A modified artificial immune system based pattern recognition approach--an application to clinical diagnostics

Objective

This paper introduces a modified artificial immune system (AIS)-based pattern recognition method to enhance the recognition ability of the existing conventional AIS-based classification approach and demonstrates the superiority of the proposed new AIS-based method via two case studies of breast cancer diagnosis.

Methods and materials

Conventionally, the AIS approach is often coupled with the k nearest neighbor (k-NN) algorithm to form a classification method called AIS-kNN. In this paper we discuss the basic principle and possible problems of this conventional approach, and propose a new approach where AIS is integrated with the radial basis function - partial least square regression (AIS-RBFPLS). Additionally, both the two AIS-based approaches are compared with two classical and powerful machine learning methods, back-propagation neural network (BPNN) and orthogonal radial basis function network (Ortho-RBF network).

Results

The diagnosis results show that: (1) both the AIS-kNN and the AIS-RBFPLS proved to be a good machine leaning method for clinical diagnosis, but the proposed AIS-RBFPLS generated an even lower misclassification ratio, especially in the cases where the conventional AIS-kNN approach generated poor classification results because of possible improper AIS parameters. For example, based upon the AIS memory cells of “replacement threshold = 0.3”, the average misclassification ratios of two approaches for study 1 are 3.36% (AIS-RBFPLS) and 9.07% (AIS-kNN), and the misclassification ratios for study 2 are 19.18% (AIS-RBFPLS) and 28.36% (AIS-kNN); (2) the proposed AIS-RBFPLS presented its robustness in terms of the AIS-created memory cells, showing a smaller standard deviation of the results from the multiple trials than AIS-kNN. For example, using the result from the first set of AIS memory cells as an example, the standard deviations of the misclassification ratios for study 1 are 0.45% (AIS-RBFPLS) and 8.71% (AIS-kNN) and those for study 2 are 0.49% (AIS-RBFPLS) and 6.61% (AIS-kNN); and (3) the proposed AIS-RBFPLS classification approaches also yielded better diagnosis results than two classical neural network approaches of BPNN and Ortho-RBF network.

Conclusion

In summary, this paper proposed a new machine learning method for complex systems by integrating the AIS system with RBFPLS. This new method demonstrates its satisfactory effect on classification accuracy for clinical diagnosis, and also indicates its wide potential applications to other diagnosis and detection problems.     

Use of bactec 460 TB system in the diagnosis of tuberculosis

PurposeTo evaluate, the efficacy of BACTEC 460 TB system for the diagnosis of tuberculosis in a tertiary care hospital in Mumbai, India. Methods: We compared 12,726 clinical specimens using BACTEC 460 TB system and conventional method for detection of Mycobacterium tuberculosis over a period of six years. Result: The overall recovery rate was 39% by BACTEC technique and 29% using Lowenstein-Jensen (LJ) medium. An average detection time for BACTEC 460 TB system was found to be 13.3 days and 15.3 days as against 31.2 days and 35.3 days by LJ method for respiratory and nonrespiratory specimens respectively. The average reporting time for drug susceptibility results ranged from 6-10 days for the BACTEC 460 TB system. Conclusions: The BACTEC system is a good system for level II laboratories, especially in the diagnosis of extrapulmonary and smear negative tuberculosis.     

Real-time recognition of patient intentions from sequences of pressure maps using artificial neural networks

Objective: In this paper we address the problem of recognising the movement intentions of patients restricted to a medical bed. The developed recognition system will be used to implement a natural human–machine interface to move a medical bed by means of the slight movements of patients with reduced mobility.Methods and material: Our proposal uses pressure map sequences as input and presents a novel system based on artificial neural networks to recognise the movement intentions. The system analyses each pressure map in real-time and classifies the raw information into output classes which represent these intentions. The complexity of the recognition problem is high because of the multiple body characteristics and distinct ways of communicating intentions. To address this problem, a complete processing chain was developed consisting of image processing algorithms, a knowledge extraction process, and a multilayer perceptron (MLP) classification model.Results: Different configurations of the MLP have been investigated and quantitatively compared. The accuracy of our approach is high, obtaining an accuracy of 87%. The model was compared with five well-known classification paradigms. The performance of a reduced model, obtained by through feature selection algorithms, was found to be better and less time-consuming than the original model. The whole proposal has been validated with real patients in pre-clinical tests using the final medical bed prototype.Conclusions: The proposed approach produced very promising results, outperforming existing classification approaches. The excellent behaviour of the recognition system will enable its use in controlling the movements of the bed, in several degrees of freedom, by the patient with his/her own body.     

