Elsevier

Tuberculosis

Volume 97, March 2016, Pages 26-32
Tuberculosis

Diagnostics
Evaluation of a lens-free imager to facilitate tuberculosis diagnostics in MODS

https://doi.org/10.1016/j.tube.2015.12.001Get rights and content

Summary

Tuberculosis (TB) control efforts are hampered by a mismatch in diagnostic technology. Lack of adequate early diagnostics and Multi-drug resistant (MDR) detection is a critical problem in control efforts. Alternate and novel diagnostic approaches are required, especially in low-resources settings where they are needed most.

The Microscopic Observation Drug Susceptibility (MODS) assay is a cost-effective, highly sensitive, and specific method based on the detection of characteristic cording growth patterns of Mycobacterium tuberculosis (MTB), in microscopic examination of a liquid culture under an inverted microscope. By adding antimicrobials to the wells, MODS also determines antimicrobial susceptibility in both MDR and Extreme Drug Resistant (XDR) tuberculosis.

The interpretation of a MODS culture performed in a 24 well plate, requires an extensive inspection over the entire surface to detect TB cords. This process requires significant time and effort from a trained microscopist.

We evaluated a lens-free imager system, able to render microscopic images of live specimens, for the proof of principle to be used for MODS culture interpretation. The lens-free imager system is able to digitalize a 24-mm2 surface with approximately 40X magnification in a single capture.

The evaluation of the lens-free imager found that it produced microscopic images that were adequate for MODS interpretation by a human expert. Compared to the average time that takes a microscopist to completely examine a MODS culture sample, the lens free imager notably reduced the time of inspection.

Therefore, lens-free imager variants may constitute promising systems to aid in the diagnostics of tuberculosis, by simplifying and reducing the time of inspection and permitting automatization of MODS interpretation.

Introduction

Tuberculosis (TB) is a contagious airborne disease caused by the bacteria Mycobacterium tuberculosis (MTB) affecting the poorest population of the world. TB is widespread and causes an important number of deaths worldwide. An estimated of 8.7 million of new cases of TB and 1.4 million deaths occurs per year [1].

In recent years, drug-resistant TB has emerged, largely due to delays in treatment, gaps in treatment protocol, and delayed drug-susceptibility testing [2]. Multi-drug resistant tuberculosis (MDR-TB) is defined by the resistance of the bacillus to isoniazid and rifampin, while extensively drug-resistant TB (XDR-TB) is defined by the resistance to any fluoroquinolone and at least one of three injectable second-line drugs (amikacin, kanamycin or capreomycin) [3], [4].

In 2011 the World Health Organization endorsed the Microscopic Observation of Drug Susceptibility (MODS) as a new phenotypic technique for TB diagnosis and drug susceptibility testing. MODS is based on the visual inspection of M. tuberculosis colonies and the identification of microscopic cording patterns in a sputum liquid culture. This characteristic phenotype is observed in an inverted microscope at magnifications between 40× and 100× [5], [6]. MODS has proven to be highly sensitive (98% sensitivity), fast in average it requires 7 days to detect a TB positive sample, and cheap (about $3–4 per sample to diagnose TB and to determine MDR-TB) [6], [8]. The incorporation of TB drugs into the culture medium allows for the determination of antimicrobial susceptibility. Besides detecting MDR-TB, MODS has recently been validated for the detection of Extreme Drug Resistant Tuberculosis (XDR-TB) [7].

The cording pattern of colonies observed between days 7–10 is specific of MTB and is the basis of MODS [6], [8]. This characteristic cording pattern has led to the development of an automated pattern recognition algorithm to facilitate automatic MODS interpretation in microscopic digital images [9].

Despite the advantages of MODS, it requires an inverted microscope, which is a relatively costly piece of equipment and a trained technician to interpret MODS cultures. To examine one sample in a 24-well plate of a MODS culture at 40X magnification, a microscopist takes approximately 1–2 min reading at least 4 fields per well.

The lens-free imager systems are able to autonomously render microscopic images of live specimens in a wide field and in a single step. This technique leverages the broad availability of high-performance sensors to provide a simple, cost-effective and automated microscopy solution [10]. The basic functioning of a lens-free imager consists in placing the specimen directly on the surface of an image sensor that is illuminated by a point source of light from the top. In this configuration, the image of the shadow is generated on the surface of the sensor, which is captured without the need of any lens or focusing procedure. This image reproduces the detail of the specimen when observed in a standard optical microscope. Notably, this process takes a single capture step to digitalize the complete surface of the sensor.

In this study, we evaluate a lens-free imager system based on a 24 mm2 CMOS sensor for digitalizing and interpreting MODS cultures.

Section snippets

Samples

Ten sputum samples were collected anonymously from the Mycobacterium Laboratory of the Faculty of Science, at the Universidad Peruana Cayetano Heredia in Lima–Peru during 2014. Five TB positive (smear positive 2+) and five TB negative sputum samples used were remnants of routine processes.

MODS culture

MODS culture was performed following the protocol previously described [5], [6], [8]. Briefly, 2 mL of sample after its decontamination and concentration by the standard NaOH–N-acetyl-l-cysteine methodology

MODS culture

Inspection of the culture plates are initiated on day 5 after incubation and are repeated every other day. TB colonies go through three visually distinct stages of growth (Figure 2). In average, in the first 4–6 days, small ‘comma’-shaped objects appear. Most characteristically, between 7 and 9 days, mycobacteria colonies show a characteristic cording pattern. After 10 days, colonies start to aggregate and form ‘conglomerates’. Both the ‘comma’ shaped colonies observed at day 4 and the

Discussion

This study confirms the proof of concept that a lens-free imager system is able to digitally capture in a single step a 24 mm2 field of a MODS culture with enough magnification and quality, appropriate for TB cords recognition and diagnostics of tuberculosis.

The MODS assay is based on the recognition of cording patterns in a liquid culture of a sputum sample. Because TB cords precipitate on the bottom of the well, this pattern is observed under an inverted microscope. The configuration of the

Acknowledgments

We are grateful with Dr. Changhuei Yang from the Electrical Engineering Department at the California Institute of Technology, for donating the complete lens-free imager system evaluated in this study, and Dr. Yingmin Wang and Dr. Benjamin Judekwitz for their assistance in setting up the lens-free imager system.

References (13)

  • WHO

    Global tuberculosis control

    (2012)
  • M.J. van der Werf et al.

    Multidrug resistance after inappropriate tuberculosis treatment: a meta-analysis

    Eur Respir J

    (2012)
  • WHO

    Treatment of tuberculosis-guidelines

    (2009)
  • M. Abdel et al.

    Epidemiology of antituberculosis drug resistance (the global project on anti-tuberuculosis drug resistance Surveillance): un updated analysis

    Lancet

    (2006)
  • WHO

    Noncommercial culture and drug-susceptibility testing methods for screening patients at risk for multidrug-resistant tuberculosis

    (2011)
  • L. Caviedes et al.

    Rapid, efficient detection and drug susceptibility testing of Mycobacterium tuberculosis in sputum by microscopic observation of broth cultures. The tuberculosis working group in Peru

    J Clin Microbiol

    (2000)
There are more references available in the full text version of this article.
View full text