An ANN model for the differential diagnosis of tuberculosis and sarcoidosis

Vijayaraj, Mahalakshmi; Abhinand, P A; Venkatesan, P; Ragunath, P K

doi:10.6026/97320630016539

Cited by 4 publications

(4 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, due to the continuous deepening of clinicians' understanding of SA and the gradual improvement of examination techniques, its rate of diagnosis has risen significantly. The diagnosis of SA often depends on radiological and clinical imaging, which are related to the histology of epithelioid granulomas; nevertheless, granulomas are not pathognomonic of SA (22). In particular, the imaging features of pulmonary SAs are similar to those of pulmonary TB and mediastinal lymph node TB (23).…”

Section: Discussionmentioning

confidence: 99%

BATF2 and PDK4 as diagnostic molecular markers of sarcoidosis and their relationship with immune infiltration

He¹,

Li²,

Zhou³

et al. 2022

Ann Transl Med

View full text Add to dashboard Cite

Background: Sarcoidosis (SA) is an immune disorder disease featuring granulomas formation with a controversial etiopathogenesis. We aimed to uncover potential markers for SA diagnosis and explore how immune cell infiltration contributes to the pathogenesis of SA.Methods: Sarcoidosis GSE83456 samples and GSE42834 from Gene Expression Omnibus (GEO) were analyzed as the training and external validation sets, respectively. Firstly, R statistical software was employed to uncover the differentially expressed genes (DEGs) of GSE83456. Weighted gene co-expression network analysis (WGCNA) was used to reveal the key module of DEGs. Secondly, the genes of the key module were used to analyze functional correlations. Thirdly, support vector machine (SVM) algorithms and least absolute shrinkage and selection operator (LASSO) logistic regression were applied for screening and verification of the diagnostic markers for key module genes. Finally, the infiltration of immune cells in SA patients' blood samples was assessed by Cell-type Identification by Estimating Relative Subsets of RNA Transcripts (CIBERSORT), after which correlations of diagnostic markers immune cell and infiltration were explored. Serum samples collected from human research participants were used for confirmatory analysis of bioinformatics. A receiver operating characteristic (ROC) curve was used to evaluate the diagnostic value of BATF2 and PDK4.Results: In total, 580 DEGs were identified from the key module. The genes PDK4 [area under the curve (AUC) =0.942] and BATF4 (AUC =0.980) were revealed as diagnostic markers of SAs. The diagnostic power of the 2 genes was verified in GSE42834. We found that monocytes, T cells regulatory (Tregs), mast cells, macrophages, natural killer (NK) cells, and dendritic cells (DC) may contribute to SA development through immune cell infiltration study. In addition, PDK4 and BATF4 were closely associated with these immune cells.Both BATF2 and PDK4 were highly expressed in active pulmonary SA. Using non-SA patients as controls, the AUC of serum PDK4, BATF2, and their combination for the diagnosis of active pulmonary SA was 0.798 (95% CI: 0.701 to 0.876), 0.895 (95% CI: 0.813 to 0.950), and 0.910 (95% CI: 0.831 to 0.960), respectively. Conclusions:The genes PDK4 and BATF4 could be used as diagnostic markers of active pulmonary SA.Immune cell infiltration severs an important role in SA.

show abstract

Section: Discussionmentioning

confidence: 99%

BATF2 and PDK4 as diagnostic molecular markers of sarcoidosis and their relationship with immune infiltration

He¹,

Li²,

Zhou³

et al. 2022

Ann Transl Med

View full text Add to dashboard Cite

show abstract

“…In recent years, due to the continuous deepening of clinicians' understanding of sarcoidosis, and the gradual improvement of examination techniques, the diagnosis rate of sarcoidosis is also signi cantly higher than before. Sarcoidosis diagnosis often is dependents on radiological and clinical imaging, which are related to the histology of epithelioid granulomas; nevertheless, granulomas are not unique pathognomonic for sarcoidosis [22]. In particular, the imaging features of pulmonary sarcoidosis are similar to those of pulmonary tuberculosis and mediastinal lymph node tuberculosis [23].…”

Section: Discussionmentioning

confidence: 99%

BATF2 and PDK4 As Diagnostic Molecular Markers of Sarcoidosis and Their Relationship With Immune Infiltration

