A decision tree for differentiating tuberculous from malignant pleural effusions

Enhanced differential evolution algorithm for feature selection in tuberculous pleural effusion clinical characteristics analysis

Tuberculous pleural effusion poses a significant threat to human health due to its potential for severe disease and mortality. Without timely treatment, it may lead to fatal consequences. Therefore, early identification and prompt treatment are crucial for preventing problems such as chronic lung disease, respiratory failure, and death. This study proposes an enhanced differential evolution algorithm based on colony predation and dispersed foraging strategies. A series of experiments conducted on the IEEE CEC 2017 competition dataset validated the global optimization capability of the method. Additionally, a binary version of the algorithm is introduced to assess the algorithm's ability to address feature selection problems. Comprehensive comparisons of the effectiveness of the proposed algorithm with 8 similar algorithms were conducted using public datasets with feature sizes ranging from 10 to 10,000. Experimental results demonstrate that the proposed method is an effective feature selection approach. Furthermore, a predictive model for tuberculous pleural effusion is established by integrating the proposed algorithm with support vector machines. The performance of the proposed model is validated using clinical records collected from 140 tuberculous pleural effusion patients, totaling 10,780 instances. Experimental results indicate that the proposed model can identify key correlated indicators such as pleural effusion adenosine deaminase, temperature, white blood cell count, and pleural effusion color, aiding in the clinical feature analysis of tuberculous pleural effusion and providing early warning for its treatment and prediction.

Construction of a multi-classified decision tree model for identifying malignant pleural effusion and tuberculous pleural effusion

OBJECTIVE

Pleural effusion (PE) is a common clinical complication associated with various disorders. We aimed to utilize laboratory variables and their corresponding ratios in serum and PE for the differential diagnosis of multiple types of PE based on a decision tree (DT) algorithm.

METHODS

A total of 1435 untreated patients with PE admitted to The First Affiliated Hospital of Ningbo University were enrolled. The demographic and laboratory variables were collected and compared. The receiver operating characteristic curve was used to select important variables for diagnosing malignant pleural effusion (MPE) or tuberculous pleural effusion (TPE) and included in the DT model. The data were divided into the training set and the test set at a ratio of 7:3. The training data was used to develop the DT model, and the test data was for evaluating the model. Independent data was collected as external validation.

RESULTS

Three PE indicators (carcinoembryonic antigen, adenosine deaminase [ADA], and total protein), two serum indicators (neuron-specific enolase and cytokeratin 19 fragments), and two ratios [high-sensitivity C-reactive protein (hsCRP)/ PE lymphocyte and hsCRP/PE ADA] were used to construct the DT model. The area under the curve (AUC), sensitivity, and specificity for diagnosing MPE were 0.963, 84.0%, 91.6% in the training set, 0.976, 84.1%, 88.6% in the test set, and 0.955,83.3%, 86.7% in the external validation set. The AUC, sensitivity, and specificity of diagnosing TPE were 0.898, 86.8%, 92.3% in the training set, 0.888, 88.8%, 92.7% in the test set, and 0.778, 84.8%, 94.3% in the external validation set.

CONCLUSION

The DT model showed good diagnostic efficacy and could be applied for the differential diagnosis of MPE and TPE in clinical settings.

Expert Review on Contemporary Management of Common Benign Pleural Effusions

Heart failure (HF) and cirrhosis are frequently associated with pleural effusions (PEs). Despite their apparently benign nature, both HF-related effusions and hepatic hydrothorax (HH) have poor prognosis because they represent an advanced stage of the disease. Optimization of medical therapy in these two entities involve not only the use of diuretics, but also other pharmacological therapies. For instance, all HF patients with reduced or mildly reduced left ventricular ejection fraction can benefit from angiotensin receptor-neprilysin inhibitors, beta blockers, mineralocorticoid receptor antagonists, and sodium-glucose cotransporter 2 inhibitors. Conversely, it is better for HH patients to avoid nonselective beta blockers. Refractory cardiac- and cirrhosis-related PEs are commonly managed by iterative therapeutic thoracentesis. When repeated aspirations are needed, thereby diminishing quality of life, the insertion of an indwelling pleural catheter (IPC) may be warranted. However, in selected HH patients who are diuretic-resistant or diuretic-intractable, placement of transjugular intrahepatic portosystemic shunts should be considered as a bridge to liver transplantation, whereas in transplant candidates the role of IPC is debatable. Another benign condition, pleural tuberculosis (TB) is a serious health problem in developing countries. Diagnostic certainty is still a concern due to the paucibacillary nature of the infection, although the use of more sensitive nucleic acid amplification tests is becoming more widespread. Its treatment is the same as that of pulmonary TB, but the potential drug interactions between antiretroviral and anti-TB drugs in HIV-coinfected patients as well as the current recommended guidelines for the different types of anti-TB drugs resistance should be followed.

Tuberculous pleural effusion prediction using ant colony optimizer with grade-based search assisted support vector machine

Introduction

Although tuberculous pleural effusion (TBPE) is simply an inflammatory response of the pleura caused by tuberculosis infection, it can lead to pleural adhesions and cause sequelae of pleural thickening, which may severely affect the mobility of the chest cavity.

