We demonstrate a lensfree dual-mode holographic microscope that can image specimens in both transmission and reflection geometries using in-line transmission and off-axis reflection holography, respectively. This field-portable dual-mode holographic microscope has a weight of ~200 g with dimensions of 15 x 5.5 x 5cm, where a laser source is powered by two batteries. Based on digital in-line holography, our transmission microscope achieves a sub-pixel lateral resolution of ≤2 µm over a wide field-of-view (FOV) of ~24 mm(2) due to its unit fringe magnification geometry. Despite its simplicity and ease of operation, in-line transmission geometry is not suitable to image dense or connected objects such as tissue slides since the reference beam gets distorted causing severe aberrations in reconstruction of such objects. To mitigate this challenge, on the same cost-effective and field-portable assembly we built a lensless reflection mode microscope based on digital off-axis holography where a beam-splitter is used to interfere a tilted reference wave with the reflected light from the object surface, creating an off-axis hologram of the specimens on a CMOS sensor-chip. As a result of the reduced space-bandwidth product of the off-axis geometry compared to its in-line counterpart, the imaging FOV of our reflection mode is reduced to ~9 mm(2), while still achieving a similar sub-pixel resolution of ≤2 µm. We tested the performance of this compact dual-mode microscopy unit by imaging a US-air force resolution test target, various micro-particles as well as a histopathology slide corresponding to skin tissue. Due to its compact, cost-effective, and lightweight design, this dual-mode lensless holographic microscope might especially be useful for field-use or for conducting microscopic analysis in resource-poor settings.

In this work we investigate whether the innate visual recognition and learning capabilities of untrained humans can be used in conducting reliable microscopic analysis of biomedical samples toward diagnosis. For this purpose, we designed entertaining digital games that are interfaced with artificial learning and processing back-ends to demonstrate that in the case of binary medical diagnostics decisions (e.g., infected vs. uninfected), with the use of crowd-sourced games it is possible to approach the accuracy of medical experts in making such diagnoses. Specifically, using non-expert gamers we report diagnosis of malaria infected red blood cells with an accuracy that is within 1.25% of the diagnostics decisions made by a trained medical professional.