Detection of colon cancer using k-means and deep learning algorithms on histopathological images

Ulaş Yurtsever; Hayrettin Evirgen; Mustafa Avunduk

Article Details

Detection of colon cancer using k-means and deep learning algorithms on histopathological images

Authors:

Ulaş Yurtsever | Hayrettin Evirgen | Mustafa Avunduk

Publisher:

Pamukkale University

Main Topics:

Bilgisayar Sistem Yazılımı

Journal:

Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi Vol.31 Issue 5, pp.821-832

Document Type:

Article

Publication Date:

10/2025

Document Language:

English

Document ID (Org No.):

2001360215

pISSN:

2147-5881

Keywords:

Abstract:

 In this research, a novel approach for classifying colon cancer was developed by employing two convolutional neural network CNN models, namely GoogLeNet and AlexNet. This approach involves training CNNs with histopathological images segmented into color clusters using an augmented k-means clustering algorithm, rather than utilizing original-raw images. This method was applied to 20 datasets with distinct structural and characteristic features, derived from larger datasets comprising both original and segmented images. The datasets were used to train and test CNN models. The results indicate that AlexNet, trained with segmented images, showed a 2 to 23 increase in accuracy performance, while GoogLeNets accuracy performance improved by 2 to 27. Notably, the proposed approach yielded higher accuracy with datasets containing non-homogeneous data.

 In this research, a novel approach for classifying colon cancer was developed by employing two convolutional neural network CNN models, namely GoogLeNet and AlexNet. This approach involves training CNNs with histopathological images segmented into color clusters using an augmented k-means clustering algorithm, rather than utilizing original-raw images. This method was applied to 20 datasets with distinct structural and characteristic features, derived from larger datasets comprising both original and segmented images. The datasets were used to train and test CNN models. The results indicate that AlexNet, trained with segmented images, showed a 2 to 23 increase in accuracy performance, while GoogLeNets accuracy performance improved by 2 to 27. Notably, the proposed approach yielded higher accuracy with datasets containing non-homogeneous data.

 In this research, a novel approach for classifying colon cancer was developed by employing two convolutional neural network CNN models, namely GoogLeNet and AlexNet. This approach involves training CNNs with histopathological images segmented into color clusters using an augmented k-means clustering algorithm, rather than utilizing original-raw images. This method was applied to 20 datasets with distinct structural and characteristic features, derived from larger datasets comprising both original and segmented images. The datasets were used to train and test CNN models. The results indicate that AlexNet, trained with segmented images, showed a 2 to 23 increase in accuracy performance, while GoogLeNets accuracy performance improved by 2 to 27. Notably, the proposed approach yielded higher accuracy with datasets containing non-homogeneous data.

 Bu araştırmada, GoogLeNet ve AlexNet olmak üzere iki evrişimli sinir ağı CNN modeli kullanılarak kolon kanserinin sınıflandırılması için yeni bir yaklaşım geliştirilmiştir. Bu yaklaşımda CNNler, orijinal ham görüntüleri kullanmak yerine, artırılmış bir k-ortalamalar kümeleme algoritması kullanılarak renk kümelerine ayrılmış histopatolojik görüntüleri kullanarak eğitilmektedir. Bu yöntem hem orijinal hem de bölütlenmiş görüntülerden oluşan daha büyük veri kümelerinden elde edilen farklı yapısal ve karakteristik özelliklere sahip 20 veri kümesine uygulanmıştır. Veri kümeleri CNN modellerini eğitmek ve test etmek için kullanılmıştır. Sonuçlar, bölümlere ayrılmış görüntülerle eğitilen AlexNetin doğruluk performansında 2 ile 23 arasında bir artış gösterdiğini, GoogLeNetin doğruluk performansının ise 2 ile 27 arasında iyileştiğini ortaya koymuştur. Özellikle, önerilen yaklaşım homojen olmayan verilere sahip veri kümelerinde daha yüksek doğruluk sağlamıştır.

Article Details

Detection of colon cancer using k-means and deep learning algorithms on histopathological images

Authors:

Ulaş Yurtsever | Hayrettin Evirgen | Mustafa Avunduk

Publisher:

Pamukkale University

Main Topics:

Bilgisayar Sistem Yazılımı

Journal:

Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi Vol.31 Issue 5, pp.821-832

Document Type:

Article

Publication Date:

10/2025

Document Language:

English

Document ID (Org No.):

2001360215

pISSN:

2147-5881

Convolutional neural network

Deep learning

Image segmentation

Image classification

Colon cancerEvrişimli sinir ağı

Derin öğrenme

Görüntü bölütleme

Görüntü sınıflandırma

Kolon kanseri

Issues

Article Details

Authors: Ulaş Yurtsever | Hayrettin Evirgen | Mustafa Avunduk

Publisher: Pamukkale University

Main Topics: Bilgisayar Sistem Yazılımı

Journal: Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi Vol.31 Issue 5, pp.821-832

Document Type: Article

Publication Date: 10/2025

Document Language: English

Document ID (Org No.): 2001360215

pISSN: 2147-5881

Issues

Authors:

Ulaş Yurtsever | Hayrettin Evirgen | Mustafa Avunduk

Publisher:

Pamukkale University

Main Topics:

Bilgisayar Sistem Yazılımı

Journal:

Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi Vol.31 Issue 5, pp.821-832

Document Type:

Article

Publication Date:

10/2025

Document Language:

English

Document ID (Org No.):

2001360215

pISSN:

2147-5881