Methods

In this study, we propose bGACO-SVM, a model with good diagnostic power, for the adjunctive diagnosis of TBPE. The model is based on an enhanced continuous ant colony optimization (ACOR) with grade-based search technique (GACO) and support vector machine (SVM) for wrapped feature selection. In GACO, grade-based search greatly improves the convergence performance of the algorithm and the ability to avoid getting trapped in local optimization, which improves the classification capability of bGACO-SVM.

Results

To test the performance of GACO, this work conducts comparative experiments between GACO and nine basic algorithms and nine state-of-the-art variants as well. Although the proposed GACO does not offer much advantage in terms of time complexity, the experimental results strongly demonstrate the core advantages of GACO. The accuracy of bGACO-predictive SVM was evaluated using existing datasets from the UCI and TBPE datasets.

Discussion

In the TBPE dataset trial, 147 TBPE patients were evaluated using the created bGACO-SVM model, showing that the bGACO-SVM method is an effective technique for accurately predicting TBPE.

Echographic septation: A potentially useful indicator discriminating tuberculous from malignant pleural effusion

BACKGROUND

Tuberculous (TBE) and malignant (MPE) pleural effusions present with similar lymphocytic exudates. As TBE is an inflammatory and hypersensitivity process, we hypothesized that echographic septation may be more prevalent in TBE than in MPE, potentially serving as a good clinical predictor for TBE.

METHODS

A total of 183 TBE and 266 MPE patients were recruited retrospectively. Multivariate logistic regression was performed to determine significant predictors for TBE.

RESULTS

TBE diagnosis was confirmed histologically (caseating granuloma) in 84.7% of the cases, while MPE was biopsy-proven in 63.9% of the cases. Echographic septation was more evident in TBE than in MPE (46.5% vs. 8.2%, p < 0.001). Multivariate logistic regression analysis showed that male sex, serum leucocyte count ≤9 × 10⁹/L or pleural fluid protein ≥50 g/L, and echographic septation (aOR: 9.28, p < 0.001) were independent predictors for TBE. These parameters collectively provided a diagnostic accuracy of 79.61% (95% CI 74.13-84.38).

CONCLUSIONS

Echographic septation may potentially facilitate discrimination between TBE and MPE as part of a clinical prediction model. Prospective validation of this prediction model in an external cohort is anticipated.

Investigating the appropriate adenosine deaminase cutoff value for the diagnosis of tuberculous pleural effusion in a country with decreasing TB burden

As the burden of tuberculosis (TB) in South Korea decreases while that of malignancy increases with an aging society, the composition of etiology for pleural effusion is changing. The aim of this study was to investigate the diagnostic value of adenosine deaminase (ADA) for diagnosis of tuberculous pleural effusion (TPE) in this circumstance. Medical records of patients who underwent medical thoracoscopy from May 2015 to September 2020 in Incheon St. Mary Hospital, Korea were retrospectively reviewed. TPE was diagnosed if one of the following criteria was met: (1) granuloma in pleura, (2) positive TB polymerase chain reaction or culture in pleural fluid or tissue with non-specific pathologic findings in pleura, or (3) bacteriologically confirmed pulmonary TB with non-specific pathologic findings in pleura. A total of 292 patients, including 156 with malignant pleural effusion (MPE), 52 with TPE, and 84 with other benign effusion, were analyzed. Among 206 patients with lymphocyte dominant pleural effusion, the area under receiver characteristic curve of ADA for diagnosis of TPE was 0.971. The sensitivity and specificity of a current cutoff value of 40 IU/L were 1.00 and 0.61, respectively, whereas those of a raised cutoff value of 70 IU/L were 0.93 and 0.93, respectively. Among 54 patients with ADA levels of 40–70 IU/L, 30 (55.6%) patients were diagnosed as MPE, 21 (38.9%) as other benign effusion, and only 3 (5.6%) as TPE. Caution is needed in clinical diagnosis of TPE with current ADA cutoff value in countries with decreasing TB incidence, due to many false positive cases.

Classification of pleural effusions using deep learning visual models: contrastive-loss

Blood and fluid analysis is extensively used for classifying the etiology of pleural effusion. However, most studies focused on determining the presence of a disease. This study classified pleural effusion etiology employing deep learning models by applying contrastive-loss. Patients with pleural effusion who underwent thoracentesis between 2009 and 2019 at the Asan Medical Center were analyzed. Five different models for categorizing the etiology of pleural effusion were compared. The performance metrics were top-1 accuracy, top-2 accuracy, and micro-and weighted-AUROC. UMAP and t-SNE were used to visualize the contrastive-loss model’s embedding space. Although the 5 models displayed similar performance in the validation set, the contrastive-loss model showed the highest accuracy in the extra-validation set. Additionally, the accuracy and micro-AUROC of the contrastive-loss model were 81.7% and 0.942 in the validation set, and 66.2% and 0.867 in the extra-validation set. Furthermore, the embedding space visualization in the contrastive-loss model exhibited typical and atypical effusion results by comparing the true and false positives of the rule-based criteria. Therefore, classifying the etiology of pleural effusion was achievable using the contrastive-loss model. Conclusively, visualization of the contrastive-loss model will provide clinicians with valuable insights for etiology diagnosis by differentiating between typical and atypical disease types.

