Microscope image processing for TB diagnosis using shape features and ellipse fitting

Reshma, S.; Beegum, T Rehannara

doi:10.1109/spices.2017.8091342

Cited by 10 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…They then resized the selected region to 20 × 20 and binary classified it with five layers of CNNs, achieving accuracy of 98.4% [ 23 , 24 ]. Reshma SR et al demonstrated that contour extraction, ellipse detection, and ellipse merging techniques could be used to count M. tuberculosis bacteria on 176 images with 91.5% accuracy [ 25 ]. Marios Zachariou et al demonstrated that ResNet50 can detect M. tuberculosis with 99.74% accuracy in 230 fluorescently stained microscope slides after segmentation by Cycle-Gan and classification of the extracted regions using ResNet, DenseNet, and SqueezeNet [ 26 ].…”

Section: Related Workmentioning

confidence: 99%

A Study of Mycobacterium tuberculosis Detection Using Different Neural Networks in Autopsy Specimens

Lee

2023

Diagnostics

View full text Add to dashboard Cite

Tuberculosis (TB) presents a substantial health risk to autopsy staff, given its three to five times higher incidence of TB compared to clinical staff. This risk is notably accentuated in South Korea, which reported the highest TB incidence rate and the third highest TB mortality rate among OECD member countries in 2020. The standard TB diagnostic method, histopathological examination of sputum or tissue for acid-fast bacilli (AFB) using Ziehl–Neelsen staining, demands microscopic examination of slides at 1000× magnification, which is labor-intensive and time-consuming. This article proposes a computer-aided diagnosis (CAD) system designed to enhance the efficiency of TB diagnosis at magnification less than 1000×. By training nine neural networks with images taken from 30 training slides and 10 evaluation slides at 400× magnification, we evaluated their ability to detect M. tuberculosis. The N model achieved the highest accuracy, with 99.77% per patch and 90% per slide. We discovered that the model could aid pathologists in preliminary TB screening, thereby reducing diagnostic time. We anticipate that this research will contribute to minimizing autopsy staff’s infection risk and rapidly determining the cause of death.

show abstract

Section: Related Workmentioning

confidence: 99%

A Study of Mycobacterium tuberculosis Detection Using Different Neural Networks in Autopsy Specimens

Lee

2023

Diagnostics

View full text Add to dashboard Cite

show abstract

“…Manual checking of Mycobacterium tuberculosis bacteria using a microscope also requires a lot of concentration of mind and body, for that it is necessary to develop an automatic detection system for Mycobacterium tuberculosis bacteria [10]. Mycobacterium tuberculosis detection results with automated procedures can improve accuracy and faster time [11].…”

Section: Introductionmentioning

confidence: 99%

Comparison of Mycobacterium Tuberculosis Image Detection Accuracy Using CNN and Combination CNN-KNN

Waluyo¹,

Isnanto²,

Rochim³

2023

J. RESTI (Rekayasa Sist. Teknol. Inf.)

View full text Add to dashboard Cite

Mycobacterium tuberculosis is a pathogenic bacterium that causes respiratory tract disease in the lungs, namely tuberculosis (TB). The problem is to find out the bacterial colonies when the observation is still done manually using a microscope with a magnification of 1000 times. It took a long time and was tiring for the observer's eye. Based on this background, an automatic detection system for Mycobacterium tuberculosis was designed. Mycobacterium tuberculosis image data were obtained from the Semarang City Health Center. The dataset used is 220 sputum images, which are divided into 180 training data and 40 testing data. The method used in this research is a combination of Convolutional Neural Network (CNN) and K-Nearest Neighbor (KNN). CNN is used for image feature extraction. Furthermore, the results of the CNN feature extraction are classified using the KNN. The results of the accuracy of the combination of CNN-KNN and CNN were also compared. The stages of the process are color transformation, feature extraction, and data training with CNN, then classification with KNN. The results of the classification test between CNN and the CNN-KNN combination show that the CNN-KNN combination is better. The result of CNN-KNN accuracy is 92.5%, while CNN's accuracy is 90%.

