Mammography is the most preferred method for breast cancer screening. In this study, computer-aided diagnosis (CAD) systems were used to improve the image quality of mammography images and to detect suspicious areas. The main contribution of this study is to reveal the optimal combination of various pre-processing algorithms to enable better interpretation and classification of mammography images because pre-processing algorithms significantly affect the accuracy of segmentation and classification methods. In this study, the effect of combinations of different preprocessing methods in differentiating benign and malignant breast lesions was investigated. All image processing algorithms used for lesion detection were used in the mini-MIAS database. In the first step, label information and pectoral muscle resulting from the acquisition of mammography images were removed. In the second step, median filter (MF), contrast limited adaptive histogram equalization (CLAHE), and unsharp masking (USM) algorithms with different combinations of the resolution and visibility of images are increased. In the third step, suspicious regions are extracted from the mammograms using the k-means clustering technique. Then, features were extracted from the obtained ROIs. Finally, feature datasets were classified as normal/abnormal, and benign/malign (two class classification) using Machine Learning algorithms. Test performance measures of the classification methods were examined. In both classifications made in the study, lower classification performance values were obtained when the CLAHE algorithm was used alone as a pre-processing method compared to other pre-processing combinations. When the median filter and unsharp masking algorithms are added to the CLAHE algorithm, the performance of the classification methods has increased. In terms of classification success, Support Vector Machines, Random Forest, and Neural Networks showed the best performance. It was found by comparing the performances of the classification methods that different preprocessing algorithms were effective in detecting the presence of breast lesions and distinguishing benign and malignant.

No abstract

Breast cancer is one of the most common types of cancer in women. To make a fast diagnosis, mammography images should have high contrast. Computer-assisted diagnosis (CAD) models are computer systems that help diagnose lesioned areas on medical images. The aim of this study is to examine the contribution of the changes in parameter values of various pre-processing methods used to increase the visibility of mammography images and reduce the noise in the images, to the classification performance. In this study, the mini-MIAS database were used. Gaussian filter, Contrast Limited Adaptive Histogram Equalization and Fast local Laplacian filtering methods were applied as pre-processing method. In this study, two different parameter values were applied for two different image processing methods (Ⅰ. Parameter values are Gauss filter σ=3, Laplacian filter σ=0.6, and α=0.6; Ⅱ. Parameter values are Gauss filter σ=1, Laplacian filter σ=2, and α=2 In the normal-abnormal tissue classification, higher accuracy and area under the curve were obtained in the 2nd parameter values in all classification methods. As a result, it has been acquired that different parameter values of the pre-processing methods used to improve mammography images can change the success of the classification methods.

Objectives The global spread of COVID-19 and associated policies have caused negative factors at the level of children, families, and services, resulting in physical, mental, and developmental issues in children, as well as limited access to healthcare. We evaluated the referral numbers, sources, and trends of a developmental-behavioral pediatrics (DBP) department in Turkey as a Eurasian country, as well as the effects of the COVID-19 pandemic on referral variables. Methods This longitudinal observational research examined patient referral data to Hacettepe University Developmental-Behavioral Pediatrics Department (HUDPD) between 2014 and 2021. We analyzed the changes in the number of referrals over time in 3-month intervals using negative binomial regression models (NBR). The impact of the COVID-19 pandemic on referral reasons was evaluated. Results The overall number of referrals increased by 1.040-fold [95% confidence interval (CI) 1.015–1.067] per year. There was a 1.070 (95% CI: 1.033–1.110)-fold increase per year before the pandemic, and a 1.077 (95% CI: 0.933–1.24)-fold increase afterward. Referrals for perinatal-neonatal risks were 1.359 (95% CI: 1.269–1.456) times higher than in the pre-pandemic period, and those for suspected autism were 1.209 (95% CI: 0.987–1.478) times higher. Conclusion Although it is encouraging that our referral trends have improved in the 1.5 years since the COVID-19 pandemic, it is thought that health service constraints caused a considerable increase in prenatal risk and suspicion of autism referrals following the pandemic. Improvement and innovation in healthcare systems to prevent the long-term detrimental impacts of periodic interruptions in healthcare on children's development and behavior is needed.