Hyporexia and cellular/biochemical characteristics of pleural fluid as predictive variables on a model for pleural tuberculosis diagnosis

Objectives

Pleural tuberculosis (PlTB) diagnosis is a challenge due to its paucibacillary nature and to the need of invasive procedures. This study aimed to identify easily available variables and build a predictive model for PlTB diagnosis which may allow earlier and affordable alternative strategy to be used in basic health care units.

Methods

An observational cross-sectional study compared PlTB and non-TB patients followed at a tertiary Brazilian hospital between 2010 and 2018. Unconditional logistic regression analysis was performed and a Decision Tree Classifier (DTC) model was validated and applied in additional PlTB patients with empiric diagnosis. The accuracy (Acc), sensitivity (Se), specificity (Sp), positive and negative predictive values were calculated.

Results

From 1,135 TB patients, 160 were considered for analysis (111 confirmed PlTB and 49 unconfirmed PlTB). Indeed, 58 non-TB patients were enrolled as controls. Hyporexia [adjusted odds ratio (aOR) 27.39 (95% CI 6.26 – 119.89)] and cellular/biochemical characteristics on pleural fluid (PF) (polimorphonuclear in two categories: 3-14% aOR 26.22, 95% CI 7.11 – 96.68 and < 3% aOR 28.67, 95% CI 5.51 – 149.25; and protein ≥ 5g/dL aOR 7.24, 95% CI 3.07 – 17.11) were associated with higher risk for TB. The DTC constructed using these variables showed Acc=87.6%, Se=89.2%, Sp=84.5% for PlTB diagnosis and was successfully applied in unconfirmed PlTB patients.

Conclusion

The DTC model showed an excellent performance for PlTB diagnosis and can be considered as an alternative diagnostic strategy by using clinical patterns in association with PF cellular/biochemical characteristics, which were affordable and easily performed in basic health care units.

Laboratory and radiological discrimination between tuberculous and malignant pleural effusions with high adenosine deaminase levels

BACKGROUND/AIMS

Pleural fluid adenosine deaminase (ADA) levels are useful in discriminating tuberculous pleural effusions (TPEs) from malignant pleural effusions (MPEs). However, some patients with MPE exhibit high-ADA levels, which may mimic TPEs. There is limited data regarding the differential diagnosis between high-ADA MPE and high-ADA TPE. This study aimed to identify the predictors for distinguishing high-ADA MPEs from high-ADA TPEs.

METHODS

Patients with TPE and MPE with pleural fluid ADA levels ≥ 40 IU/L were included in this study. Clinical, laboratory, and radiological data were compared between the two groups. Independent predictors and their diagnostic performance for high-ADA MPEs were evaluated using multivariate logistic regression analysis and receiver operating characteristic curve.

RESULTS

A total of 200 patients (high-ADA MPE, n = 30, and high-ADA TPE, n = 170) were retrospectively included. In the multivariate analysis, pleural fluid ADA, pleural fluid carcinoembryonic antigen (CEA), and pleural nodularity were independent discriminators between high-ADA MPE and high-ADA TPE groups. Using pleural ADA level of 40 to 56 IU/L (3 points), pleural CEA level ≥ 6 ng/mL (6 points), and presence of pleural nodularity (3 points) for predicting high-ADA MPEs, a sum score ≥ 6 points yielded a sensitivity of 90%, specificity of 96%, positive predictive value of 82%, negative predictive value of 98%, and area under the receiver operating characteristic curve of 0.965.

CONCLUSION

A scoring system using three parameters may be helpful in guiding the differential diagnosis between high-ADA MPEs and high-ADA TPEs.

A retrospective study on the combined biomarkers and ratios in serum and pleural fluid to distinguish the multiple types of pleural effusion

Purpose

Pleural effusion (PE) is a common clinical manifestation, and millions of people suffer from pleural disease. Herein, this retrospective study was performed to evaluate the biomarkers and ratios in serum and pleural fluid (PF) for the differential diagnosis of the multiple types of PE and search for a new diagnostic strategy for PE.

Methods

In-patients, who developed tuberculous PE (TPE), malignant PE (MPE), complicated parapneumonic effusion (CPPE), uncomplicated PPE (UPPE), or PE caused by connective tissue diseases (CTDs) and underwent thoracentesis at Peking University People’s Hospital from November 2016 to April 2019, were included in this study. Eleven biomarkers and their ratios in serum and PF were investigated and compared between pairs of the different PE groups, and a decision-tree was developed.

Results

Totally 112 PE cases, including 25 MPE, 33 TPE, 19 CPPE, 27 UPPE, and 8 PE caused by CTDs, were reviewed. Biomarkers and ratios showed good diagnostic performance with high area under the curve values, sensitivities, and specificities for the differential diagnosis of the multiple types of PE. According to the decision-tree analysis, the combination of adenosine deaminase (ADA), serum albumin, serum lactate dehydrogenase, total protein, PF-LDH/ADA, and PF-LDH/TP provided the best predictive capacity with an overall accuracy of 84.8%; the sensitivity and specificity for TPE diagnosis were 100% and 98.7%, respectively.