Li¹,

Zhou²,

He³

2021

Preprint

View full text Add to dashboard Cite

Background: Sarcoidosis (SA) is an immune disorder disease featured with granulomas formation. The work purposed to uncover potential markers for sarcoidosis (SA) diagnosis and explore how immune cell infiltration contributes to the pathogenesis of SA.Methods: Sarcoidosis GSE83456 samples and GSE42834 from Gene Expression Omnibus (GEO) were analyzed as the training and external validation sets, respectively. R statistical software was employed to uncover the differentially expressed genes (DEGs) of GSE83456. SVM algorithms and LASSO logistic regression were applied for screening and verification of the diagnostic markers for key module genes. The infiltration of immune cells in sarcoidosis patients’ blood samples was assessed by CIBERSORT. The expression of serum BATF2 and PDK4 was detected by RT-qPCR method, and the value of BATF2 and PDK4 mRNA expression in the diagnosis of pulmonary sarcoidosis was analyzed.Results: In total, 580 DEGs were identified from the key module. PDK4 (AUC=0.942) and BATF4 (AUC=0.980) were revealed as diagnostic markers of sarcoidosis. We found that monocytes, T cells regulatory (Tregs), mast cells, macrophages，NK cells, and dendritic cells may contribute to sarcoidosis development. In addition, PDK4 and BATF4 were closely associated with these immune cells. BATF2 and PDK4 were highly expressed in pulmonary sarcoidosis. BATF2 and PDK4 combined to predict the area under the ROC curve of pulmonary sarcoidosis was 0.922.Conclusions: PDK4 and BATF4 could be used as diagnostic markers of sarcoidosis, and immune cell infiltration severs an important role in sarcoidosis.

show abstract

“…On labeled tuberculosis culture images, Chang et al [ 12 ] employed the transfer learning method and achieved sensitivity and precision rates of 98 and 99%, respectively. Mahalakshmi et al [ 11 ] used an Artificial Neural Network (ANN) to identify tuberculosis and sarcoidosis based on gene expression. Several machine learning algorithms were compared by Chen et al [ 36 ].…”

Section: Related Workmentioning

confidence: 99%

“…In several studies, MTB deoxyribonucleic acid (DNA) was associated with a large proportion of tissue and bronchoalveolar lavage samples from sarcoidosis patients. There have, however, been reports of deleterious consequences [ 2 , 11 ]. Differentiating tuberculosis and sarcoidosis can be difficult, especially in cases of mediastinal lymphadenopathy, because both diseases have similar clinical presentations and histopathologically identical granulomatous inflammation [ 12 ].…”

Section: Introductionmentioning

confidence: 99%

Reliable Sarcoidosis Detection Using Chest X-rays with EfficientNets and Stain-Normalization Techniques

Baghdadi

Maklad

Malki

et al. 2022

Sensors

View full text Add to dashboard Cite

Sarcoidosis is frequently misdiagnosed as tuberculosis (TB) and consequently mistreated due to inherent limitations in radiological presentations. Clinically, to distinguish sarcoidosis from TB, physicians usually employ biopsy tissue diagnosis and blood tests; this approach is painful for patients, time-consuming, expensive, and relies on techniques prone to human error. This study proposes a computer-aided diagnosis method to address these issues. This method examines seven EfficientNet designs that were fine-tuned and compared for their abilities to categorize X-ray images into three categories: normal, TB-infected, and sarcoidosis-infected. Furthermore, the effects of stain normalization on performance were investigated using Reinhard’s and Macenko’s conventional stain normalization procedures. This procedure aids in improving diagnostic efficiency and accuracy while cutting diagnostic costs. A database of 231 sarcoidosis-infected, 563 TB-infected, and 1010 normal chest X-ray images was created using public databases and information from several national hospitals. The EfficientNet-B4 model attained accuracy, sensitivity, and precision rates of 98.56%, 98.36%, and 98.67%, respectively, when the training X-ray images were normalized by the Reinhard stain approach, and 97.21%, 96.9%, and 97.11%, respectively, when normalized by Macenko’s approach. Results demonstrate that Reinhard stain normalization can improve the performance of EfficientNet -B4 X-ray image classification. The proposed framework for identifying pulmonary sarcoidosis may prove valuable in clinical use.

show abstract

An ANN model for the differential diagnosis of tuberculosis and sarcoidosis

Abstract: The authors are responsible for the content of this article. The Editorial and the publisher has taken reasonable steps to check the content of the article with reference to publishing ethics with adequate peer reviews deposited at PUBLONS.

Cited by 4 publications

References 26 publications

BATF2 and PDK4 as diagnostic molecular markers of sarcoidosis and their relationship with immune infiltration

BATF2 and PDK4 as diagnostic molecular markers of sarcoidosis and their relationship with immune infiltration

BATF2 and PDK4 As Diagnostic Molecular Markers of Sarcoidosis and Their Relationship With Immune Infiltration

Reliable Sarcoidosis Detection Using Chest X-rays with EfficientNets and Stain-Normalization Techniques

Contact Info

Product

Resources

About