FOXO3 rs12212067: T > G Association with Active Tuberculosis in Han Chinese Population

It is well known that the human innate immune and adaptive immune response play important role in tuberculosis (TB) infection and progress. Emerging evidence shows that FOXO3 plays an important role in the human immune system. Recent research has shown that the FOXO3 genetic variants are associated malaria infection. In this study, 268 confirmed TB patients, 321 patients with latent tuberculosis infection (LTBI), and 475 TB-free controls were recruited; the single-nucleotide polymorphism (SNP) rs12212067: T > G in FOXO3 was genotyped using predesigned TaqMan® allelic discrimination assays. The results showed that the G allele of rs12212067 in FOXO3 was more common in health control and the latent TB group compared with the active TB group (p?=?0.048, odds ratio (OR) 95 % confidence intervals (CI)?=?1.37 (1.00–1.89); p?=?0.042, OR 95 % CI?=?1.42 (1.01–1.99), respectively); furthermore, within active TB patients, the G allele of rs12212067 in FOXO3 was more frequent in extra-pulmonary tuberculosis (EPTB) group compared to pulmonary tuberculosis (PTB) group (p?=?0.035, OR 95 % CI?=?0.57 (0.33–0.97). In conclusion, this study found that rs12212067 in FOXO3 was associated with increased risk of active TB. The minor G allele might be a protection factor which was found more common in latent TB patients and healthy controls than active TB patients.     

A rapid and low-cost microscopic observation drug susceptibility assay for detecting TB and MDR-TB among individuals infected by HIV in South India

Background: The converging epidemics of HIV and tuberculosis (TB) pose one of the greatest public health challenges of our time. Rapid diagnosis of TB is essential in view of its infectious nature, high burden of cases, and emergence of drug resistance. Objective: The purpose of this present study was to evaluate the feasibility of implementing the microscopic observation drug susceptibility (MODS) assay, a novel assay for the diagnosis of TB and multi-drug-resistant tuberculosis (MDR-TB) directly from sputum specimens, in the Indian setting. Materials and Methods: This study involved a cross-sectional, blinded assessment of the MODS assay on 1036 suspected cases of pulmonary TB in HIV-positive and HIV-negative patients against the radiometric method, BD-BACTEC TB 460 system. Results: Overall, the sensitivity, specificity, positive predictive value, and negative predictive value of the MODS assay in detecting MTB among TB suspected patients were 89.1%, 99.1%, 94.2%, 95.8%, respectively. In addition, in the diagnosis of drug-resistant TB, the MODS assay was 84.2% sensitive for those specimens reporting MDR, 87% sensitivity for those specimens reporting INH mono-resistance, and 100% sensitive for specimens reporting RIF mono-resistance. The median time to detection of TB in the MODS assay versus BACTEC was 9 versus 21 days (P < 0.001). Conclusion: Costing 5 to 10 times lesser than the automated culture methods, the MODS assay has the potential clinical utility as a simple and rapid method. It could be effectively used as an alternative method for diagnosing TB and detection of MDR-TB in a timely and affordable way in resource-limited settings.     

Specific features of immune complexes in patients with sarcoidosis and pulmonary tuberculosis