Conclusion

The biomarkers and ratios showed good diagnostic performance, and a decision-tree with an overall accuracy of 84.8% was developed to differentiate the five types of PE in clinical settings.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12890-021-01459-w.

Driverless artificial intelligence framework for the identification of malignant pleural effusion

Our study aimed to explore the applicability of deep learning and machine learning techniques to distinguish MPE from BPE. We initially used a retrospective cohort with 726 PE patients to train and test the predictive performances of the driverless artificial intelligence (AI), and then stacked with a deep learning and five machine learning models, namely gradient boosting machine (GBM), extreme gradient boosting (XGBoost), extremely randomized trees (XRT), distributed random forest (DRF), and generalized linear models (GLM). Furthermore, a prospective cohort with 172 PE patients was applied to detect the external validity of the predictive models. The area under the curve (AUC) in the training, test and validation set were deep learning (0.995, 0.848, 0.917), GBM (0.981, 0.910, 0.951), XGBoost (0.933, 0.916, 0.935), XRT (0.927, 0.909, 0.963), DRF (0.906, 0.809, 0.969), and GLM (0.898, 0.866, 0.892), respectively. Although the Deep Learning model had the highest AUC in the training set (AUC = 0.995), GBM demonstrated stable and high predictive efficiency in three data sets. The final AI model by stacked ensemble yielded optimal diagnostic performance with AUC of 0.991, 0.912 and 0.953 in the training, test and validation sets, respectively. Using the driverless AI framework based on the routinely collected clinical data could significantly improve diagnostic performance in distinguishing MPE from BPE.

Extrapulmonary tuberculosis in HIV-infected patients in rural Tanzania: The prospective Kilombero and Ulanga antiretroviral cohort

BACKGROUND

In sub-Saharan Africa, diagnosis and management of extrapulmonary tuberculosis (EPTB) in people living with HIV (PLHIV) remains a major challenge. This study aimed to characterize the epidemiology and risk factors for poor outcome of extrapulmonary tuberculosis in people living with HIV (PLHIV) in a rural setting in Tanzania.

METHODS

We included PLHIV >18 years of age enrolled into the Kilombero and Ulanga antiretroviral cohort (KIULARCO) from 2013 to 2017. We assessed the diagnosis of tuberculosis by integrating prospectively collected clinical and microbiological data. We calculated prevalence- and incidence rates and used Cox regression analysis to evaluate the association of risk factors in extrapulmonary tuberculosis (EPTB) with a combined endpoint of lost to follow-up (LTFU) and death.

RESULTS

We included 3,129 subjects (64.5% female) with a median age of 38 years (interquartile range [IQR] 31-46) and a median CD4+ cell count of 229/μl (IQR 94-421) at baseline. During the median follow-up of 1.25 years (IQR 0.46-2.85), 574 (18.4%) subjects were diagnosed with tuberculosis, whereof 175 (30.5%) had an extrapulmonary manifestation. Microbiological evidence by Acid-Fast-Bacillus stain (AFB-stain) or Xpert® MTB/RIF was present in 178/483 (36.9%) patients with pulmonary and in 28/175 (16.0%) of patients with extrapulmonary manifestations, respectively. Incidence density rates for pulmonary Tuberculosis (PTB and EPTB were 17.9/1000person-years (py) (95% CI 14.2-22.6) and 5.8/1000 py (95% CI 4.0-8.5), respectively. The combined endpoint of death and LTFU was observed in 1058 (33.8%) patients, most frequently in the subgroup of EPTB (47.2%). Patients with EPTB had a higher rate of the composite outcome of death/LTFU after TB diagnosis than with PTB [HR 1.63, (1.14-2.31); p = 0.006]. The adjusted hazard ratios [HR (95% CI)] for death/LTFU in EPTB patients were significantly increased for patients aged >45 years [HR 1.95, (1.15-3.3); p = 0.013], whereas ART use was protective [HR 0.15, (0.08-0.27); p <0.001].

CONCLUSIONS

Extrapulmonary tuberculosis was a frequent manifestation in this cohort of PLHIV. The diagnosis of EPTB in the absence of histopathology and mycobacterial culture remains challenging even with availability of Xpert® MTB/RIF. Patients with EPTB had increased rates of mortality and LTFU despite early recognition of the disease after enrollment.

Score for Differentiating Pleural Tuberculosis from Malignant Effusion

Differential diagnosis of lymphocytic pleural effusions between tuberculous (TBE) and malignant (ME) effusion is usually difficult in daily practice. Our aim was to develop a score to differentiate TBE from ME effusions. A cohort of 138 consecutive patients with pleural effusion was prospectively studied from May 2014 through June 2017. Glucose, lactate dehydrogenase (LDH), proteins, white cell count, lactic acid, and pH in the pleural fluid were measured. Pleural effusions other than lymphocytic, patients with a final diagnosis other than tuberculosis or malignancy, and patients who met Light’s criteria for transudate were excluded. Eighty-two samples (47 TBE and 35 ME) were included in the analysis. Using logistic regression analysis and Wald test, we developed a score including age, glucose, proteins, and lactic acid. The receiver operating characteristic curve (ROC) for the score was determined, and the area under the curve (AUC) was calculated. A cutoff of eight points was required to achieve 93.5% sensitivity, 78% specificity, and a likelihood ratio of 4.26 to distinguish tuberculosis from malignant pleural effusion. The AUC of the score was 0.915 (95% CI = 0.82–0.96).

