Machine Learning Algorithms Evaluate Immune Response to Novel Mycobacterium tuberculosis Antigens for Diagnosis of Tuberculosis

Evaluation of serological assays for the diagnosis of childhood tuberculosis disease: a study protocol

BACKGROUND

Tuberculosis (TB) poses a major public health challenge, particularly in children. A substantial proportion of children with TB disease remain undetected and unconfirmed. Therefore, there is an urgent need for a highly sensitive point-of-care test. This study aims to assess the performance of serological assays based on various antigen targets and antibody properties in distinguishing children (0-18 years) with TB disease (1) from healthy TB-exposed children, (2) children with non-TB lower respiratory tract infections, and (3) from children with TB infection.

METHODS

The study will use biobanked plasma samples collected from three prospective multicentric diagnostic observational studies: the Childhood TB in Switzerland (CITRUS) study, the Pediatric TB Research Network in Spain (pTBred), and the Procalcitonin guidance to reduce antibiotic treatment of lower respiratory tract infections in children and adolescents (ProPAED) study. Included are children diagnosed with TB disease or infection, healthy TB-exposed children, and sick children with non-TB lower respiratory tract infection. Serological multiplex assays will be performed to identify M. tuberculosis antigen-specific antibody features, including isotypes, subclasses, Fc receptor (FcR) binding, and IgG glycosylation.

DISCUSSION

The findings from this study will help to design serological assays for diagnosing TB disease in children. Importantly, those assays could easily be developed as low-cost point-of-care tests, thereby offering a potential solution for resource-constrained settings.

CLINICALTRIALS

GOV IDENTIFIER

NCT03044509.

The performance of VCS(volume, conductivity, light scatter) parameters in distinguishing latent tuberculosis and active tuberculosis by using machine learning algorithm

BACKGROUND

Tuberculosis is a chronic infectious disease caused by mycobacterium tuberculosis (MTB) and is the ninth leading cause of death worldwide. It is still difficult to distinguish active TB from latent TB,but it is very important for individualized management and treatment to distinguish whether patients are active or latent tuberculosis infection.

METHODS

A total of 220 subjects, including active TB patients (ATB, n = 97) and latent TB patients (LTB, n = 113), were recruited in this study .46 features about blood routine indicators and the VCS parameters (volume, conductivity, light scatter) of neutrophils(NE), monocytes(MO), and lymphocytes(LY) were collected and was constructed classification model by four machine learning algorithms(logistic regression(LR), random forest(RF), support vector machine(SVM) and k-nearest neighbor(KNN)). And the area under the precision-recall curve (AUPRC) and the area under the receiver operating characteristic curve (AUROC) to estimate of the model's predictive performance for dentifying active and latent tuberculosis infection.

RESULTS

After verification,among the four classifications, LR and RF had the best performance (AUROC = 1, AUPRC = 1), followed by SVM (AUROC = 0.967, AUPRC = 0.971), KNN (AUROC = 0.943, AUPRC = 0.959) in the training set. And LR had the best performance (AUROC = 0.977, AUPRC = 0.957), followed by SVM (AUROC = 0.962, AUPRC = 0.949), RF (AUROC = 0.903, AUPRC = 0.922),KNN(AUROC = 0.883, AUPRC = 0.901) in the testing set.

CONCLUSIONS

The machine learning algorithm classifier based on leukocyte VCS parameters is of great value in identifying active and latent tuberculosis infection.

