An Efficient Approach for Tuberculosis Diagnosis on Chest X-Ray

Ngoc, Hoang Nguyen; Hoang, Vu; Bui, Trung; Truong, Steven Q. H.; Minh, Thanh Huynh; Van, Duong Nguyen; Minh, Trang Nguyen Thi; Van, Cong Cung

doi:10.1109/isbi52829.2022.9761426

Cited by 4 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…While the model was tested on the Shenzhen and Tuberculosis X-ray (TBX11K) datasets, it can be argued that the results need to be improved to obtain convincing scores. A new segmentation model was proposed for the segmentation of tuberculosis by using the Lightweight U-Net model ( Ngoc et al, 2022 ). The Lightweight U-Net model focused on standard up-convolution and skip connections in the study.…”

Section: Rngu-netmentioning

confidence: 99%

See 1 more Smart Citation

RNGU-NET: a novel efficient approach in Segmenting Tuberculosis using chest X-Ray images

Turk

2024

PeerJ Computer Science

View full text Add to dashboard Cite

Tuberculosis affects various tissues, including the lungs, kidneys, and brain. According to the medical report published by the World Health Organization (WHO) in 2020, approximately ten million people have been infected with tuberculosis. U-NET, a preferred method for detecting tuberculosis-like cases, is a convolutional neural network developed for segmentation in biomedical image processing. The proposed RNGU-NET architecture is a new segmentation technique combining the ResNet, Non-Local Block, and Gate Attention Block architectures. In the RNGU-NET design, the encoder phase is strengthened with ResNet, and the decoder phase incorporates the Gate Attention Block. The key innovation lies in the proposed Local Non-Local Block architecture, overcoming the bottleneck issue in U-Net models. In this study, the effectiveness of the proposed model in tuberculosis segmentation is compared to the U-NET, U-NET+ResNet, and RNGU-NET algorithms using the Shenzhen dataset. According to the results, the RNGU-NET architecture achieves the highest accuracy rate of 98.56%, Dice coefficient of 97.21%, and Jaccard index of 96.87% in tuberculosis segmentation. Conversely, the U-NET model exhibits the lowest accuracy and Jaccard index scores, while U-NET+ResNet has the poorest Dice coefficient. These findings underscore the success of the proposed RNGU-NET method in tuberculosis segmentation.

show abstract

Section: Rngu-netmentioning

confidence: 99%

“…The lightweight U-Net method was also used for segmentation using the Shenzhen dataset. This approach is the modification of the U-Net model to reduce the latency ( Ngoc et al, 2022 ). The employed lightweight U-Net provided a Dice value of 0.7252 for the Shenzhen dataset.…”

Section: Introductionmentioning

confidence: 99%

RNGU-NET: a novel efficient approach in Segmenting Tuberculosis using chest X-Ray images

Turk

2024

PeerJ Computer Science

View full text Add to dashboard Cite

show abstract

“…Diversas técnicas computacionais e métodos de inteligência artificial têm sido apresentadas para este fim, utilizando desde aprendizado de máquina clássico até o aprendizado profundo (DL) (5,13,14,15,16,17) . Além do método de aprendizado de máquina, tais abordagens variam entre si na extração de recursos das imagens de RXT, e também na seleção do conjunto final, que pode ocorrer de forma manual ou automática (7,8,18) .…”

Section: Introductionunclassified

Diagnosticando Tuberculose com Redes Neurais Artificiais e Recursos BPPC

et al. 2023

View full text Add to dashboard Cite

Tuberculose é uma doença grave e contagiosa que mata milhões de pessoas, sendo um problema de saúde pública global. Enquanto isso, o uso da Inteligência Artificial na radiologia tem despertado crescente interesse de pesquisadores e da indústria. Soluções para auxiliar no diagnóstico já são uma realidade, mas ainda distante de populações vulneráveis e regiões subdesenvolvidas. Logo, soluções acessíveis são essenciais para populações altamente dependentes de ações e serviços públicos. Assim, propomos um método de baixo custo computacional e alta eficiência para auxiliar no diagnóstico de tuberculose. Utilizamos imagens de radiografia torácicas e construímos um modelo de rede neural artificial com recursos BPPC com e sem a geração de dados sintéticos. Os resultados equivalentes à literatura relacionada mostram o desempenho excepcional e de baixo custo da solução, colocando-a como uma solução alternativa viável.

show abstract

“…Studying the periodicity of solutions is one of the great problems for the qualitative theory of evolution equations. The existence and uniqueness of periodic solutions have been proved for several important classes of densely defined evolution equations by using classical approaches such as the fixed point method [10,19,22], the use of ultimate boundedness of solutions and the compactness of Poincaré map over compact embedding [17,18,20], the spectral theory of functions [11,15,16], ergodic approach [14]. As indicated in [6], we sometimes need to deal with non-densely defined operators.…”

Section: Introductionmentioning

confidence: 99%

“…Luong et al introduced the new condition for the unique existence bounded solution to be asymptotic 1-periodic on the half line. More precisely, they showed that a bounded and continuous function g : R → X is asymptotic 1-periodic if and only if its circular spectrum σ(g) (see [11] for more detail of this notion) satisfies σ(g) ⊂ {1}. Therefore, the existence of asymptotic 1periodic solutions is reduced to that of solutions x(•) such that σ(x(•)) ⊂ {1}.…”

Section: Introductionmentioning

confidence: 99%