An ANN model for the differential diagnosis of tuberculosis and sarcoidosis.
Ontology highlight
ABSTRACT:
SUBMITTER: Vijayaraj M
PROVIDER: S-EPMC7505241 | biostudies-literature | 2020
REPOSITORIES: biostudies-literature
Publications
An ANN model for the differential diagnosis of tuberculosis and sarcoidosis.
Vijayaraj Mahalakshmi M Abhinand P A PA Venkatesan P P Ragunath P K PK
Bioinformation 20200731 7
Sarcoidosis is often misdiagnosed as tuberculosis and consequently mistreated owing to inherent limitations in histopathological and radiological presentations. It is known that the differential diagnosis of Tuberculosis and Sarcoidosis is often non-trivial and requires expertise and experience from clinicians. Therefore, it is of interest to describe a multilayer neural network model to differentiate pulmonary tuberculosis from Sarcoidosis using signal intensity data from blood transcriptional ...[more]
Similar Datasets
Project description:In this study, we sought to identify differentially expressed proteins in the serum of patients with sarcoidosis or tuberculosis and to evaluate these proteins as markers for the differential diagnosis of sarcoidosis and sputum-negative tuberculosis.Using protein microarrays, we identified 3 proteins exhibiting differential expression between patients with sarcoidosis and tuberculosis. Elevated expression of these proteins was verified using the enzyme-linked immunosorbent assay (ELISA) and was further confirmed by immunohistochemistry. Receiver operating characteristic (ROC) curve, logistic regression analysis, parallel, and serial tests were used to evaluate the diagnostic efficacy of the proteins.Intercellular Adhesion Molecule 1(ICAM-1) and leptin were screened for differentially expressed proteins relevant to sarcoidosis and tuberculosis. Using ROC curves, we found that ICAM-1 (cutoff value: 57740 pg/mL) had an area under the curve (AUC), sensitivity, and specificity of 0.718, 62.3%, and 79.5% respectively, while leptin (cutoff value: 1193.186 pg/mL) had an AUC, sensitivity, and specificity of 0.763, 88.3%, and 65.8%, respectively. Logistic regression analysis revealed that the AUC, sensitivity, and specificity of combined leptin and ICAM-1 were 0.787, 89.6%, and 65.8%, respectively, while those of combined leptin, ICAM-1, and body mass index (BMI) were 0.837, 90.9%, and 64.4%, respectively, which had the greatest diagnostic value. Parallel and serial tests indicated that the BMI-leptin parallel with the ICAM-1 serial was the best diagnostic method, achieving a sensitivity and specificity of 86.5% and 73.1%, respectively. Thus, our results identified elevated expression of ICAM-1 and leptin in serum and granulomas of sarcoidosis patients.ICAM-1 and leptin were found to be potential markers for the diagnosis of sarcoidosis and differential diagnosis of sarcoidosis and sputum-negative tuberculosis.
| S-EPMC4569088 | biostudies-literature
Project description:BackgroundThe aim of this study was to evaluate the role of Xpert MTB/RIF assay in the detection of Mycobacterium tuberculosis for differentiating tuberculosis intrathoracic lymphadenopathy from sarcoidosis intrathoracic lymphadenopathy.MethodsThe patients who were suspected to having sarcoidosis or tuberculosis intrathoracic lymphadenopathy at the Shanghai Pulmonary Hospital between October 1, 2020 and June 30, 2021 were retrospectively evaluated in this study. All patients underwent endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) and Xpert analysis. Differences in clinical and radiological features were recorded. The diagnostic performances of EBUS-TBNA Xpert, acid-fast bacilli, culture, and peripheral blood QuantiFERON-TB Gold (QFT) for differentiating sarcoidosis from tuberculosis intrathoracic lymphadenopathy were analyzed.ResultsA total of 119 patients were included in this analysis. Of those, 83 patients were finally diagnosed with sarcoidosis (N = 50) and tuberculosis (N = 33) intrathoracic lymphadenopathy. Young individuals were more likely to have tuberculosis versus sarcoidosis intrathoracic lymphadenopathy (P = 0.006). Markers of inflammation, including fever, leukocytes, and serum ferritin levels, were significantly higher in tuberculosis versus sarcoidosis intrathoracic lymphadenopathy (P < 0.01). Bilateral lung involvement and symmetry intrathoracic lymphadenopathy were more common in sarcoidosis intrathoracic lymphadenopathy (P < 0.01). In addition, the longest diameter of intrathoracic lymphadenopathy (in cm) was significantly larger in sarcoidosis intrathoracic lymphadenopathy (P = 0.001). However, the largest diameter of lung lesions was significantly shorter (P = 0.005). The sensitivity and specificity values of Xpert and QFT for differentiating these two diseases were 69.70% and 100%, and 96.43% and 91.84%, respectively.ConclusionXpert MTB/RIF is recommended for the diagnosis of tuberculosis intrathoracic lymphadenopathy using EBUS-TBNA samples. A negative QFT suggests the exclusion of the diagnosis of tuberculosis intrathoracic lymphadenopathy.