Machine learning to predict bacteriologic confirmation of Mycobacterium tuberculosis in infants and very young children

Diagnosis of tuberculosis (TB) among young children (<5 years) is challenging due to the paucibacillary nature of clinical disease and clinical similarities to other childhood diseases. We used machine learning to develop accurate prediction models of microbial confirmation with simply defined and easily obtainable clinical, demographic, and radiologic factors. We evaluated eleven supervised machine learning models (using stepwise regression, regularized regression, decision tree, and support vector machine approaches) to predict microbial confirmation in young children (<5 years) using samples from invasive (reference-standard) or noninvasive procedure. Models were trained and tested using data from a large prospective cohort of young children with symptoms suggestive of TB in Kenya. Model performance was evaluated using areas under the receiver operating curve (AUROC) and precision-recall curve (AUPRC), accuracy metrics. (i.e., sensitivity, specificity), F-beta scores, Cohen's Kappa, and Matthew's Correlation Coefficient. Among 262 included children, 29 (11%) were microbially confirmed using any sampling technique. Models were accurate at predicting microbial confirmation in samples obtained from invasive procedures (AUROC range: 0.84-0.90) and from noninvasive procedures (AUROC range: 0.83-0.89). History of household contact with a confirmed case of TB, immunological evidence of TB infection, and a chest x-ray consistent with TB disease were consistently influential across models. Our results suggest machine learning can accurately predict microbial confirmation of M. tuberculosis in young children using simply defined features and increase the bacteriologic yield in diagnostic cohorts. These findings may facilitate clinical decision making and guide clinical research into novel biomarkers of TB disease in young children.

Monocyte, Lymphocyte and Neutrophil Ratios - Easy-to-Use Biomarkers for the Diagnosis of Pediatric Tuberculosis

BACKGROUND

The neutrophil-to-lymphocyte-ratio (NLR), neutrophil-to-monocyte-plus-lymphocyte-ratio (NMLR) and monocyte-to-lymphocyte-ratio (MLR) may have diagnostic potential for tuberculosis (TB).

METHODS

Data of two prospective multicenter studies in Switzerland were used, which included children <18 years with TB exposure, infection or disease or with febrile non-TB lower-respiratory-tract infection (nTB-LRTI).

RESULTS

Of the 389 children included 25 (6.4%) had TB disease, 12 (3.1%) TB infection, 28 (7.2%) were healthy TB exposed and 324 (83.3%) nTB-LRTI. Median (IQR) NLR was highest with 2.0 (1.2, 2.2) in children with TB disease compared to TB exposed [0.8 (0.6, 1.3); P = 0.002] and nTB-LRTI [0.3 (0.1, 1.0); P < 0.001]. Median (IQR) NMLR was highest with 1.4 (1.2, 1.7) in children with TB disease compared to healthy exposed [0.7 (0.6, 1.1); P = 0.003] and children with nTB-LRTI [0.2 (0.1, 0.6); P < 0.001). Receiver operating characteristic curves to detect TB disease compared to nTB-LRTI for NLR and NMLR had an area under the curve of 0.82 and 0.86, the sensitivity of 88% and 88%, and specificity of 71% and 76%, respectively.

CONCLUSION

NLR and NMLR are promising, easy-to-obtain diagnostic biomarkers to differentiate children with TB disease from other lower respiratory tract infections. These results require validation in a larger study and in settings with high and low TB endemicity.

An Immunoinformatic Strategy to Develop New Mycobacterium tuberculosis Multi-epitope Vaccine

Mycobacterium tuberculosis causes a life-threatening disease known as tuberculosis (TB). In 2021, tuberculosis was the second cause of death after COVID-19 among infectious diseases. Latent life cycle and development of multidrug resistance in one hand and lack of an effective vaccine in another hand have made TB a global health issue. Here, a multi-epitope vaccine have been designed against TB using five new antigenic protein and immunoinformatic tools. To do so, immunodominant MHC-I/MHC-II binding epitopes of Rv2346, Rv2347, Rv3614, Rv3615 and Rv2031 antigenic proteins have been selected using advanced computational procedures. The vaccine was designed by linking ten epitopes from the antigenic proteins and flagellin and TpD as adjuvant. Three-dimensional (3D) structure of the vaccine was modeled, was refined and was evaluated using bioinformatics tools. The 3D structure of the vaccine was docked into the toll-like-receptors (TLR3, 4, 8) to evaluate potential interaction between the vaccine and TLRs. Evaluation of immunological and physicochemical properties of the constructed vaccine have demonstrated the vaccine construct can induce significant humoral and cellular immune responses, the vaccine is non-allergenic and can be recognized by TLR proteins. The immunoinformatic results reported in the present study demonstrates that it is worth following the designed vaccine by experimental investigations.

