Improving the performance of EM and K-means algorithms for breast lesion segmentation

Fuldem Mutlu; Sevda Gül

Makale detayları

Improving the performance of EM and K-means algorithms for breast lesion segmentation

Yazarlar:

Fuldem Mutlu | Sevda Gül

Yayıncı:

MediHealth Academy Yayıncılık

Ana Başlıklar:

Sağlıkta Bilgi İşleme | Radyoloji ve Organ Görüntüleme

Dergi:

Anatolian Current Medical Journal Vol.5 Issue 4, pp.492-497

Belge Tipi:

Makale

Yayın Tarihi:

10/2023

Belge Dili:

İngilizce

Document ID (Org No.):

2001302562

pISSN:

2718-0115

Anahtar kelimeler:

Özet:

 Aims: Breast cancer is the most common type of cancer in women and accounts for a large portion of cancer-related deaths. As in the other types of cancer, the prevention and early diagnosis of breast cancer gain importance day after day. For this purpose, the artificial intelligence-based decision support systems become popular in recent years. In this study, an automatic breast lesion segmentation process is proposed to detect breast lesions in the images taken with magnetic resonance imaging MRI protocol. Methods: Two most popular segmentation methods: expectation maximization EM and K-means algorithms are used to determine the region of breast lesions. Furthermore, superpixel based fuzzy C-means SPFCM algorithm is applied after EM and K-means methods to improve the lesion segmentation performance. Results: The proposed methods are evaluated on the private database constructed by the authors with ethical permission. The performances of the utilized methods are analyzed by comparing the lesion areas determined by a radiologist ground-truth and areas that are achieved by automatic segmentation algorithms. Conclusion: Dice coefficient, Jaccard index JI, and area under curve AUC metrics are calculated for performance comparison. According to the simulation results, EM, K-means, EMSPFCM, and K-meansSPFCM methods provides good segmentation performance on breast MRI database. The best segmentation results are obtained by using EMSPFCM hybrid method. The results of the EMSPFCM method are 0,8711, 0,8979, and 0,9981 for JI, Dice, and AUC, respectively.

 Aims: Breast cancer is the most common type of cancer in women and accounts for a large portion of cancer-related deaths. As in the other types of cancer, the prevention and early diagnosis of breast cancer gain importance day after day. For this purpose, the artificial intelligence-based decision support systems become popular in recent years. In this study, an automatic breast lesion segmentation process is proposed to detect breast lesions in the images taken with magnetic resonance imaging MRI protocol. Methods: Two most popular segmentation methods: expectation maximization EM and K-means algorithms are used to determine the region of breast lesions. Furthermore, superpixel based fuzzy C-means SPFCM algorithm is applied after EM and K-means methods to improve the lesion segmentation performance. Results: The proposed methods are evaluated on the private database constructed by the authors with ethical permission. The performances of the utilized methods are analyzed by comparing the lesion areas determined by a radiologist ground-truth and areas that are achieved by automatic segmentation algorithms. Conclusion: Dice coefficient, Jaccard index JI, and area under curve AUC metrics are calculated for performance comparison. According to the simulation results, EM, K-means, EMSPFCM, and K-meansSPFCM methods provides good segmentation performance on breast MRI database. The best segmentation results are obtained by using EMSPFCM hybrid method. The results of the EMSPFCM method are 0,8711, 0,8979, and 0,9981 for JI, Dice, and AUC, respectively.

 Aims: Breast cancer is the most common type of cancer in women and accounts for a large portion of cancer-related deaths. As in the other types of cancer, the prevention and early diagnosis of breast cancer gain importance day after day. For this purpose, the artificial intelligence-based decision support systems become popular in recent years. In this study, an automatic breast lesion segmentation process is proposed to detect breast lesions in the images taken with magnetic resonance imaging MRI protocol. Methods: Two most popular segmentation methods: expectation maximization EM and K-means algorithms are used to determine the region of breast lesions. Furthermore, superpixel based fuzzy C-means SPFCM algorithm is applied after EM and K-means methods to improve the lesion segmentation performance. Results: The proposed methods are evaluated on the private database constructed by the authors with ethical permission. The performances of the utilized methods are analyzed by comparing the lesion areas determined by a radiologist ground-truth and areas that are achieved by automatic segmentation algorithms. Conclusion: Dice coefficient, Jaccard index JI, and area under curve AUC metrics are calculated for performance comparison. According to the simulation results, EM, K-means, EMSPFCM, and K-meansSPFCM methods provides good segmentation performance on breast MRI database. The best segmentation results are obtained by using EMSPFCM hybrid method. The results of the EMSPFCM method are 0,8711, 0,8979, and 0,9981 for JI, Dice, and AUC, respectively.

