Harnessing Big Data to Optimize an Algorithm for Rapid Diagnosis of Pulmonary Tuberculosis in a Real-World Setting

Evaluation of mycobacteria infection prevalence and optimization of the identification process in North Sardinia, Italy

IMPORTANCE

Mycobacterial infection is a major threat to public health worldwide. Accurate identification of infected species and drug resistance detection are critical factors in treatment. We focused on shortening the turn-around time of identifying mycobacteria species and antibiotic resistance tests.

Cerebrospinal fluid protein levels are elevated 100 times in a Leptomeningeal metastasis patient: a case report and literature review

Leptomeningeal metastasis (LM) has a high degree of malignancy and high mortality. We describe a patient admitted to hospital with acute lower extremity weakness, dysuria, and high intracranial pressure. Enhanced magnetic resonance imaging (MRI) showed extensive enhancement of the leptomeningeal and spinal meninges with multiple nodular changes and extensive fusion. His cerebrospinal fluid (CSF) was yellow and cloudy, the Pandy test was strongly positive (++++), the protein was 46 g/L (normal range 0.15-0.45 g/L), which attracted our attention. Initially, miliary TB with associated tuberculous meningitis (TBM) was diagnosed, and neurosarcoidosis cannot be ruled out. After poor therapeutic effect of standard antituberculosis (anti-TB) therapy, further inspection found that malignant cells were detected by cerebrospinal fluid (CSF) cytology. PET/CT suggested the diagnosis of LM. The purpose of this paper is to describe the characteristics of atypical diffuse LM. In conclusion, when patient with unexplained high levels of CSF protein, it is necessary to be alert to the diagnosis of LM. Multiple examinations of fresh CSF are helpful to increase the positive detection rate of tumor cells. Early diagnosis and active treatment are conducive to improving survival rate.

Convolutional neural network based on T-SPOT.TB assay promoting the discrimination between active tuberculosis and latent tuberculosis infection

OBJECTIVES

The study aims to investigate the potential of convolutional neural network (CNN) based on spot image of T-SPOT assay for distinguishing active tuberculosis (ATB) from latent tuberculosis infection (LTBI).

METHODS

CNN was applied to recognize and classify T-SPOT spot image. Logistic regression was used to establish prediction model based on CNN.

RESULTS

Areas under the receiver operating characteristic curve (AUCs) of early secreted antigenic target 6 (ESAT-6) CNN, culture filtrate protein 10 (CFP-10) CNN, and phytohemagglutinin (PHA) CNN were more than 0.7 in differentiating ATB from LTBI, while the performance of these indicators was significantly better than that of spot number. Furthermore, prediction model based on the combination of CNNs yielded an AUC of 0.898. The model presented a sensitivity of 85.76% and a specificity of 90.23%.

CONCLUSIONS

The current study identified CNN based on T-SPOT spot image with the potential to serve as a tool for TB diagnostics.

Combination of HLA-DR on Mycobacterium tuberculosis-Specific Cells and Tuberculosis Antigen/Phytohemagglutinin Ratio for Discriminating Active Tuberculosis From Latent Tuberculosis Infection

Background

Novel approaches for tuberculosis (TB) diagnosis, especially for distinguishing active TB (ATB) from latent TB infection (LTBI), are urgently warranted. The present study aims to determine whether the combination of HLA-DR on Mycobacterium tuberculosis (MTB)-specific cells and TB antigen/phytohemagglutinin (TBAg/PHA) ratio could facilitate MTB infection status discrimination.

Methods

Between June 2020 and June 2021, participants with ATB and LTBI were recruited from Tongji Hospital (Qiaokou cohort) and Sino-French New City Hospital (Caidian cohort), respectively. The detection of HLA-DR on MTB-specific cells upon TB antigen stimulation and T-SPOT assay were simultaneously performed on all subjects.

Results

A total of 116 (54 ATB and 62 LTBI) and another 84 (43 ATB and 41 LTBI) cases were respectively enrolled from Qiaokou cohort and Caidian cohort. Both HLA-DR on IFN-γ⁺TNF-α⁺ cells and TBAg/PHA ratio showed discriminatory value in distinguishing between ATB and LTBI. Receiver operator characteristic (ROC) curve analysis showed that HLA-DR on IFN-γ⁺TNF-α⁺ cells produced an area under the ROC curve (AUC) of 0.886. Besides, TBAg/PHA ratio yield an AUC of 0.736. Furthermore, the combination of these two indicators resulted in the accurate discrimination with an AUC of 0.937. When the threshold was set as 0.36, the diagnostic model could differentiate ATB from LTBI with a sensitivity of 92.00% and a specificity of 81.82%. The performance obtained in Qiaokou cohort was further validated in Caidian cohort.

Conclusions

The combination of HLA-DR on MTB-specific cells and TBAg/PHA ratio could serve as a robust tool to determine TB disease states.