Development of a clinical prediction rule for the diagnosis of pleural tuberculosis in Peru

OBJECTIVES

To develop a clinical prediction rule (CPR) for the diagnosis of pleural tuberculosis (PT) in patients with pleural exudates in Peru.

METHODS

Clinical and laboratory information was collected from patients with exudative pleural effusion attending two reference hospitals in Lima, Peru. Predictive findings associated with PT in a multiple logistic regression model were used to develop the CPR. A definite diagnosis of PT was based on a composite reference standard including bacteriological and/or histological analysis of pleural fluid and pleural biopsy specimens.

RESULTS

A total of 238 patients were included in the analysis, of whom 176 had PT. Age, sex, previous contact with a TB patient, presence of lymphadenopathy, and pleural adenosine deaminase (ADA) levels were found to be independently associated with PT. These predictive findings were used to construct a CPR, for which the area under the receiver operating characteristics curve (AUC) was 0.92. The single best cut-off point was a score of ≥60 points, which had a sensitivity of 88%, specificity of 92%, a positive likelihood ratio of 10.9, and a negative likelihood ratio of 0.13.

CONCLUSIONS

The CPR is accurate for the diagnosis of PT and could be useful for treatment initiation while avoiding pleural biopsy. A prospective evaluation is needed before its implementation in different settings.

Developing a new intelligent system for the diagnosis of tuberculous pleural effusion

BACKGROUND AND OBJECTIVE

In countries with high prevalence of tuberculosis (TB), clinicians often diagnose tuberculous pleural effusion (TPE) by using diagnostic tests, which have not only poor sensitivity, but poor availability as well. The aim of our study is to develop a new artificial intelligence based diagnostic model that is accurate, fast, non-invasive and cost effective to diagnose TPE. It is expected that a tool derived based on the model be installed on simple computer devices (such as smart phones and tablets) and be used by clinicians widely.

METHODS

For this study, data of 140 patients whose clinical signs, routine blood test results, blood biochemistry markers, pleural fluid cell type and count, and pleural fluid biochemical tests' results were prospectively collected into a database. An Artificial intelligence based diagnostic model, which employs moth flame optimization based support vector machine with feature selection (FS-MFO-SVM), is constructed to predict the diagnosis of TPE.

RESULTS

The optimal model results in an average of 95% accuracy (ACC), 0.9564 the area under the receiver operating characteristic curve (AUC), 93.35% sensitivity, and 97.57% specificity for FS-MFO-SVM.

CONCLUSIONS

The proposed artificial intelligence based diagnostic model is found to be highly reliable for diagnosing TPE based on simple clinical signs, blood samples and pleural effusion samples. Therefore, the proposed model can be widely used in clinical practice and further evaluated for use as a substitute of invasive pleural biopsies.

Performance of clinical prediction rules for diagnosis of pleural tuberculosis in a high-incidence setting

OBJECTIVES

Diagnosis of pleural tuberculosis (PT) is still a challenge, particularly in resource-constrained settings. Alternative diagnostic tools are needed. We aimed at evaluating the utility of Clinical Prediction Rules (CPRs) for diagnosis of pleural tuberculosis in Peru.

METHODS

We identified CPRs for diagnosis of PT through a structured literature search. CPRs using high-complexity tests, as defined by the FDA, were excluded. We applied the identified CPRs to patients with pleural exudates attending two third-level hospitals in Lima, Peru, a setting with high incidence of tuberculosis. Besides pleural fluid analysis, patients underwent closed pleural biopsy for reaching a final diagnosis through combining microbiological and histopathological criteria. We evaluated the performance of the CPRs against this composite reference standard using classic indicators of diagnostic test validity.

RESULTS

We found 15 eligible CPRs, of which 12 could be validated. Most included ADA, age, lymphocyte proportion and protein in pleural fluid as predictive findings. A total of 259 patients were included for their validation, of which 176 (67%) had PT and 50 (19%) malignant pleural effusion. The overall accuracy of the CPRs varied from 41% to 86%. Two had a positive likelihood ratio (LR) above 10, but none a negative LR below 0.1. ADA alone at a cut-off of ≥40 IU attained 87% diagnostic accuracy and had a positive LR of 6.6 and a negative LR of 0.2.

CONCLUSION

Many CPRs for PT are available. In addition to ADA alone, none of them contributes significantly to diagnosis of PT.

