Elegant: A point-of-care microscope for TB diagnosis

Brian McCall, Randall J. Olsen, Nicole J. Nelles, Dawn L. Williams, Kevin Jackson, Rebecca Richards-Kortum, Edward A. Graviss, and Tomasz S. Tkaczyk (2014) Evaluation of a Miniature Microscope Objective Designed for Fluorescence Array Microscopy Detection of Mycobacterium tuberculosis. Archives of Pathology & Laboratory Medicine: March 2014, Vol. 138, No. 3, pp. 379-389.

ORIGINAL ARTICLES

Evaluation of a Miniature Microscope Objective Designed for Fluorescence Array Microscopy Detection of Mycobacterium tuberculosis

Brian McCall , PhD; Randall J. Olsen , MD, PhD; Nicole J. Nelles , MD; Dawn L. Williams , MD; Kevin Jackson , BS; RebeccaRichards-Kortum , PhD; Edward A. Graviss , MPH, PhD; Tomasz S. Tkaczyk , PhD

From the Departments of Bioengineering (Drs McCall, Richards-Kortum, and Tkaczyk and Mr Jackson) and Electrical Engineering (Drs Richards-Kortum and Tkaczyk), Rice University, Houston, Texas; and the Center for Molecular and Translational Infectious Disease Research, The Methodist Hospital Research Institute, Houston, Texas (Drs Olsen, Nelles, Williams, and Graviss).

Context.—A prototype miniature objective that was designed for a point-of-care diagnostic array microscope for detection of Mycobacterium tuberculosis and previously fabricated and presented in a proof of concept is evaluated for its effectiveness in detecting acid-fast bacteria.

Objective.—To evaluate the ability of the microscope to resolve submicron features and details in the image of acid-fast microorganisms stained with a fluorescent dye, and to evaluate the accuracy of clinical diagnoses made with digital images acquired with the objective.

Design.—The lens prescription data for the microscope design are presented. A test platform is built by combining parts of a standard microscope, a prototype objective, and a digital single-lens reflex camera. Counts of acid-fast bacteria made with the prototype objective are compared to counts obtained with a standard microscope over matched fields of view. Two sets of 20 smears, positive and negative, are diagnosed by 2 pathologists as sputum smear positive or sputum smear negative, using both a standard clinical microscope and the prototype objective under evaluation. The results are compared to a reference diagnosis of the same sample.

Results.—More bacteria are counted in matched fields of view in digital images taken with the prototype objective than with the standard clinical microscope. All diagnostic results are found to be highly concordant.

Conclusions.—An array microscope built with this miniature lens design will be able to detect M tuberculosis with high sensitivity and specificity.