Clinical and radiological features of tuberculosis and sarcoidosis are quite overlapping, and therefore, a diagnostic dilemma often persists. There are no commonly accepted criteria for the diagnosis of sarcoidosis due to the lack of data on the etiology of the disease. The exclusion of tuberculosis in every patient with suspected sarcoidosis is a mandatory stage of diagnosis, especially in countries with a high burden of tuberculosis. A prospective study was conducted with two groups of patients: group I (n?=?50)—patients with pulmonary sarcoidosis established according to standard criteria; group II (n?=?28)—patients with pulmonary tuberculosis with bacterial excretion. The control group (n?=?24) was presented by healthy subjects. The examination complex included x-ray, bacteriological, immunological (Mantoux test with 2 TE, TB.SPOT test), and histological methods. All patients and healthy subjects were assessed for immune complexes with the use of the dynamic light scattering (DLS) method and adding of “healthy lung tissue extract” antigens and specific tuberculosis antigens ESAT-6 and SFP-10 in vitro. Significant differences were found in determining specific immune complexes in patients with pulmonary sarcoidosis and pulmonary tuberculosis. Registration of specific immune complex formation with “healthy lung tissue extract” in 100% cases may indicate the autoimmune nature of sarcoidosis. The absence of the immune complex formation in response to ESAT-6/SFP-10 antigens can be used for the differential diagnosis of two diseases. The diagnostic significance of the DLS method was 100% for sarcoidosis and 92.2% for tuberculosis. The data obtained in the study allows not only understanding the etiology of sarcoidosis, but also obtaining new criteria for the differential diagnosis of tuberculosis and pulmonary sarcoidosis.     

Therapeutic host-directed strategies to improve outcome in tuberculosis

Bacille Calmette-Guérin (BCG) is the only licenced tuberculosis (TB) vaccine, but has limited efficacy against pulmonary TB disease development and modest protection against extrapulmonary TB. Preventative antibiotic treatment for Mycobacterium tuberculosis (Mtb) infections in high-prevalence settings is unfeasible due to unclear treatment durability, drug toxicity, logistical constraints related to directly observed treatment strategy (DOTS) and the lengthy treatment protocols. Together, these factors promote non-adherence, contributing to relapse and establishment of drug-resistant Mtb strains. Although antibiotic treatment of drug-susceptible Mtb is generally effective, drug-resistant TB has a treatment efficacy below 50% and can, in a proportion, develop into progressive, untreatable disease. Other immune compromising co-infections and/or co-morbidities require more complex prevention/treatment approaches, posing huge financial burdens to national health services. Novel TB treatment strategies, such as host-directed therapeutics, are required to complement pathogen-targeted approaches. Pre-clinical studies have highlighted promising candidates that enhance endogenous pathways and/or limit destructive host responses. This review discusses promising pre-clinical candidates and forerunning compounds at advanced stages of clinical investigation in TB host-directed therapeutic (HDT) efficacy trials. Such approaches are rationalized to improve outcome in TB and shorten treatment strategies.     

Potential Role of Immunodiagnosis for Pulmonary Tuberculosis Using Induced Sputum Cells

PurposeTo evaluate the diagnostic utility and predictors for determinate results of an enzyme-linked immunospot assay using induced sputum cells (IS ELISPOT) for a rapid diagnosis of pulmonary tuberculosis (TB).ResultsA total of 43 subjects, including 25 with TB (TB group) and 18 with non-TB disease (non-TB group) were enrolled. Results of IS ELISPOT were determinate in only 17/43 (39%) subjects, but all of determinate results were consistent with the final diagnosis. Of the 43 sputum samples, 11 (26%) were inadequate to perform IS ELISPOT. Of 32 adequate sputum samples, the proportion of determinate results was significantly higher in the TB group (75%, 15/20) than in the non-TB group (17%, 2/12) (p=0.002). The status of active TB was a unique predictor but smear positivity was not a significant predictor for determinate results. In addition, sensitivity of IS ELISPOT (75%, 9/12) in smear negative TB was higher than that of TB-polymerase chain reaction (25%, 3/12).ConclusionIS ELISPOT showed relatively high diagnostic value and accuracy in the TB group, independent of smear positivity. IS ELISPOT may provide additional diagnostic yield for microbiological tools in the rapid diagnosis of smear-negative TB.     

Application of adaptive neuro-fuzzy inference system for epileptic seizure detection using wavelet feature extraction

Intelligent computing tools such as artificial neural network (ANN) and fuzzy logic approaches are demonstrated to be competent when applied individually to a variety of problems. Recently, there has been a growing interest in combining both these approaches, and as a result, neuro-fuzzy computing techniques have been evolved. In this study, a new approach based on an adaptive neuro-fuzzy inference system (ANFIS) was presented for epileptic seizure detection. The proposed ANFIS model combined the neural network adaptive capabilities and the fuzzy logic qualitative approach. Decision making was performed in two stages: feature extraction using the wavelet transform (WT) and the ANFIS trained with the backpropagation gradient descent method in combination with the least squares method. Some conclusions concerning the impacts of features on the detection of epileptic seizures were obtained through analysis of the ANFIS. The results are highly promising, and a comparative analysis suggests that the proposed modeling approach outperforms ANN model in terms of training performances and classification accuracies. The results confirmed that the proposed ANFIS model has some potential in epileptic seizure detection. The ANFIS model achieved accuracy rates which were higher than that of the stand-alone neural network model.     