| S-EPMC10601222 | biostudies-literature
Project description:Active pulmonary tuberculosis (ATB), which is more infectious and has a higher mortality rate compared with non-active pulmonary tuberculosis (non-ATB), needs to be diagnosed accurately and timely to prevent the tuberculosis from spreading and causing deaths. However, traditional differential diagnosis methods of active pulmonary tuberculosis involve bacteriological testing, sputum culturing and radiological images reading, which is time consuming and labour intensive. Therefore, an artificial intelligence model for ATB differential diagnosis would offer great assistance in clinical practice. In this study, computer tomography (CT) scans images and corresponding clinical information of 1160 ATB patients and 1131 patients with non-ATB were collected and divided into training, validation, and testing sets. A 3-dimension (3D) Nested UNet model was utilized to delineate lung field regions in the CT images, and three different pre-trained deep learning models including 3D VGG-16, 3D EfficientNet and 3D ResNet-50 were used for classification and differential diagnosis task. We also collected an external testing set with 100 ATB cases and 100 Non-ATB cases for further validation of the model. In the internal and external testing set, the 3D ResNet-50 model outperformed other models, reaching an AUC of 0.961 and 0.946, respectively. The 3D ResNet-50 model reached even higher levels of diagnostic accuracy than experienced radiologists, while the CT images reading and diagnosing speed was 10 times faster than human experts. The model was also capable of visualizing clinician interpretable lung lesion regions important for differential diagnosis, making it a powerful tool assisting ATB diagnosis. In conclusion, we developed an auxiliary tool to differentiate active and non-active pulmonary tuberculosis, which would have broad prospects in the bedside.
| S-EPMC9760807 | biostudies-literature
Project description:Many aspects of pathogenic granuloma formation are poorly understood, requiring new relevant laboratory models that represent the complexity (genetics and diversity) of human disease. To address this need, we developed an in vitro model of granuloma formation using human peripheral blood mononuclear cells (PBMCs) derived from patients with active sarcoidosis, latent tuberculosis (TB) infection (LTBI), or normal healthy control subjects. PBMCs were incubated for 7 days with uncoated polystyrene beads or beads coated with purified protein derivative (PPD) or human serum albumin. In response to PPD-coated beads, PBMCs from donors with sarcoidosis and LTBI formed robust multicellular aggregates resembling granulomas, displaying a typical T-helper cell type 1 immune response, as assessed by cytokine analyses. In contrast, minimal PBMC aggregation occurred when control PBMCs were incubated with PPD-coated beads, whereas the response to uncoated beads was negligible in all groups. Sarcoidosis PBMCs responded to human serum albumin-coated beads with modest cellular aggregation and inflammatory cytokine release. Whereas the granuloma-like aggregates formed in response to PPD-coated beads were similar for sarcoidosis and LTBI, molecular profiles differed significantly. mRNA expression patterns revealed distinct pathways engaged in early granuloma formation in sarcoidosis and LTBI, and they resemble molecular patterns reported in diseased human tissues. This novel in vitro human granuloma model is proposed as a tool to investigate mechanisms of early granuloma formation and for preclinical drug discovery research of human granulomatous disorders. Clinical trial registered with www.clinicaltrials.gov (NCT01857401).
| S-EPMC5650085 | biostudies-literature
Project description:As an ancient infectious disease, tuberculosis (TB) is still the leading cause of death from a single infectious agent worldwide. Latent TB infection (LTBI) has been recognized as the largest source of new TB cases and is one of the biggest obstacles to achieving the aim of the End TB Strategy. The latest data indicate that a considerable percentage of the population with LTBI and the lack of differential diagnosis between LTBI and active TB (aTB) may be potential reasons for the high TB morbidity and mortality in countries with high TB burdens. The tuberculin skin test (TST) has been used to diagnose TB for > 100 years, but it fails to distinguish patients with LTBI from those with aTB and people who have received Bacillus Calmette-Guérin vaccination. To overcome the limitations of TST, several new skin tests and interferon-gamma release assays have been developed, such as the Diaskintest, C-Tb skin test, EC-Test, and T-cell spot of the TB assay, QuantiFERON-TB Gold In-Tube, QuantiFERON-TB Gold-Plus, LIAISON QuantiFERON-TB Gold Plus test, and LIOFeron TB/LTBI. However, these methods cannot distinguish LTBI from aTB. To investigate the reasons why all these methods cannot distinguish LTBI from aTB, we have explained the concept and definition of LTBI and expounded on the immunological mechanism of LTBI in this review. In addition, we have outlined the research status, future directions, and challenges of LTBI differential diagnosis, including novel biomarkers derived from Mycobacterium tuberculosis and hosts, new models and algorithms, omics technologies, and microbiota.