 Özet Amaçlar: Meme kanseri, kadınlarda en yaygın görülen kanser türüdür ve kanserle ilgili ölümlerin büyük bir bölümünü oluşturur. Diğer kanser türlerinde olduğu gibi, meme kanserinin önlenmesi ve erken teşhisi gün geçtikçe daha da önem kazanmaktadır. Bu amaçla, yapay zeka tabanlı karar destek sistemleri son yıllarda popüler hale gelmektedir. Bu çalışmada, manyetik rezonans görüntüleme MRG protokolü ile çekilen görüntülerde meme lezyonlarını tespit etmek için otomatik bir meme lezyonu segmentasyon süreci önerilmektedir. Yöntemler: İki en popüler segmentasyon yöntemi olan beklenti maksimizasyonu EM ve K-ortalama algoritmaları, meme lezyonlarının bölgesini belirlemek için kullanılmıştır. Ayrıca, EM ve K-ortalama yöntemlerinden sonra süper piksel tabanlı bulanık C-ortalama SPFCM algoritması, lezyon segmentasyon performansını artırmak için uygulanmıştır. Sonuçlar: Önerilen yöntemler, yazarlar tarafından etik izinle oluşturulan özel bir veritabanında değerlendirilmiştir. Kullanılan yöntemlerin performansları, bir radyolog tarafından belirlenen lezyon alanlarıyla gerçek veri otomatik segmentasyon algoritmalarıyla elde edilen alanların karşılaştırılmasıyla analiz edilmiştir. Sonuç: Performans karşılaştırması için Dice katsayısı, Jaccard endeksi JI ve eğri altı alan AUC ölçüleri hesaplanmıştır. Simülasyon sonuçlarına göre, EM, K-ortalama, EMSPFCM ve K-ortalamaSPFCM yöntemleri meme MRG veritabanında iyi segmentasyon performansı sağlamaktadır. En iyi segmentasyon sonuçları EMSPFCM hibrit yöntemi kullanılarak elde edilmektedir.

Makale detayları

Improving the performance of EM and K-means algorithms for breast lesion segmentation

Yazarlar:

Fuldem Mutlu | Sevda Gül

Yayıncı:

MediHealth Academy Yayıncılık

Ana Başlıklar:

Sağlıkta Bilgi İşleme | Radyoloji ve Organ Görüntüleme

Dergi:

Anatolian Current Medical Journal Vol.5 Issue 4, pp.492-497

Belge Tipi:

Makale

Yayın Tarihi:

10/2023

Belge Dili:

İngilizce

Document ID (Org No.):

2001302562

pISSN:

2718-0115

breast cancer

automatic segmentation

magnetic resonance imaging

image processingmeme kanseri

otomatik segmentasyon

manyetik rezonans görüntüleme

görüntü işleme

Sayı

Makale detayları

Yazarlar: Fuldem Mutlu | Sevda Gül

Yayıncı: MediHealth Academy Yayıncılık

Ana Başlıklar: Sağlıkta Bilgi İşleme | Radyoloji ve Organ Görüntüleme

Dergi: Anatolian Current Medical Journal Vol.5 Issue 4, pp.492-497

Belge Tipi: Makale

Yayın Tarihi: 10/2023

Belge Dili: İngilizce

Document ID (Org No.): 2001302562

pISSN: 2718-0115

Sayı

Yazarlar:

Fuldem Mutlu | Sevda Gül

Yayıncı:

MediHealth Academy Yayıncılık

Ana Başlıklar:

Sağlıkta Bilgi İşleme | Radyoloji ve Organ Görüntüleme

Dergi:

Anatolian Current Medical Journal Vol.5 Issue 4, pp.492-497

Belge Tipi:

Makale

Yayın Tarihi:

10/2023

Belge Dili:

İngilizce

Document ID (Org No.):

2001302562

pISSN:

2718-0115