show abstract

“…Selain membutuhkan keterampilan analis laboratorium untuk membuat sediaan mikroskopis pada preparat, untuk membaca gambaran mikroskopis sediaan hapusan darah atau sputum juga dibutuhkan ketelitian mata untuk melihat berbagai bentuk parasit malaria atau bakteri tuberculosis di bawah lensa objektif mikroskop secara terus menerus [6,7]. Pembacaan sediaan pada preparat ini seringkali bersifat subjektif bagi masing-masing pemeriksa sehingga sering terdapat perbedaan pendapat pada pengambilan kesimpulan berapa jumlah keseluruhan parasit malaria ataupun bakteri tuberculosis pada satu sediaan mikroskopik [8,9]. Untuk mengurangi kesalahan akibat subjektivitas pembacaan sediaan mikroskopis, diperlukan seorang analis ahli yang berpengalaman membaca slide hapusan darah malaria atau sputum tuberculosis sebagai standar dan rujukan bagi para analis lainnya, yang disebut sebagai cross-checker [10].…”

unclassified

Pemeriksaan Citra Mikroskop Menggunakan Graphical User Interface dengan Python pada Raspberry Pi

Hidayat,

Muttaqin,

Algifari

et al. 2024

View full text Add to dashboard Cite

Pengamatan objek mikroskopis seperti bakteri, parasit, atau virus menggunakan mikroskop cahaya untuk tujuan diagnosis berbagai penyakit memerlukan ketelitian yang tinggi, sangat melelahkan, memerlukan waktu yang lama, serta subjektifitas pengamat juga cukup tinggi. Oleh karena itu, mikroskop digital, otomatis, portabel, dan compact dirancang berbasis komputer mini raspberry pi agar mengurangi kelelahan dan subjektivitas dalam pengamatan sampel berukuran mikroskopis dalam jumlah yang banyak. Untuk mendukung alat bantu tersebut, platform graphical user interface (GUI) dirancang menggunakan bahasa pemrograman python untuk memfasilitasi interaksi pengamat dan mikroskop digital, sehingga dapat menyimpan citra pengamatan sampel laboratorium dalam format digital dan dapat dilakukan operasi pengolahan citra maupun pengujian model Convolutional Neural Network (CNN) pada layar LCD berbasis raspbery pi. Pengujian penggunaan platform GUI, pengolahan citra, dan model CNN telah dilakukan pada raspberry pi melalui interaksi lewat layar LCD. Operasi pengolahan citra dan penghitungan jumlah objek dengan nilai akurasi 56,21%, serta pengujian model CNN dengan nilai akurasi 98,76% dapat dieksekusi dengan baik pada platform GUI ini dengan waktu eksekusi selama 2-3 detik. Dengan dirancangnya platform GUI pada mikroskop digital berbasis komputer mini raspberry pi ini diharapkan dapat mempermudah tenaga kesehatan untuk menginterpretasi hasil pemeriksaan laboratorium dibandingkan dengan inspeksi manual sampel melalui penglihatan yang memerlukan ketelitian tinggi serta melelahkan sehingga ia dapat meningkatkan layanan kesehatan di berbagai daerah terpencil di Indonesia dan berkontribusi dalam penurunan dan eliminasi berbagai penyakit.

show abstract

Microscope image processing for TB diagnosis using shape features and ellipse fitting

Cited by 10 publications

References 11 publications

A Study of Mycobacterium tuberculosis Detection Using Different Neural Networks in Autopsy Specimens

A Study of Mycobacterium tuberculosis Detection Using Different Neural Networks in Autopsy Specimens

Comparison of Mycobacterium Tuberculosis Image Detection Accuracy Using CNN and Combination CNN-KNN

Pemeriksaan Citra Mikroskop Menggunakan Graphical User Interface dengan Python pada Raspberry Pi

Contact Info

Product

Resources

About