A Novel Architecture for Improving Tuberculosis Detection from Microscopic Sputum Smear Images

Angayarkanni, S. Pitchumani; Vanitha, V.; Karan, V.; Sivant, M.

doi:10.1007/978-3-031-12413-6_5

Cited by 4 publications

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In assessing the chest radiograph to assess the strength of the exposure can be seen from the lower thoracic spine in which images are taken during inspiration (diaphragm at the level of the 5th or 6th rib anterior) and rotation (processus spinosus of the upper thoracic vertebra is felt in the middle of the medial end of the clavicle). Based on the radiologists' reading of the radiographs, most diagnose in the form of infiltrates in unilateral lung apex or the presence of cavities, pleural effusions, infiltrates in the inferior lobes, and lymphadenopathy images without cavitation and infiltrates in the upper lobes of the lungs in which lymphadenopathy is the most common manifestation of TB [28].This image is formed due to the presence of a primary complex, namely a combination of primary focus, enlarged regional lymph nodes (lymphadenitis), and inflamed lymph ducts (lymphangitis) [18].…”

Section: Results Of Radiograph Optical Density In Pneumonia (Tb) Casesmentioning

confidence: 99%

Determination Image Quality on Thorax COVID-19 and Tuberculosis Using Optical Density Image Analysis

Sari¹,

Mayani²,

Putri³

et al. 2023

AMM

View full text Add to dashboard Cite

Covid-19 is a type of pneumonia disease currently affecting people around the world. Covid-19 and tuberculosis are lung diseases that are dangerous and spread quickly to other humans. Chest radiography is one of the main modalities in the management of suspected COVID-19 and tuberculosis patients as it provides radiological information on lung infections that can be used as diagnostic guidelines and patient care. This study aims to determine the differences in optical density and contrast in Covid-19 and pneumonia tuberculosis cases in order to find the specific characteristics of Covid-19 and tuberculosis. This research processed secondary data from Covid-19 and tuberculosis positive patients using image-J software that can be easily obtained and operated by anyone. After processing, the density and contrast were analyzed, particularly for the lungs. Based on image processing results, the average density for Cnovid-19 radiographs is 1,066, while for Tuberculosis radiograph is 1,519. The average contrast values for Covid-19 and tuberculosis radiographs are 0.37 and 1.03. Thus, it can be concluded that the contrast of the Covid-19 radiograph is lower than the tuberculosis radiograph. The difference in optical density on the tuberculosis and Covid-19 cases is 0.5 as the opacity on the chest radiographs of Covid-19 patients is evenly distributed over the entire lung surface.

show abstract

Section: Results Of Radiograph Optical Density In Pneumonia (Tb) Casesmentioning

confidence: 99%

Determination Image Quality on Thorax COVID-19 and Tuberculosis Using Optical Density Image Analysis

Sari¹,

Mayani²,

Putri³

et al. 2023

AMM

View full text Add to dashboard Cite

show abstract

“…Furthermore, the U-Net proposed by Ronneberger et al [ 13 ] adds the “skip connection” to FCN to alleviate the loss of shallow information, which has demonstrated excellent performance on many segmentation tasks, and U-Net-based methods have been applied to many segmentation tasks [ 14 – 16 ]. Angayarkanni et al [ 17 ] enhanced the quality of sputum smears using several image preprocessing methods and then fed the preprocessed images to U-Net for Mycobacterium tuberculosis segmentation. Ali et al [ 18 ] proposed a modified U-Net and applied it to the segmentation of bacilli in ZN-stained sputum smear images.…”

Section: Introductionmentioning

confidence: 99%

DB-DCAFN: dual-branch deformable cross-attention fusion network for bacterial segmentation

Wang

Cao

et al. 2023

Vis. Comput. Ind. Biomed. Art

View full text Add to dashboard Cite

Sputum smear tests are critical for the diagnosis of respiratory diseases. Automatic segmentation of bacteria from sputum smear images is important for improving diagnostic efficiency. However, this remains a challenging task owing to the high interclass similarity among different categories of bacteria and the low contrast of the bacterial edges. To explore more levels of global pattern features to promote the distinguishing ability of bacterial categories and maintain sufficient local fine-grained features to ensure accurate localization of ambiguous bacteria simultaneously, we propose a novel dual-branch deformable cross-attention fusion network (DB-DCAFN) for accurate bacterial segmentation. Specifically, we first designed a dual-branch encoder consisting of multiple convolution and transformer blocks in parallel to simultaneously extract multilevel local and global features. We then designed a sparse and deformable cross-attention module to capture the semantic dependencies between local and global features, which can bridge the semantic gap and fuse features effectively. Furthermore, we designed a feature assignment fusion module to enhance meaningful features using an adaptive feature weighting strategy to obtain more accurate segmentation. We conducted extensive experiments to evaluate the effectiveness of DB-DCAFN on a clinical dataset comprising three bacterial categories: Acinetobacter baumannii, Klebsiella pneumoniae, and Pseudomonas aeruginosa. The experimental results demonstrate that the proposed DB-DCAFN outperforms other state-of-the-art methods and is effective at segmenting bacteria from sputum smear images.

show abstract

AI-Based Analysis of Ziehl–Neelsen-Stained Sputum Smears for Mycobacterium tuberculosis as a Screening Method for Active Tuberculosis

Witarto,

Ceachi,

Popp

et al. 2024

Life

View full text Add to dashboard Cite

Tuberculosis is the primary cause of death due to infection in the world. Identification of Mycobacterium tuberculosis in sputum is a diagnostic test, which can be used in screening programs—especially in countries with a high incidence of tuberculosis—to identify and treat those persons with the highest risk of disseminating the infection. We previously developed an algorithm which is able to automatically detect mycobacteria on tissue; in particular, our algorithm identified acid-fast bacilli on tissue with 100% specificity, 95.65% sensitivity, and 98.33% accuracy. We tested this algorithm on 1059 Ziehl–Neelsen-stained sputum smears to evaluate its results as a possible tool for screening. The results were displayed as a heat map of 32 × 32 pixel patches. Analysis of the positive patches revealed a good specificity (86.84%) and 100% sensitivity for patches with a level of confidence over 90; furthermore, the accuracy remained over 95% for all levels of confidence over 80, except the class (95–100]. The modest specificity is caused by the peculiarities of smears (uneven thickness, dust contamination, lack of coverslip). We will train the algorithm on sputum smears to increase the specificity to over 95%. However, as our algorithm showed no false negatives, it is suitable for screening.

show abstract

A Novel Architecture for Improving Tuberculosis Detection from Microscopic Sputum Smear Images

Cited by 4 publications

References 17 publications

Determination Image Quality on Thorax COVID-19 and Tuberculosis Using Optical Density Image Analysis

Determination Image Quality on Thorax COVID-19 and Tuberculosis Using Optical Density Image Analysis

DB-DCAFN: dual-branch deformable cross-attention fusion network for bacterial segmentation

AI-Based Analysis of Ziehl–Neelsen-Stained Sputum Smears for Mycobacterium tuberculosis as a Screening Method for Active Tuberculosis

Contact Info

Product

Resources

About