A Study of the Recent Trends of Immunology: Key Challenges, Domains, Applications, Datasets, and Future Directions

The human immune system is very complex. Understanding it traditionally required specialized knowledge and expertise along with years of study. However, in recent times, the introduction of technologies such as AIoMT (Artificial Intelligence of Medical Things), genetic intelligence algorithms, smart immunological methodologies, etc., has made this process easier. These technologies can observe relations and patterns that humans do and recognize patterns that are unobservable by humans. Furthermore, these technologies have also enabled us to understand better the different types of cells in the immune system, their structures, their importance, and their impact on our immunity, particularly in the case of debilitating diseases such as cancer. The undertaken study explores the AI methodologies currently in the field of immunology. The initial part of this study explains the integration of AI in healthcare and how it has changed the face of the medical industry. It also details the current applications of AI in the different healthcare domains and the key challenges faced when trying to integrate AI with healthcare, along with the recent developments and contributions in this field by other researchers. The core part of this study is focused on exploring the most common classifications of health diseases, immunology, and its key subdomains. The later part of the study presents a statistical analysis of the contributions in AI in the different domains of immunology and an in-depth review of the machine learning and deep learning methodologies and algorithms that can and have been applied in the field of immunology. We have also analyzed a list of machine learning and deep learning datasets about the different subdomains of immunology. Finally, in the end, the presented study discusses the future research directions in the field of AI in immunology and provides some possible solutions for the same.

HIV-Infected Patients Developing Tuberculosis Disease Show Early Changes in the Immune Response to Novel Mycobacterium tuberculosis Antigens

Background: In individuals living with HIV infection the development of tuberculosis (TB) is associated with rapid progression from asymptomatic TB infection to active TB disease. Sputum-based diagnostic tests for TB have low sensitivity in minimal and subclinical TB precluding early diagnosis. The immune response to novel Mycobacterium tuberculosis in-vivo expressed and latency associated antigens may help to measure the early stages of infection and disease progression and thereby improve early diagnosis of active TB disease. Methods: Serial prospectively sampled cryopreserved lymphocytes from patients of the Swiss HIV Cohort Study developing TB disease ("cases") and matched patients with no TB disease ("controls") were stimulated with 10 novel Mycobacterium tuberculosis antigens. Cytokine concentrations were measured in cases and controls at four time points prior to diagnosis of TB: T1-T4 with T4 being the closest time point to diagnosis. Results: 50 samples from nine cases and nine controls were included. Median CD4 cell count at T4 was 289/ul for the TB-group and 456/ul for the control group. Viral loads were suppressed in both groups. At T4 Rv2431c-induced and Rv3614/15c-induced interferon gamma-induced protein (IP)-10 responses and Rv2031c-induced and Rv2346/Rv2347c-induced tumor necrosis factor (TNF)-α responses were significantly higher in cases compared to controls (p < 0.004). At T3 - being up to 2 years prior to TB diagnosis - Rv2031c-induced TNF-α was significantly higher in cases compared to controls (p < 0.004). Area under the receiver operating characteristics (AUROC) curves resulted in an AUC > 0.92 for all four antigen-cytokine pairs. Conclusion: The in vitro Mycobacterium tuberculosis-specific immune response in HIV-infected individuals that progress toward developing TB disease is different from those in HIV-infected individuals that do not progress to developing TB. These differences precede the clinical diagnosis of active TB up to 2 years, paving the way for the development of immune based diagnostics to predict TB disease at an early stage.