İnsan ölümlerinin en büyük nedenlerinden biri kanserdir. Kadınlar arasındaki kanser ölümlerinin başlıca sebebi ise meme kanseridir. Bu kanser türü sebebiyle yaşanan ölümleri azaltmanın yolu erken teşhistir. Uzman sistemler, yapay zeka ve makine öğrenmesi tekniklerinin tıp alanında kullanılmasının temel amaçlarından biri hastalıkları erken teşhis etmede doktorlara yardımcı olmaktır. Kanser türleri arasında özellikle meme kanserinde erken teşhis sayesinde ölüm riski büyük oranda düşürülebilir. Bu çalışmada temel bileşen analizi (Principal Component Analysis-PCA) ve ileri beslemeli sinir ağı (Feed Forward Neural Network-FFNN) temelli yeni bir kanser teşhisi yöntemi önerilmiştir. Önerilen yöntemin performansı Meme Kanseri Coimbra Veri Seti (Breast Cancer Coimbra Dataset-BCCD) üzerinde sınıflandırma doğruluğu, kesinlik, duyarlılık ve F-ölçütü metrikleri ile test edilmiştir. Ayrıca önerilen yöntemin klasik makine öğrenmesi teknikleri ve literatürdeki çalışmalar ile ayrıntılı olarak karşılaştırmalı performans analizi yapılmıştır. Deneysel sonuçlar önerilen yöntemin etkin olduğunu ve erken teşhis için doktorlar tarafından kullanılabileceğini göstermektedir.
One of the biggest causes of human death is cancer. The main cause of cancer death among women is breast cancer. The way to reduce the deaths caused by this type of cancer is early diagnosis. One of the main objectives of the use of specialized systems, artificial intelligence and machine learning techniques in the field of medicine is to help doctors in early diagnosis of diseases. Due to early diagnosis among cancers, especially in breast cancer, the risk of death can be significantly reduced. This study proposed a new cancer diagnosis method based on the basic component analysis (PCA) and the feed forward neural network (FFNN). The performance of the recommended method has been tested on the Breast Cancer Coimbra Dataset (BCCD) with classification accuracy, accuracy, sensitivity and F-dimensional metric. A detailed comparative performance analysis was also made with the classic machine learning techniques of the recommended method and the studies in literature. Experimental results show that the recommended method is effective and can be used by doctors for early diagnosis.
One of the major causes of human death is cancer. Breast cancer is the main reason for cancer deaths among women. Early diagnosis is the way to reduce deaths due to this cancer type. One of the main objectives of the use of expert systems, artificial intelligence and machine learning techniques in medicine is to assist doctors in early diagnosis of diseases. Among cancer types, the risk of death can be greatly reduced by early diagnosis, especially in breast cancer. In this study, a new cancer diagnosis method based on Principal Component Analysis (PCA) and Feed Forward Neural Network (FFNN) has been proposed. The performance of the proposed method is tested on the Breast Cancer Coimbra Dataset (BCCD) with classification accuracy, precision, recall and F-measure metrics. Besides, the comparative performance analysis of the proposed method with conventional machine learning techniques and studies in the literature is performed. Experimental results show that the proposed method is effective and can be utilized by doctors for early diagnosis.
Alan : Fen Bilimleri ve Matematik; Mühendislik
Dergi Türü : Ulusal
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|