Figure 1.
Comparison of mobile phone microscopes.
A) Left panel: Cartoon schematic of a ball lens mobile phone microscope. Red brackets indicate microscope attachment optics outside of the phone (a ball lens), and blue brackets indicate mobile phone camera optics inside the phone (a lens group and CMOS sensor). Middle panel: Image of stained cheek epithelial cells taken with a 6 mm ball lens. Right panel: Enlargement of the area indicated within the dashed line in the middle panel. B) Left panel: Cartoon schematic of a standard finite objective microscope attachment to a mobile phone, consisting of an objective and an eyepiece. Middle panel: Image of stained cheek epithelial cells taken with a 4X/0.10 NA objective and a 20X eyepiece. Right panel: Enlargement of the area indicated within the dashed line in the middle panel. Note that despite the image being in-focus at the center of the field of view, some image degradation due to field curvature is detectable at the edge of the field. C) Left panel: Cartoon schematic of the reversed lens microscope presented in this paper, with opposing identical lens groups outside the phone (red brackets) and inside the phone (blue brackets). Middle panel: Image of stained cheek epithelial cells taken with the opposed lens group setup. Right panel: Enlarged area of the area indicated within the dashed line in the middle panel. Note that despite the image being focused at the center of the field, no field curvature is detectable in the reversed lens microscope image, in contrast to the ball lens A) and standard finite objective B) microscope images.
Figure 2.
Resolution of the reversed lens microscope.
A) Ray-tracing model of a reversed mobile phone camera lens as an objective for a mobile phone microscope. Performance is predicted to be best on axis (α), falling off by >2X at the edge of the field (δ) for a 1.0 mm spacing between lenses. Optical resolution is in microns; to account for variations in sagittal and tangential point-spread at higher field angles, resolution was defined as the first-zero radius of an Airy disk chosen such that its 70% encircled energy radius is the same as that computed for the sample point via ZEMAX. Field positions for α, β, γ, and δ are 0.0, 0.7, 1.5, and 2.1 mm, respectively. B) Measurements of resolution achieved by the reversed lens microscope. The resolution measurements are based on the smallest resolvable group of a 1951 USAF resolution target imaged at different radial distances from the optic axis; asymmetric NA at high field angles (and thus field radii) results in differing sagittal and tangential resolution, as seen in c and d. The dashed line connects to enlargements of the target at the different field positions.
Figure 3.
Illumination of the reversed lens microscope.
A) Cartoon schematic of the illumination optics together with the collection optics. Red and blue brackets indicate optics outside and inside the phone, respectively. Green brackets indicate the illumination system. A single LED illuminates the sample through an illumination shaping filter (ISF, dashed line) and a diffuser (solid line). B) Methods for correcting image intensity variation caused by vignetting. Columns correspond to the method used. For each column, the top panel is an image of a blank sample showing the illumination uniformity (or lack thereof). The middle panel is a line scan of this image from corner to corner. The lower panel is the standard deviation of a 10×10 pixel box at the indicated positions. Column 1 shows the results of using an LED to directly illuminate the sample. Column 2 shows the results of adding a diffuser between the LED and the sample. Column 3 shows the results of adding illumination shaping filters between the LED and the diffuser. Column 4 shows the results of incorporating a modified form of high-dynamic-range imaging with the diffuser and illumination shaping filters. Images at multiple illumination levels are taken and combined into a single image. Parts of the sample that fall into vignetted regions on the sensor are substituted with the corresponding region of the images taken with brighter illumination levels (see Methods). Note that this image has not yet been flat fielded based on the calibration image. C) An image of a 0.05 mm spacing Ronchi ruling taken with the reversed lens microscope and the combined illumination correction methods described in B). A 10X zoom of a portion of the Ronchi ruling is shown in the upper right corner.
Figure 4.
Implementation of the reversed lens microscope.
A) Image of the US$30 reversed lens microscope with integrated LED illumination (cost excludes illumination intensity control). This proof-of-concept device features a rack and pinion x- and y-translation mechanism and a screw-based focus mechanism. The reversed mobile phone lens group is fit into a piece of 4.5 mm acrylic, and an LED light source is positioned below the lens. B) Image of a simple lens in a removable lens-mount, all of which can be constructed for nearly the cost of the lens alone. The lens is mounted in acrylic, and slides in an out of tracks that are attached to the phone by a removable adhesive. The contrast of the image area around the camera was enhanced using Photoshop to improve visibility. C) Image of a planarian, taken with illumination shaping. D) Image of a stool sample containing eggs from the soil-transmitted helminth Ascaris lumbricoides taken with illumination shaping (sample courtesy of Dr. Isaac Bogoch, Toronto General Hospital). Arrows indicate helminth eggs. Zoomed image of an Ascaris lumbricoides helminth egg is shown in the upper right. E) Zoomed image of a blood smear. Arrows indicate white blood cells.