Red blood cell classification on thin blood smear images for malaria diagnosis

Sunarko, Budi; Djuniadi, D.; Bottema, Murk J.; Iksan, Nur; Hudaya, Khakim A.N.; Hanif, Muhammad

doi:10.1088/1742-6596/1444/1/012036

Cited by 12 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The use of softmax activation in this study as a probability function where this function predicts the output of a particular class with a predicted value (Mustafa et al, 2018). As previously described, we used categorical cross-entropy to determine the activation loss function, with an optimization algorithm, namely the stochastic gradient optimizer (SGD) (Sunarko et al, 2020). The adjustment of the learning rate is 0.001 and momentum 0.5.…”

Section: Fig 3 Proposed Cnn Model Architecturementioning

confidence: 99%

Image Recognition of Malaria-infected Red Blood Cells among Other Normal and Cancer-Mutated Cells Using CNN

Nivaan

2022

jinav j. inf. vis.

View full text Add to dashboard Cite

Malaria is a contagious infectious disease that is still threatening human life. Malaria morbidity when viewed by province shows that Eastern Indonesia is the area with the highest Annual Parasite Incidence (API), namely Papua, West Papua, NTT, and Maluku. This is a concern for the continued efforts to control and eliminate malaria in these high malaria-endemic areas. There are many strategies to help and prevent, include the possibility of innovation in the diagnostic process. Therefore, to answer how to provide innovation in technology to accelerate the elimination of malaria, this study aims to identify the image of red blood cells which infected with malaria among other normal and leukemia cancer-mutated cells (non-malaria) by making improvements through the proposed new model used. This model is meant to do deep learning using Convolutional Neural Network (CNN). The results obtained in this study show that the success of using the proposed model is influenced by the pre-processing stage, the dropout regularization function, learning rate, and momentum value used. The accuracy value obtained is 0.9660, 0.9693 precision, 0.9626 recall, and an F1 score of 0.9659.

show abstract

Section: Fig 3 Proposed Cnn Model Architecturementioning

confidence: 99%

Image Recognition of Malaria-infected Red Blood Cells among Other Normal and Cancer-Mutated Cells Using CNN

Nivaan

2022

jinav j. inf. vis.

View full text Add to dashboard Cite

show abstract

“…In [13], the Otsu thresholding segmentation method generated a binary image to separate red blood cells and malaria parasites. The method succeeded in achieving 94.60% accuracy in classification.…”

Section: Introductionmentioning

confidence: 99%

Kinerja Algoritme Pengelompokan Fuzzy C-Means pada Segmentasi Citra Leukosit

Hudaya

Sunarko

Nugroho

2022

JNTETI

View full text Add to dashboard Cite

Segmentasi citra merupakan salah satu tahap paling penting dalam computer-aided diagnosis yang berpotensi mempercepat proses diagnosis leukemia. Leukemia merupakan salah satu penyakit mematikan yang masuk ke dalam kategori kanker darah. Umumnya, pendeteksian leukemia limfoblastik akut (LLA) dilakukan secara manual di rumah sakit dengan cara menghitung secara manual sel leukosit yang terdapat pada citra apusan darah tepi hasil pewarnaan dengan metode immunohistochemical (IHC). Sayangnya, proses diagnosis manual memakan waktu 3−24 jam hingga hasil dapat diketahui dan cenderung tidak akurat karena tingkat kelelahan operator. Hasil yang tidak akurat dapat mengakibatkan kesalahan diagnosis yang berakibat fatal pada pasien. Sebuah metode segmentasi citra yang diusulkan oleh Vogado mampu mencapai akurasi 98,5% untuk menyegmentasi dataset ALL-IDB2. Namun, metode ini menggunakan algoritme pengelompokan k-means yang tidak optimal dalam menangani masukan citra yang memiliki intensitas derau yang tinggi. Dalam makalah ini, fuzzy c-means diterapkan guna mengetahui perubahan kinerja pada metode segmentasi tersebut. Dataset yang digunakan adalah ALL-IDB2 yang berjumlah 260 citra dengan ukuran masing-masing citra adalah 257×257 piksel dalam format tagged image file (TIF). Tahap awal metode ini adalah membagi citra dataset leukemia akut ALL-IDB2 menjadi skema warna cyan, magenta, yellow, key (CMYK) dan L*a*b yang kemudian mensubstraksi komponen M, yaitu dikurangkan dengan komponen *b. Hasil pengurangan tersebut kemudian dipisahkan dengan algoritme fuzzy c-means (FCM), yang kemudian menghasilkan keluaran citra yang hanya mengandung bagian nukleus dan latar belakang. Keluaran metode ini kemudian dievaluasi dan diukur menggunakan metrik accuracy, specificity, sensitivity, kappa index, dice-coefficient, dan time complexity. Hasil analisis menunjukkan bahwa pengubahan algoritme pengelompokan pada metode segmentasi citra tersebut tidak memberikan perubahan hasil yang signifikan, dengan rincian peningkatan terjadi pada metrik specificity dan precision rata-rata 0,1-0,4%, waktu eksekusi bertambah lama rata-rata 23,10%, dan penurunan terjadi pada metrik accuracy, yaitu turun menjadi 95,4238% dan nilai dice-coefficient 79,3682%. Maka, dapat disimpulkan bahwa penerapan algoritme FCM pada metode segmentasi yang diusulkan tidak memberikan peningkatan kinerja yang optimal.

show abstract

“…For example, the traditional machine vision method requires the design of complex feature extractors (such as morphological features and texture features), and a large number of images need to be preprocessed before training [ 5–8 ]. Previous studies mainly focused on the recognition and property retrieval of single-cell types [ 9 , 10 ], and few studies have focused on automatic recognition and localization of other common cells, such as epithelial cells (Epi cells), red blood cells (RBCs), white blood cells (WBCs) and molds.…”

Section: Introductionmentioning

confidence: 99%

Morphological components detection for super-depth-of-field bio-micrograph based on deep learning

Wang

et al. 2021

Microscopy

View full text Add to dashboard Cite

Accompanied with the clinical routine examination demand increase sharply, the efficiency and accuracy are the first priority. However, automatic classification and localization of cells in microscopic images in super depth of Field (SDoF) system remains great challenges. In this paper, we advance an object detection algorithm for cells in the SDoF micrograph based on Retinanet model. Compared with the current mainstream algorithm, the mean average precision (mAP) index is significantly improved. In the experiment of leucorrhea samples and fecal samples, mAP indexes are 83.1% and 88.1%, respectively, with an average increase of 10%. The object detection model proposed in this paper can be applied to feces and leucorrhea detection equipment, and significantly improve the detection efficiency and accuracy.

show abstract

Red blood cell classification on thin blood smear images for malaria diagnosis

Cited by 12 publications

References 15 publications

Image Recognition of Malaria-infected Red Blood Cells among Other Normal and Cancer-Mutated Cells Using CNN

Image Recognition of Malaria-infected Red Blood Cells among Other Normal and Cancer-Mutated Cells Using CNN

Kinerja Algoritme Pengelompokan Fuzzy C-Means pada Segmentasi Citra Leukosit

Morphological components detection for super-depth-of-field bio-micrograph based on deep learning

Contact Info

Product

Resources

About