose: Ovarian cancer is a common type of cancer and a leading cause of death in women. Therefore, accurate and fast prediction of ovarian tumors is crucial. One of the appropriate and precise methods for predicting and diagnosing this cancer is to build a model based on artificial intelligence methods. These methods provide a tool for predicting ovarian cancer according to the characteristics and conditions of each person. Method: In this study, a data set included records related to 171 cases of benign ovarian tumors, and 178 records related to cases of ovarian cancer were analyzed. The data set contains the records of blood test results and tumor markers of the patients. After data preprocessing, including removing outliers and replacing missing values, the weight of the effective factors was determined using information gain indices and the Gini index. In the next step, predictive models were created using random forest (RF), support vector machine (SVM), decision trees (DT), and artificial neural network (ANN) models. The performance of these models was evaluated using the 10‐fold cross‐validation method using the indicators of specificity, sensitivity, accuracy, and the area under the receiver operating characteristic curve. Finally, by comparing the performance of the models, the best predictive model of ovarian cancer was selected. Results: The most important predictive factors were HE4, CA125, and NEU. The RF model was identified as the best predictive model, with an accuracy of more than 86%. The predictive accuracy of DT, SVM, and ANN models was estimated as 82.91%, 85.25%, and 79.35%, respectively. Various artificial intelligence (AI) tools can be used with high accuracy and sensitivity in predicting ovarian cancer. Conclusion: Therefore, the use of these tools can help specialists and patients with early, easier, and less expensive diagnosis of ovarian cancer. Future studies can leverage AI to integrate image data with serum biomarkers, thereby facilitating the creation of novel models and advancing the diagnosis and treatment of ovarian cancer. [ABSTRACT FROM AUTHOR]

Background: The privacy issue in web-based registry systems is very important and patients are concerned about it. Protecting patients' privacy in medical databases is an important challenge in health informatics. Objectives: The aim of this study was to develop an instrument to assess privacy challenges of a web-based liver transplantation registry (WLTR), and to determine the views of patients on the privacy of the implemented WLTR. Methods: This study consisted of two parts: 1) an instrument development phase, in which we developed and implemented an instrument, and 2) a cross-sectional study. The instrument included a total of 59 questions based on literature review. Content validity method was performed to confirm the questionnaire validity and its reliability was evaluated based on the test-retest method. The second phase of the study included all patients who received liver transplantation (n=81) between February 2017 and April 2017 in the only organ transplant center in northeastern and eastern Iran. The data were collected using the instrument that we have developed in the first phase of the study. Data were analyzed by SPSS version 16, using descriptive statistics, Mann-Whitney U test, and Spearman's Rank-Order Correlation test. P-value less than 0.05 was considered statistically significant. Results: In the first phase, a researcher-made questionnaire consisting of 71 questions was prepared based on the extensive literature review. Necessary modifications were made to match the questionnaire with research topic by Delphi technique (12 questions were removed and 10 questions modified). The final questionnaire included 59 questions about demographic information (age, gender, education level), different dimensions of privacy (24 questions), tendency to trust (9 questions), availability of identifiable and unidentifiable health information for different users (18 question), and the goals of using identifiable and unidentifiable health information (8 questions). Reliability was evaluated based on the test-retest method (r=0.82). In the second phase, we determine the views of patients who received liver transplantation on privacy in the WLTR. Most patients (74.1%) stated that they were worried about privacy risks in online systems. The sensitivity to using online personal health information in women was higher in comparison to men (p=0.008). Older people have less tendency to use Internet systems to perform activities (p=0.023). In addition, age had a negative relationship with the convenience of using web-based registry (p=0.033, rho= -239). Regarding the purposes of using the WLTR information from the patients' views, the use of information for purposes of health care had the highest mean, and the use of this information for research purposes had the lowest mean. Conclusions: Designers and developers of these systems must, in accordance with the objectives of the registry program, and the users of these systems, develop policies for protecting the privacy of patients. [ABSTRACT FROM AUTHOR]

Background: Breast cancer is the most common cancer and the second cause of death in women worldwide. Thus, early and precise breast cancer detection could reduce mortality in the long term. Different techniques, such as mammography, and ultrasound, could be applied to breast cancer diagnosis. However, artificial intelligence (AI) tools help to identify chest abnormalities more efficiently and accurately. Objective: In this study, various models of AI are developed to predict breast cancer. Then the performance of the models is compared, and the best model is identified. Material and Methods: A dataset containing 1727 records was retrieved from Data PROM (Patient-reported outcomes) Baseline (2022). After preprocessing the dataset, Support Vector Machine (SVM), Naive Bayes, Decision Tree, Random Forest, Gradient Boosted Trees, K-Nearest Neighbors, and AutoMLP (Multi-layer perceptron) classifiers were created and evaluated using 10-fold cross-validation. The performance of the models was evaluated in terms of accuracy, sensitivity, specificity, F-measure, and area under the receiver operating characteristic curve. Results: The best performance was obtained using SVM classifier with an area under the curve (AUC) of 0.858 (±0.035), an accuracy of %78.72 (±2.64), a sensitivity of %82.95 (±3.47), a specificity of %73.41 (±4.12), and f-measure of %81.27 (±2.35).. Conclusion: Compared to similar studies, the findings obtained in this study have provided satisfactory results, indicating this model's high performance. Artificial intelligence techniques could be helpful in the prediction of breast cancer that results in omitting excessive diagnosis tests. [ABSTRACT FROM AUTHOR]