Performance Analysis of EfficientNet Based Segmentation Models for Automatic Detection of Malaria Disease

Murat Uçar; Emine Uçar

Makale detayları

Performance Analysis of EfficientNet Based Segmentation Models for Automatic Detection of Malaria Disease

Yazarlar:

Murat Uçar | Emine Uçar

Yayıncı:

Gazi University

Ana Başlıklar:

Bilgisayar Yazılımı

Dergi:

Bilişim Teknolojileri Dergisi Vol.16 Issue 3, pp.167-176

Belge Tipi:

Makale

Yayın Tarihi:

7/2023

Belge Dili:

İngilizce

Document ID (Org No.):

2001293445

pISSN:

1307-9697

Anahtar kelimeler:

Özet:

 Malaria is a disease caused by the Plasmodium parasite, which is common in the tropics. The traditional methods commonly used to diagnose malaria, one of the worlds deadliest diseases, are microscopic diagnostic methods in which blood samples taken from suspected individuals are manually examined, or rapid diagnostic tests that are sensitive to human errors. These processes are inexpensive, but experienced and qualified clinicians are needed. Due to this shortcoming, modern diagnostic tools are crucial in the struggle against the disease. In this study, an approach based on deep learning methods was used, which offers beneficial solutions in the diagnosis of disease from medical images. In the proposed approach, U-Net, Pyramid Scene Parsing Network, LinkNet, and Feature Pyramid Network segmentation methods were modified with 8 different pre-trained variants of the EfficientNet deep learning model to obtain improved models. In the malaria segmentation performed with these models, the highest Dice score of 91.50 was achieved in the use of the U-Net model with EfficientNetB6. This model offers a faster and more robust solution to detecting parasites compared to traditional methods.

 Malaria is a disease caused by the Plasmodium parasite, which is common in the tropics. The traditional methods commonly used to diagnose malaria, one of the worlds deadliest diseases, are microscopic diagnostic methods in which blood samples taken from suspected individuals are manually examined, or rapid diagnostic tests that are sensitive to human errors. These processes are inexpensive, but experienced and qualified clinicians are needed. Due to this shortcoming, modern diagnostic tools are crucial in the struggle against the disease. In this study, an approach based on deep learning methods was used, which offers beneficial solutions in the diagnosis of disease from medical images. In the proposed approach, U-Net, Pyramid Scene Parsing Network, LinkNet, and Feature Pyramid Network segmentation methods were modified with 8 different pre-trained variants of the EfficientNet deep learning model to obtain improved models. In the malaria segmentation performed with these models, the highest Dice score of 91.50 was achieved in the use of the U-Net model with EfficientNetB6. This model offers a faster and more robust solution to detecting parasites compared to traditional methods.

 Malaria is a disease caused by the Plasmodium parasite, which is common in the tropics. The traditional methods commonly used to diagnose malaria, one of the worlds deadliest diseases, are microscopic diagnostic methods in which blood samples taken from suspected individuals are manually examined, or rapid diagnostic tests that are sensitive to human errors. These processes are inexpensive, but experienced and qualified clinicians are needed. Due to this shortcoming, modern diagnostic tools are crucial in the struggle against the disease. In this study, an approach based on deep learning methods was used, which offers beneficial solutions in the diagnosis of disease from medical images. In the proposed approach, U-Net, Pyramid Scene Parsing Network, LinkNet, and Feature Pyramid Network segmentation methods were modified with 8 different pre-trained variants of the EfficientNet deep learning model to obtain improved models. In the malaria segmentation performed with these models, the highest Dice score of 91.50 was achieved in the use of the U-Net model with EfficientNetB6. This model offers a faster and more robust solution to detecting parasites compared to traditional methods.

 Sıtma, tropik bölgelerde yaygın olan Plasmodium parazitinin neden olduğu bir hastalıktır. Dünyanın en ölümcül hastalıklarından biri olan sıtmanın teşhisinde yaygın olarak kullanılan geleneksel yöntemler, şüpheli kişilerden alınan kan örneklerinin manuel olarak incelendiği mikroskobik teşhis yöntemleri veya insan hatalarına duyarlı hızlı teşhis testleridir. Bu işlemler ucuzdur, ancak deneyimli ve nitelikli klinisyenlere ihtiyaç vardır. Bu eksiklik nedeniyle, modern teşhis araçları hastalıkla mücadelede çok önemlidir. Bu çalışmada tıbbi görüntülerden hastalık teşhisinde faydalı çözümler sunan derin öğrenme yöntemlerine dayalı bir yaklaşım kullanılmıştır. Önerilen yaklaşımda, U-Net, Pyramid Scene Parsing Network, LinkNet ve Feature Pyramid Network segmentasyon yöntemleri, EfficientNet derin öğrenme modelinin 8 farklı önceden eğitilmiş varyantı ile modifiye edilerek gelişmiş modeller elde edilmiştir. Bu modeller ile yapılan sıtma segmentasyonunda 91,50 ile en yüksek Dice skoru EfficientNetB6 ile U-Net modelinin kullanımında elde edilmiştir. Bu model, geleneksel yöntemlere kıyasla parazitleri tespit etmek için daha hızlı ve daha sağlam bir çözüm sunar.

Makale detayları

Performance Analysis of EfficientNet Based Segmentation Models for Automatic Detection of Malaria Disease

Yazarlar:

Murat Uçar | Emine Uçar

Yayıncı:

Gazi University

Ana Başlıklar:

Bilgisayar Yazılımı

Dergi:

Bilişim Teknolojileri Dergisi Vol.16 Issue 3, pp.167-176

Belge Tipi:

Makale

Yayın Tarihi:

7/2023

Belge Dili:

İngilizce

Document ID (Org No.):

2001293445

pISSN:

1307-9697

Malaria detection

deep learning

segmentation

EfficientNetSıtma tespiti

derin öğrenme

segmentasyon

EfficientNet

Sayı

Makale detayları

Yazarlar: Murat Uçar | Emine Uçar

Yayıncı: Gazi University

Ana Başlıklar: Bilgisayar Yazılımı

Dergi: Bilişim Teknolojileri Dergisi Vol.16 Issue 3, pp.167-176

Belge Tipi: Makale

Yayın Tarihi: 7/2023

Belge Dili: İngilizce

Document ID (Org No.): 2001293445

pISSN: 1307-9697

Sayı

Yazarlar:

Murat Uçar | Emine Uçar

Yayıncı:

Gazi University

Ana Başlıklar:

Bilgisayar Yazılımı

Dergi:

Bilişim Teknolojileri Dergisi Vol.16 Issue 3, pp.167-176

Belge Tipi:

Makale

Yayın Tarihi:

7/2023

Belge Dili:

İngilizce

Document ID (Org No.):

2001293445

pISSN:

1307-9697