Evaluation of Five User-Friendly Whole Genome Sequencing Software for Mycobacterium tuberculosis in Clinical Application

BackgroundWhole genome sequencing (WGS) is an increasingly useful tool for tuberculosis (TB) diagnosis and disease management. In this study, we evaluated the utility of user-friendly WGS tools in reporting resistance profiles and identifying lineages of clinical TB isolates from South Korea.MethodsForty clinical samples from TB patients showing discrepancies between their rapid molecular and conventional drug susceptibility tests were used in this study. Among these clinical isolates, 37 strains were successfully evaluated via WGS software, using the GenTB, TB Profiler, PhyResSE, CASTB, and Mykrobe.ResultsMore accurate and faster susceptibility results could be obtained with isoniazid than with rifampin. Using the phenotypic test as the gold standard, the isoniazid concordance rate between phenotypic drug susceptibility test (DST) and WGS (GenTB: 45.9%, TB profiler: 40.5%, PhyResSE: 40.5%, CASTB: 48.6%, and Mykrobe: 43.2%) was much higher than between phenotypic DST and rapid molecular genotypic DST (18.9%) among the 37 strains. In contrast, the rifampin concordance rate between phenotypic DST and WGS and that between phenotypic DST and rapid molecular genotypic DST was similar (81.1–89.2%). We also found novel mutations associated with INH in katG and ahpC gene region, not covered by the line probe assay. In addition, lineage analysis identified 81.1% of these samples as L2 East Asian lineage strains, and 18.9% as L4 Euro-American lineage strains.ConclusionWGS may play a pivotal role in TB diagnosis and the detection of drug resistance, genetic diversity, and transmission dynamics in the near future because of its accuracy, speed, and extensibility.     

An Artificial Immune System-Based Support Vector Machine Approach for Classifying Ultrasound Breast Tumor Images

A rapid and highly accurate diagnostic tool for distinguishing benign tumors from malignant ones is required owing to the high incidence of breast cancer. Although various computer-aided diagnosis (CAD) systems have been developed to interpret ultrasound images of breast tumors, feature selection and the setting of parameters are still essential to classification accuracy and the minimization of computational complexity. This work develops a highly accurate CAD system that is based on a support vector machine (SVM) and the artificial immune system (AIS) algorithm for evaluating breast tumors. Experiments demonstrate that the accuracy of the proposed CAD system for classifying breast tumors is 96.67 %. The sensitivity, specificity, PPV, and NPV of the proposed CAD system are 96.67, 96.67, 95.60, and 97.48 %, respectively. The receiver operator characteristic (ROC) area index A_z is 0.9827. Hence, the proposed CAD system can reduce the number of biopsies and yield useful results that assist physicians in diagnosing breast tumors.     

Electrocardiogram analysis using a combination of statistical, geometric, and nonlinear heart rate variability features

Objective

The paper addresses a common and recurring problem of electrocardiogram (ECG) classification based on heart rate variability (HRV) analysis. Current understanding of the limits of HRV analysis in diagnosing different cardiac conditions is not complete. Existing research suggests that a combination of carefully selected linear and nonlinear HRV features should significantly improve the accuracy for both binary and multiclass classification problems. The primary goal of this work is to evaluate a proposed combination of HRV features. Other explored objectives are the comparison of different machine learning algorithms in the HRV analysis and the inspection of the most suitable period T between two consecutively analyzed R-R intervals for nonlinear features.