Use of Data Mining to Predict the Risk Factors Associated With Osteoporosis and Osteopenia in Women

Osteoporosis has recently been acknowledged as a major public health issue in developed countries because of the decrease in the quality of life of the affected person and the increase in public costs due to complete or partial physical disability. The aim of this study was to use the J48 algorithm as a classification task for data from women exhibiting changes in bone densitometry. The study population included all patients treated at the diagnostic center for bone densitometry since 2010. Census sample data collection was conducted as all elements of the population were included in the sample. The service in question provides care to patients via the Brazilian Unified Health System and private plans. The results of the classification task were analyzed using the J48 algorithm, and among the dichotomized variables associated with a diagnosis of osteoporosis, the mean accuracy was 74.0 (95% confidence interval [CI], 61.0-68.0) and the mean area under the curve of the receiver operating characteristic (ROC) curve was 0.65 (95% CI, 0.64-0.66), with a mean sensitivity of 76.0 (95% CI, 76.0-76.0) and a mean specificity of 48.0 (95% CI, 46.0-49.0). The analyzed results showed higher values of sensitivity, accuracy, and curve of the ROC area in experiments conducted with individuals with osteoporosis. Most of the generated rules were consistent with the literature, and the few differences might serve as hypotheses for further studies.

Novel biomarker analysis of pleural effusion enhances differentiation of tuberculous from malignant pleural effusion

Lymphocytic pleurisy is commonly observed in tuberculosis and cancer. Noninvasive biomarkers are needed to distinguish tuberculous pleural effusion (TPE) from malignant pleural effusion (MPE) because current clinical diagnostic procedures are often invasive. We identified immune response biomarkers that can discriminate between TPE and MPE. Fourteen pleural effusion biomarkers were compared in 22 MPE patients and five TPE patients. Of the innate immunity biomarkers, the median levels of interleukin (IL)-1β and interferon-induced protein-10 (IP-10) were higher in TPE patients than in MPE patients (P<0.05 and P<0.01, respectively). Of the adaptive immunity biomarkers, the median levels of IL-13 and interferon-γ (IFN-γ) were higher in TPE patients than in MPE patients (P<0.05). In addition, the levels of basic fibroblast growth factor were higher in MPE patients than in TPE patients (P<0.05). Receiver operator characteristic analysis of these biomarkers was performed, resulting in the highest area under the curve (AUC) for IP-10 (AUC =0.95, 95% confidence interval, P<0.01), followed by IL-13 (AUC =0.86, 95% confidence interval, P<0.05). Our study shows that five biomarkers (IL-1β, IP-10, IFN-γ, IL-13, and basic fibroblast growth factor) have a potential diagnostic role in differentiating TPE from MPE, particularly in lung cancer-related MPE.

Age and sex distribution in malignant and tuberculous serous effusions: A study of 127 patients and review of the literature

AIM

Tuberculosis and carcinomatosis are the two most frequent causes of pleural effusion and exudative ascites, and both are characterized by lymphocyte-rich effusion. We attempted to discover if there is any significant difference in the age and sex distribution between patients presenting with these two conditions.

METHODS

A total of 161 serous effusion samples from 127 patients (89 with pleural effusion and 38 with ascites) having follow-up biopsy and histopathological examination were included in the present study. Three groups - malignancy (47 patients), tuberculosis (47) and non-tuberculous benign lesions (26) as per histopathological diagnoses - were compared in respect to age and sex distribution.

RESULTS

A total of 29 (61.7%) patients with malignancy were aged ≥50 years as compared with three (6.4%) tuberculosis patients with serous effusions (P = 0.00000). A similar trend was observed in the ≥60 years age group (18 or 38.3% malignancy vs none with tuberculosis, P = 0.00000). A total of 36 (76.6%) tuberculous effusion patients were aged less than 40 years as opposed to eight (17.0%) patients with malignant effusions (P = 0.00000). There was also s significant difference between tuberculous and non-tuberculous benign lesions in the ≥50 years age group (6.4% vs 69.2%, P = 0.00000), but no significant difference between malignancy and non-tuberculous benign lesions (P = 0.61385). There were 31 female (66.0%) patients with malignancy, which was significantly higher than that of patients with tuberculosis (16, [34%], P = 0.00365) and non-specific inflammation/benign lesions (23.1%, P = 0.00059). However, the difference between tuberculosis and non-tuberculous benign lesions was not significant (P = 0.42756).

CONCLUSION

Whereas malignancy in serous effusions is found in older and middle-aged people, tuberculous effusion is a disease of younger people.

Tuberculous pleurisy: an update

Tuberculous pleurisy is the most common form of extrapulmonary tuberculosis in Korea. Tuberculous pleurisy presents a diagnostic and therapeutic problem due to the limitations of traditional diagnostic tools. There have been many clinical research works during the past decade. Recent studies have provided new insight into the tuberculous pleurisy, which have a large impact on clinical practice. This review is a general overview of tuberculous pleurisy with a focus on recent findings on the diagnosis and management.

Tuberculosis disease diagnosis using artificial immune recognition system

BACKGROUND

There is a high risk of tuberculosis (TB) disease diagnosis among conventional methods.

OBJECTIVES

This study is aimed at diagnosing TB using hybrid machine learning approaches.

MATERIALS AND METHODS

Patient epicrisis reports obtained from the Pasteur Laboratory in the north of Iran were used. All 175 samples have twenty features. The features are classified based on incorporating a fuzzy logic controller and artificial immune recognition system. The features are normalized through a fuzzy rule based on a labeling system. The labeled features are categorized into normal and tuberculosis classes using the Artificial Immune Recognition Algorithm.

