Fig 1.
The ESPressoscope concept (left) consists of an ESP32 microcontroller development board with an embedded graphical user interface and an integrated camera which is combined with other modules which are selected depending on imaging configuration and which are combined through a layered structure. This paper demonstrates the concept with a prototype set of modules (center, refer to S1 Fig) which we combined in various ways to achieve a variety of prototype optical configurations, such as a compact general-purpose microscope (Matchboxscope), an underwater microscope (Anglerfish), a flow-imaging microscope with an embedded fluidic device (ESPlanktoscope), a spectrophotometer (ESPectrophotometer), and a lensless holographic microscope (HoloESP).
Fig 2.
(a) A Matchboxscope configuration featuring periscopic illumination and spring-based focusing mechanism. (b) This simple microscope can resolve features as small as 4–5μm inside the USAF chart (group 6,7) also demonstrated with (c) the lineplot along the vertical and horizontal direction (red, blue). Examples of micrographs obtained with the Matchboxscope: (d, e) A mosquito larvae found in a pond and (f) red blood cells.
Fig 3.
The Anglerfish: a submersible and automatic microscope for underwater operation, built using (a) a Matchboxscope with additional 3D-printed components, and (b) by combining an ESP32 camera module together with modules from the UC2 microscopy toolkit. With deployment in a pond (c) and application of flat-field correction to acquired images, the microscope allows observation of initial biofilm formation and microbial behaviors, including (d) a Paramecium crossing the field-of-view. (e) A variance projection of the timelapse visualizes the movement of different particles over time, with yellow arrows indicating the movement of the Paramecium at different timepoints and the blue arrow indicating the movement of a smaller microorganism. The optical path is visualized as a ray diagram, which also indicates the resulting magnification (a represents the distance from object to the lens, a′ the distance between the lens and the camera, d the overall distance between sensor and sample d = −a + a′).
Fig 4.
The ESPlanktoscope combines the standard Matchboxscope (top) with an additional ESP32 board for driving a peristaltic pump. Being a Matchboxscope extension, all possible Matchboxscope modifications explained before (different light sources and focusing capabilities) can also be applied to the ESPlanktoscope.
Fig 5.
(a) Adapting the Matchboxscope into a spectrophotometer the objective lens to be mounted on the camera and a 3D printing nozzle to be added as a pinhole. (b) An adjacent grating diffracts the light before it gets focused on the sensor. (c) The resulting image on the camera sensor can be converted into a spectrum by drawing a d.) line plot through the intensity signal along the first diffraction order. e) A browser-based tool derived from the [34] receives frames from the ESP32 camera and draws the lineplot. For quantitative pixel/wavelength measurements, the ESPectrophotometer would require calibration.
Fig 6.
(a) Inline holographic microscope, with an orange arm holding an LED whose output is spatially filtered by an FDM 3D printer’s hotend nozzle. (b) The point source of light sends out spherical waves so that scattered and unscattered waves interfere on the sensor. Using the Fresnel transform, the raw digital hologram (c, d) can be numerically refocused as in (e).
Fig 7.
a) The light weight and small size of the minimal Matchboxscope configuration make it suitable in remote locations for on-site measurements, as already demonstrated by various users under the hashtag “#matchboxscope” on Twitter/X. (b) The Matchboxscope can easily be mass-produced to deploy it in different settings such as field work (d) or STEM education (c). (e) The recent Seeed Studio XIAO Sense ESP32 camera board makes the device even more compact, as presented at the 2023 Eindhoven Maker Faire.