Methods and material

We extracted 11 features from 5 min of R-R interval recordings: SDNN, RMSSD, pNN20, HRV triangular index (HTI), spatial filling index (SFI), correlation dimension, central tendency measure (CTM), and four approximate entropy features (ApEn1-ApEn4). Analyzed heart conditions included normal heart rhythm, arrhythmia (any), supraventricular arrhythmia, and congestive heart failure. One hundred patient records from six online databases were analyzed, 25 for each condition. Feature vectors were extracted by a platform designed for this purpose, named ECG Chaos Extractor. The vectors were then analyzed by seven clustering and classification algorithms in the Weka system: K-means, expectation maximization (EM), C4.5 decision tree, Bayesian network, artificial neural network (ANN), support vector machines (SVM) and random forest (RF). Four-class and two-class (normal vs. abnormal) classification was performed. Relevance of particular features was evaluated using 1-Rule and C4.5 decision tree in the cases of individual features classification and classification with features’ pairs.

Results

Average total classification accuracy obtained for top three classification methods in the two classes’ case was: RF 99.7%, ANN 99.1%, SVM 98.9%. In the four classes’ case the best results were: RF 99.6%, Bayesian network 99.4%, SVM 98.4%. The best overall method was RF. C4.5 decision tree was successful in the construction of useful classification rules for the two classes’ case. EM and K-means showed comparable clustering results: around 50% for the four classes’ case and around 75% for the two classes’ case. HTI, pNN20, RMSSD, ApEn3, ApEn4 and SFI were shown to be the most relevant features. HTI in particular appears in most of the top-ranked pairs of features and is the best analyzed feature. The choice of the period T for nonlinear features was shown to be arbitrary. However, a combination of five different periods significantly improved classification accuracy, from 70% for a single period up to 99% for five periods.

Conclusions

Analysis shows that the proposed combination of 11 linear and nonlinear HRV features gives high classification accuracy when nonlinear features are extracted for five periods. The features’ combination was thoroughly analyzed using several machine learning algorithms. In particular, RF algorithm proved to be highly efficient and accurate in both binary and multiclass classification of HRV records. Interpretable and useful rules were obtained with C4.5 decision tree. Further work in this area should elucidate which features should be extracted for the best classification results for specific types of cardiac disorders.     

Prévalence de la co-infection VIH-tuberculose et impact de l’infection VIH sur l’évolution de la tuberculose pulmonaire au Togo

The aim of this study was to determine the prevalence of HIV infection in tuberculosis patients and its impact on the TB treatment. We enrolled 569 pulmonary TB patients in four diagnosis and treatment centres in Togo. All patients were new TB cases and received the first-line TB drugs: two months of rifampicin-pyrazinamide-isoniazid-ethambutol and six months of isoniazid-ethambutol. HIV testing was done according to the national guidelines, using rapid diagnosis tests. The CD4 lymphocyte counting was performed by Facscalibur (BD, Sciences) for all HIV-positive patients. Of the 569 TB patients enrolled, 135 (23.7%) were HIV positive (TB/HIV+). HIV prevalence was 22.4% (76 of 339) among men and 25.6% (59 of 230) among women without statistical difference. The global rate of treatment success was 82.2%. The rate of treatment success was lower (64.3%) in TB/HIV+ patients than in TB/HIV? patients (87.5%) (p <0.01). The mortality rates were 25.6% and 11.8% in TB/HIV+ patients and TB/HIV? patients, respectively, with a statistically significant difference (p <0.01). We did not found any statistical difference between the rates of treatment success among TB/HIV? (87.5%) patients and TB/HIV+ patients who had TCD4 lymphocyte counts above 200/µl (84.4%). TB program in Togo must take into account HIV infection to improve its performance.     

Identification, Molecular Cloning, and Evaluation of Potential Use of Isocitrate Dehydrogenase II of Mycobacterium bovis BCG in Serodiagnosis of Tuberculosis

Diagnosis of tuberculosis is time-consuming and requires infrastructures which are often not available in countries with high incidences of the disease. In the present study, an 82-kDa protein antigen was isolated by affinity chromatography and was identified by peptide mass fingerprinting as isocitrate dehydrogenase II, which is encoded by the icd2 gene of Mycobacterium bovis BCG. The icd2 gene of BCG was cloned by PCR, and the product of recombinant gene expression was purified and analyzed by two-dimensional polyacrylamide gel electrophoresis. The recombinant protein, named rICD2, was tested for its recognition by immunoglobulin G (IgG) antibodies from the sera of 16 patients with tuberculosis (TB) and 23 healthy individuals by Western blotting. The results showed that rICD2 is recognized by IgG antibodies from the sera of all TB patients tested at serum dilutions of ≥1:640. At a serum dilution of 1:1,280, the sensitivity was 50% and the specificity was 86.9%. These results indicate that rICD2 might represent a candidate for use in a new assay for the serodiagnosis of TB.     