RESULTS

Overall, the highest classification accuracy reached was for the 0.8 learning rate (α) values. The artificial immune recognition system (AIRS) classification approaches using fuzzy logic also yielded better diagnosis results in terms of detection accuracy compared to other empirical methods. Classification accuracy was 99.14%, sensitivity 87.00%, and specificity 86.12%.

A decision tree for differentiating multiple system atrophy from Parkinson's disease using 3-T MR imaging

OBJECTIVE

To develop a decision tree based on standard magnetic resonance imaging (MRI) and diffusion tensor imaging to differentiate multiple system atrophy (MSA) from Parkinson's disease (PD).

METHODS

3-T brain MRI and DTI (diffusion tensor imaging) were performed on 26 PD and 13 MSA patients. Regions of interest (ROIs) were the putamen, substantia nigra, pons, middle cerebellar peduncles (MCP) and cerebellum. Linear, volumetry and DTI (fractional anisotropy and mean diffusivity) were measured. A three-node decision tree was formulated, with design goals being 100 % specificity at node 1, 100 % sensitivity at node 2 and highest combined sensitivity and specificity at node 3.

RESULTS

Nine parameters (mean width, fractional anisotropy (FA) and mean diffusivity (MD) of MCP; anteroposterior diameter of pons; cerebellar FA and volume; pons and mean putamen volume; mean FA substantia nigra compacta-rostral) showed statistically significant (P < 0.05) differences between MSA and PD with mean MCP width, anteroposterior diameter of pons and mean FA MCP chosen for the decision tree. Threshold values were 14.6 mm, 21.8 mm and 0.55, respectively. Overall performance of the decision tree was 92 % sensitivity, 96 % specificity, 92 % PPV and 96 % NPV. Twelve out of 13 MSA patients were accurately classified.

CONCLUSION

Formation of the decision tree using these parameters was both descriptive and predictive in differentiating between MSA and PD.

KEY POINTS

• Parkinson's disease and multiple system atrophy can be distinguished on MR imaging. • Combined conventional MRI and diffusion tensor imaging improves the accuracy of diagnosis. • A decision tree is descriptive and predictive in differentiating between clinical entities. • A decision tree can reliably differentiate Parkinson's disease from multiple system atrophy.

Pattern Discovery from Biological Data

Extracting useful information from structured and unstructured biological data is crucial in the health industry. Some examples include medical practitioner’s need to identify breast cancer patient in the early stage, estimate survival time of a heart disease patient, or recognize uncommon disease characteristics which suddenly appear. Currently there is an explosion in biological data available in the data bases. But information extraction and true open access to data are require time to resolve issues such as ethical clearance. The emergence of novel IT technologies allows health practitioners to facilitate the comprehensive analyses of medical images, genomes, transcriptomes, and proteomes in health and disease. The information that is extracted from such technologies may soon exert a dramatic change in the pace of medical research and impact considerably on the care of patients. The current research will review the existing technologies being used in heart and cancer research. Finally this research will provide some possible solutions to overcome the limitations of existing technologies. In summary the primary objective of this research is to investigate how existing modern machine learning techniques (with their strength and limitations) are being used in the indent of heartbeat related disease and the early detection of cancer in patients. After an extensive literature review these are the objectives chosen: to develop a new approach to find the association between diseases such as high blood pressure, stroke and heartbeat, to propose an improved feature selection method to analyze huge images and microarray databases for machine learning algorithms in cancer research, to find an automatic distance function selection method for clustering tasks, to discover the most significant risk factors for specific cancers, and to determine the preventive factors for specific cancers that are aligned with the most significant risk factors. Therefore we propose a research plan to attain these objectives within this chapter. The possible solutions of the above objectives are: new heartbeat identification techniques show promising association with the heartbeat patterns and diseases, sensitivity based feature selection methods will be applied to early cancer patient classification, meta learning approaches will be adopted in clustering algorithms to select an automatic distance function, and Apriori algorithm will be applied to discover the significant risks and preventive factors for specific cancers. We expect this research will add significant contributions to the medical professional to enable more accurate diagnosis and better patient care. It will also contribute in other area such as biomedical modeling, medical image analysis and early diseases warning.

Myelomatous pleural effusion: a case series in a single institution and literature review

Background

Myelomatous pleural effusion (MPE) is rare in myeloma patients. We present a consecutive series of patients with MPE in a single institution.

Methods

We retrospectively reviewed the medical records of 19 patients diagnosed with MPE between 1989 and 2008 at the Asan Medical Center. Diagnoses were confirmed by cytologic identification of malignant plasma cells in the pleural fluid.

Results

Our patients showed dominance of IgA (36.8%) and IgD (31.6%) subtypes. Of 734 myeloma patients, the incidence of MPE was remarkably high for the IgD myeloma subtype (16.7%), compared to the other subtypes (1.4% for IgG and 4.6% for IgA). At the time of diagnosis of MPE, elevated serum β2-microglobulin, anemia, elevated serum lactate dehydrogenase, and elevated creatinine levels were found in 100%, 89.5%, 83.3%, and 57.9% of the patients, respectively. Approximately one-third (31.3%) of the patients had adenosine deaminase (ADA) activities in their pleural fluid exceeding the upper limit of the reported cutoff values for tuberculous pleural effusion (55.8 U/L). Chromosome 13 abnormality was seen in 77.8% of the tested patients. The median survival period from the development of MPE was 2.8 months.