| S-EPMC8570039 | biostudies-literature
Project description:BackgroundThe diagnosis of sarcoidosis is made by the combination of clinical features and biopsy results. The clinical features of sarcoidosis can be quite variable. We developed a Sarcoidosis Diagnostic Score (SDS) to summarize the clinical features of patients with possible sarcoidosis.MethodsBiopsy-confirmed patients with sarcoidosis seen during a 7-month period at the University of Cincinnati sarcoidosis clinic were prospectively identified. Patients with nonsarcoidosis seen at the same clinic were used as control patients. Using a modified World Association of Sarcoidosis and Other Granulomatous Disorders organ assessment instrument, we scored all patients for presence of biopsy, ≥1 highly probable symptom, and ≥1 at least probable symptom for each area. Two sarcoidosis scores were generated: SDS biopsy (with biopsy) and SDS clinical (without biopsy).ResultsThe 980 evaluable patients were divided into two cohorts: an initial 600 patients (450 with biopsy-confirmed sarcoidosis, 150 control patients) to establish cutoff values for SDS biopsy and an SDS clinical and a validation cohort of 380 patients (103 biopsy-confirmed patients with sarcoidosis and 277 control patients). The best cutoff value for SDS biopsy was ≥ 6 (sensitivity, 99.3%; specificity, 100%). For the total the 980 patients, an SDS clinical ≥ 3 had a sensitivity of 90.6%, specificity of 88.5%, and a likelihood ratio of 7.9. An SDS clinical score ≥ 4 had a lower sensitivity of (76.9%) but higher specificity (98.6%).ConclusionsFor sarcoidosis, the presence of specific clinical features, especially multiorgan involvement, can enhance the diagnostic certainty. The SDS scoring system quantitated the clinical features consistent with sarcoidosis.
| S-EPMC6859250 | biostudies-literature
Project description:Sarcoidosis is a disease involving abnormal collection of inflammatory cells forming nodules, called granulomas. Such granulomas occur in the lung and the mediastinal lymph nodes, in the heart, and in other vital and nonvital organs. The origin of the disease is unknown, and there are only limited clinical data on lung tissue of patients. No current model of sarcoidosis exists. In this paper we develop a mathematical model on the dynamics of the disease in the lung and use patients' lung tissue data to validate the model. The model is used to explore potential treatments.
| S-EPMC4234561 | biostudies-literature
Project description:Sarcoidosis is a multisystem granulomatous disease with nonspecific clinical manifestations that commonly affects the pulmonary system and other organs including the eyes, skin, liver, spleen, and lymph nodes. Sarcoidosis usually presents with persistent dry cough, eye and skin manifestations, weight loss, fatigue, night sweats, and erythema nodosum. Sarcoidosis is not influenced by sex or age, although it is more common in adults (< 50 years) of African-American or Scandinavians decent. Diagnosis can be difficult because of nonspecific symptoms and can only be verified following histopathological examination. Various factors, including infection, genetic predisposition, and environmental factors, are involved in the pathology of sarcoidosis. Exposures to insecticides, herbicides, bioaerosols, and agricultural employment are also associated with an increased risk for sarcoidosis. Due to its unknown etiology, early diagnosis and detection are difficult; however, the advent of advanced technologies, such as endobronchial ultrasound-guided biopsy, high-resolution computed tomography, magnetic resonance imaging, and 18F-fluorodeoxyglucose positron emission tomography has improved our ability to reliably diagnose this condition and accurately forecast its prognosis. This review discusses the causes and clinical features of sarcoidosis, and the improvements made in its prognosis, therapeutic management, and the recent discovery of potential biomarkers associated with the diagnostic assay used for sarcoidosis confirmation.
| S-EPMC7230978 | biostudies-literature
Project description:Paragonimiasis is a foodborne trematode infection of the lungs caused by Paragonimus spp., presenting clinically with similar symptoms to active tuberculosis (TB). Worldwide, an estimated 20.7 million people are infected with paragonimiasis, but relatively little epidemiological data exists for Africa. Given a recently reported case, we sought to establish whether paragonimiasis should be considered as an important differential diagnosis for human TB in The Gambia, West Africa.We developed a novel PCR-based diagnostic test for Paragonimus species known to be found in West Africa, which we used to examine archived TB negative sputum samples from a cross-sectional study of volunteers with tuberculosis-like symptoms from communities in the Western coastal region of The Gambia. Based on a "zero patient" design for detection of rare diseases, 300 anonymised AFB smear negative sputum samples, randomly selected from 25 villages, were screened for active paragonimiasis by molecular detection of Paragonimus spp. DNA. No parasite DNA was found in any of the sputa of our patient group. Despite the recent case report, we found no evidence of active paragonimiasis infection masking as TB in the Western region of The Gambia.
| S-EPMC5769439 | biostudies-literature
Project description:Two types of antiviral treatments, namely, interferon and nucleoside/nucleotide analogues are available for hepatitis infections. The selection of drug and dose determined using known pharmacokinetics and pharmacodynamics data is important. The lack of sufficient information for pharmacokinetics of a drug may not produce the desired results. Artificial neural network (ANN) provides a novel model-independent approach to pharmacokinetics and pharmacodynamics data. ANN model is created by supervised learning of 90 patients sample to predict the treatment strategy (lamivudine only and Lamivudine + Interferon) on the basis of viral load, liver function test, visit number, treatment duration, ethnic area, sex, and age. The model was trained with 68 (77.3%) samples and tested with 20 (22.7%) samples. The model produced 92% accuracy with 92.8% sensitivity and 83.3% specificity.
| S-EPMC3124792 | biostudies-literature