Diagnosis of acute abdominal pain using a three-stage classifier

The present paper deals with an application of a three-stage classifier based on a decision tree logic to the diagnosis of acute abdominal pain. On the basis of clinical information collected from a series of 476 patients suffering from abdominal pain of acute onset, the method of multistage classifier synthesis is presented. The results of classification accuracy using a modified version of k-nearest neighbours strategy for different features used at interior nodes of a tree are given.     

Diagnostic potential of IS6110, 38kDa, 65kDa and 85B sequence-based polymerase chain reaction in the diagnosis of Mycobacterium tuberculosis in clinical samples

Purpose: The correlation between the presence of specific gene sequence of M. tuberculosis and specific diagnosis of clinical tuberculosis is not known. This study compared the results of polymerase chain reaction (PCR) amplification of M. tuberculosis specific DNA sequences (IS6110, 65kDa, 38kDa and mRNA coding for 85 B protein) from different clinical samples of pulmonary and extrapulmonary tuberculosis. Methods: One hundred and seventy-two clinical samples from suspected tuberculosis patients were tested for smear examination, culture (LJ and rapid BACTEC 460 TB system) and PCR. PCR was performed with specific primers for the targets: IS6110, 65kDa, 38kDa and 85B. Results: Each PCR test was found to have a much higher positivity than conventional test and BACTEC culture (P 0.05). Smear positive samples (56) and the samples (36) showing positive results by conventional methods (smear and LJ medium culture) and BACTEC were found to be positive by all PCR protocols. No significant difference was found between the four PCR protocols (P >0.05). The primer specific for amplifying the 123bp IS6110 fragment gave the highest positivity (83%), followed by 65kDa, 38kDa and 85B RT-PCR in descending order. Conclusions: These data suggest that the presence of IS6110 correlates more closely with the diagnosis of clinical tuberculosis than that of 65kDa, 38kDa and 85B     

Auxiliary Diagnosis for COVID-19 with Deep Transfer Learning

To assist physicians identify COVID-19 and its manifestations through the automatic COVID-19 recognition and classification in chest CT images with deep transfer learning. In this retrospective study, the used chest CT image dataset covered 422 subjects, including 72 confirmed COVID-19 subjects (260 studies, 30,171 images), 252 other pneumonia subjects (252 studies, 26,534 images) that contained 158 viral pneumonia subjects and 94 pulmonary tuberculosis subjects, and 98 normal subjects (98 studies, 29,838 images). In the experiment, subjects were split into training (70%), validation (15%) and testing (15%) sets. We utilized the convolutional blocks of ResNets pretrained on the public social image collections and modified the top fully connected layer to suit our task (the COVID-19 recognition). In addition, we tested the proposed method on a finegrained classification task; that is, the images of COVID-19 were further split into 3 main manifestations (ground-glass opacity with 12,924 images, consolidation with 7418 images and fibrotic streaks with 7338 images). Similarly, the data partitioning strategy of 70%-15%-15% was adopted. The best performance obtained by the pretrained ResNet50 model is 94.87% sensitivity, 88.46% specificity, 91.21% accuracy for COVID-19 versus all other groups, and an overall accuracy of 89.01% for the three-category classification in the testing set. Consistent performance was observed from the COVID-19 manifestation classification task on images basis, where the best overall accuracy of 94.08% and AUC of 0.993 were obtained by the pretrained ResNet18 (P < 0.05). All the proposed models have achieved much satisfying performance and were thus very promising in both the practical application and statistics. Transfer learning is worth for exploring to be applied in recognition and classification of COVID-19 on CT images with limited training data. It not only achieved higher sensitivity (COVID-19 vs the rest) but also took far less time than radiologists, which is expected to give the auxiliary diagnosis and reduce the workload for the radiologists.     