Conclusions

Patients with MPE have aggressive clinical and laboratory characteristics. The preponderance of IgD myeloma in MPE patients is a noteworthy finding because IgD myeloma is a rare subtype. Elevated ADA activity in the pleural fluid is also noteworthy, and may be helpful for detecting MPE. Physicians treating myeloma patients should monitor the development of MPE and consider the possibility of a worse clinical course.

Translational advances in pleural malignancies

Pleural malignancies, including primary malignant pleural mesothelioma and secondary pleural metastasis of various tumours resulting in malignant pleural effusion, are frequent and lethal diseases that deserve devoted translational research efforts for improvements to be introduced to the clinic. This paper highlights select clinical advances that have been accomplished recently and that are based on preclinical research on pleural malignancies. Examples are the establishment of folate antimetabolites in mesothelioma treatment, the use of PET in mesothelioma management and the discovery of mesothelin as a marker of mesothelioma. In addition to established translational advances, this text focuses on recent research findings that are anticipated to impact clinical pleural oncology in the near future. Such progress has been substantial, including the development of a genetic mouse model of mesothelioma and of transplantable models of pleural malignancies in immunocompetent hosts, the deployment of stereological and imaging methods for integral assessment of pleural tumour burden, as well as the discovery of the therapeutic potential of aminobiphosphonates, histone deacetylase inhibitors and ribonucleases against malignant pleural disease. Finally, key obstacles to overcome towards a more rapid advancement of translational research in pleural malignancies are outlined. These include the dissection of cell-autonomous and paracrine pathways of pleural tumour progression, the study of mesothelioma and malignant pleural effusion separately from other tumours at both the clinical and preclinical levels, and the expansion of tissue banks and consortia of clinical research of pleural malignancies.

III Brazilian Thoracic Association Guidelines on tuberculosis

New scientific articles about tuberculosis (TB) are published daily worldwide. However, it is difficult for health care workers, overloaded with work, to stay abreast of the latest research findings and to discern which information can and should be used in their daily practice on assisting TB patients. The purpose of the III Brazilian Thoracic Association (BTA) Guidelines on TB is to critically review the most recent national and international scientific information on TB, presenting an updated text with the most current and useful tools against TB to health care workers in our country. The III BTA Guidelines on TB have been developed by the BTA Committee on TB and the TB Work Group, based on the text of the II BTA Guidelines on TB (2004). We reviewed the following databases: LILACS (SciELO) and PubMed (Medline). The level of evidence of the cited articles was determined, and 24 recommendations on TB have been evaluated, discussed by all of the members of the BTA Committee on TB and of the TB Work Group, and highlighted. The first version of the present Guidelines was posted on the BTA website and was available for public consultation for three weeks. Comments and critiques were evaluated. The level of scientific evidence of each reference was evaluated before its acceptance for use in the final text.

Pleural fluid characteristics of tuberculous pleural effusions

Mycobacterium tuberculosis (TB) infection of the pleural space is an important cause of pleural effusion in areas of high TB prevalence. Microbiological analyses of pleural fluid in the acute setting may be negative. Consequently, investigations may proceed to more invasive techniques, such as pleural biopsy or thoracoscopy. Ongoing research has led to implementing a number of additional fluid analyses that may lead to a diagnosis without a need for further invasive procedures. In this review, we discuss the characteristics of tuberculous pleural fluid that may assist in this important diagnosis, and we highlight the benefits of specific biomarker analyses. English-language publications from a MEDLINE search and references from relevant articles from January 1, 1990 to September 1, 2009 were reviewed. The key words searched included tuberculosis, pleural fluid, effusion, diagnosis, adenosine deaminase, and interferon.

Tuberculous pleural effusion

Tuberculous pleural effusion is one of the most common forms of extrapulmonary tuberculosis (TB). The immediate cause of the effusion is a delayed hypersensitivity response to mycobacterial antigens in the pleural space. For this reason microbiological analyses are often negative and limited by the lengthy delay in obtaining results. In areas with high TB prevalence, pleural fluid adenosine deaminase (ADA) levels greater than 40 U/l argue strongly for TB; in contrast, low levels of pleural ADA have high negative predictive value in low-prevalence countries. The specificity of this enzyme increases if only lymphocytic exudates are considered. The shortcoming of the ADA test is its inability to provide culture and drug sensitivity information, which is paramount in countries with a high degree of resistance to anti-TB drugs. Sputum induction (in addition to pleural fluid) for acid-fast bacilli and culture is a recommended procedure in all patients with TB pleurisy. The microscopic-observation drug-susceptibility assay performed on pleural fluid or pleural tissue increases by two to three times the detection of TB over conventional cultures, and it allows for the identification of multidrug-resistant TB. A reasonable management strategy for pleural TB would be to initiate a four-drug regimen and perform a therapeutic thoracentesis in patients with large, symptomatic effusions.