Difference in TB10.4 T‐cell epitope recognition following immunization with recombinant TB10.4, BCG or infection with Mycobacterium tuberculosis

Most novel vaccines against infectious diseases are based on recombinant Ag; however, only few studies have compared Ag‐specific immune responses induced by natural infection with that induced by the same Ag in a recombinant form. Here, we studied the epitope recognition pattern of the tuberculosis vaccine Ag, TB10.4, in a recombinant form, or when expressed by the pathogen Mycobacterium tuberculosis (M.tb), or by the current anti‐tuberculosis vaccine, Mycobacterium bovis BCG. We showed that BCG and M.tb induced a similar CD4⁺ T‐cell specific TB10.4 epitope‐pattern, which differed completely from that induced by recombinant TB10.4. This difference was not due to post‐translational modifications of TB10.4 or because TB10.4 is secreted from BCG and M.tb as a complex with Rv0287. In addition, BCG and TB10.4/CAF01 were both taken up by DC and macrophages in vivo, and in vitro uptake experiments revealed that both TB10.4 and BCG were transported to Lamp⁺‐compartments. BCG and TB10.4 however, were directed to different types of Lamp⁺‐compartments in the same APC, which may lead to different epitope recognition patterns. In conclusion, we show that different vectors can induce completely different recognition of the same protein.     

Epitope-Specific Antibody Levels Demonstrate Recognition of New Epitopes and Changes in Titer but Not Affinity during Treatment of Tuberculosis

Antibody levels rise during treatment of tuberculosis. This study examined when this rise occurred, whether there was recognition of new antigen binding sites (epitopes) on the same or different antigens, and how long specific antibody persisted. Forty patients with smear-positive pulmonary tuberculosis provided serum before and during treatment. Antibody levels were measured using a monoclonal antibody competition assay to epitopes restricted to the Mycobacterium tuberculosis complex and an enzyme-linked immunosorbent assay for lipoarabinomannan. Significant increases in antibody levels were apparent after 7 days of treatment. Five samples (12.5%) had positive titers to all epitopes at the start of treatment, and this increased to 23 (58%) during treatment. Antibody to epitopes with the poorest sensitivity (the TB23 epitope of the 19-kDa antigen and the TB78 epitope of hsp65) showed the greatest increases after treatment. Antibody to these two epitopes was also absent in some patients with relapsed tuberculosis until after treatment. Antibody titers showed a biphasic response, with a fall at 2 to 3 months of treatment. Sera from two patients showed changes in the affinity of epitope-specific antibody during treatment, whereas the majority did not. Those infected with isoniazid-resistant strains of M. tuberculosis showed a late rise in antibody. Antibody to the TB68 epitope of the 16-kDa α-crystallin homolog was short-lived, but it recurred with bacteriological relapse during treatment. Positive antibody titers persisted for at least 3 to 18 months after treatment. Diagnostic tests for tuberculosis should be evaluated using only pretreatment sera. Delayed antigenic recognition could be due to active suppression and/or failure to engage internal antigens of M. tuberculosis.     

Medical decision support system for diagnosis of neuromuscular disorders using DWT and fuzzy support vector machines

The motor unit action potentials (MUAPs) in an electromyographic (EMG) signal provide a significant source of information for the assessment of neuromuscular disorders. In this work, different types of machine learning methods were used to classify EMG signals and compared in relation to their accuracy in classification of EMG signals. The models automatically classify the EMG signals into normal, neurogenic or myopathic. The best averaged performance over 10 runs of randomized cross-validation is also obtained by different classification models. Some conclusions concerning the impacts of features on the EMG signal classification were obtained through analysis of the classification techniques. The comparative analysis suggests that the fuzzy support vector machines (FSVM) modelling is superior to the other machine learning methods in at least three points: slightly higher recognition rate; insensitivity to overtraining; and consistent outputs demonstrating higher reliability. The combined model with discrete wavelet transform (DWT) and FSVM achieves the better performance for internal cross validation (External cross validation) with the area under the receiver operating characteristic (ROC) curve (AUC) and accuracy equal to 0.996 (0.970) and 97.67% (93.5%), respectively. These results show that the proposed model have the potential to obtain a reliable classification of EMG signals, and to assist the clinicians for making a correct diagnosis of neuromuscular disorders.