henhao
Sign in
Cell BiologybioRxiv
Preprint

Adapting the OpenFlexure Microscope for Affordable Live-Cell Imaging

Jodie R. MalcolmiD, Olympia Physouni, Lola K Kelly, Juliana B. T. CarnielliiD, Sandy J. MacdonaldiD, Mark Bentley, Stephen P Howarth, Stuart LacyiD, Alastair DroopiD, Matthew ReillyiD, James P J Chong, Ana Paula C A Lima, Jeremy Charles MottramiD, L. F. WigginsiD, William J. BrackenburyiD, Peter J O’TooleiD

Abstract

Live-cell imaging (LCI) captures dynamic cellular behaviours missed in fixed samples. Researchers in resource-limited settings are underrepresented in the adoption of LCI due to the prohibitive cost of dedicated systems. We adapted the 3D-printed OpenFlexure Microscope (OFM) for LCI in humid tissue culture incubators. We relocated electronics to prevent corrosion, and printed in acrylonitrile styrene acrylate (ASA) to improve image stability in prolonged experiments. Current-limiting resistors reduced heat buildup inside sealed incubators, while our simple graphical user interface (GUI)-based application enabled rapid setup of timelapse experiments. Using a 48-hour docetaxel treatment of breast cancer cells, we validated how the LCI-OFM seamlessly fits into established bioimaging pipelines to generate biologically meaningful data. We further demonstrated broad applicability of the LCI-OFM by imaging Leishmania-infected macrophages in a containment level 3 laboratory and gut fungi in an anaerobic environment. This accessible platform expands opportunities for resource-limited researchers to study locally-relevant health challenges.

PDFDOICode
Full-text digest ↓

02 /Specification & Classification

Click any tag to browse matching papers
Citations
0
OpenAlex Topics

In-depth summary

✦ Multi-agent verified full-text synthesis

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 2 (17 Jul 2026) · CC BY 4.0

A 3D-printed OpenFlexure Microscope adapted for humid incubators, with relocated electronics, ASA printing and 22 Ω current-limiting resistors, records 48-hour label-free docetaxel responses that CellPhe can analyse.

Dedicated live-cell imaging systems cost too much for many researchers in low- and middle-income countries, and the low-cost OpenFlexure Microscope (OFM) could not run reliably for long periods inside humid, temperature-controlled incubators. This work adds hardware, material, thermal and software changes that close that gap and remain compatible with the original OFM design.

Key results

  1. SuggestedPrinting the microscope body in acrylonitrile styrene acrylate (ASA) gave the most stable long timelapses. Polylactic acid variants lost focus, and flame-retardant polycarbonate held its position but transmitted incubator vibration.

    24-hour timelapses (6 min interval, 40X) of MDA-MB-231 cells on bodies printed in PLA, PLA+, PC-FR or ASA: PLA and PLA+ showed substantial focal drift; PC-FR kept position but frames visibly shook; ASA showed minimal drift and fewer vibration artefacts. Judged visually from images and videos. · Printing in acrylic styrene acrylonitrile is…; Figure legends

  2. ShownPlugins that put current-limiting resistors in series with each stepper-motor winding lowered motor temperatures inside the incubator. The 22 Ω version gave the best balance between heat reduction and reliable motor response.

    Without plugins (0.68 W per winding), motors reached 71.1 °C (± 2.1 SD) at 1 h and 71.1 °C at 48 h. With 22 Ω plugins (0.17 W per energised winding), they held 49.5 °C at 1 h and 48.6 °C at 48 h. With 15 Ω plugins, 50.3 °C at 24 h. 27 Ω (0.14 W) ran similarly cool, but the motors responded poorly. The culture vessel stayed at 36.8 °C even without plugins. · Custom resistor plugins are essential for…; Figure legends

  3. ShownLabel-free timelapses captured docetaxel-induced rounding, failed cytokinesis and reduced motility in MDA-MB-231 cells. A standard segmentation, tracking and phenotyping pipeline used these images to separate treated cells from controls.

    Four biological replicates, 48 h at 6-min intervals, 1 μM docetaxel vs 0.01% DMSO. A custom Cellpose model with TrackMate and CellPhe found 88 of 1061 features that discriminated the groups, led by texture and shape. This matches features previously reported with the Phasefocus Livecyte. UMAP separated the conditions. The microscopes stayed in the incubator for 17 days total. · The modified OpenFlexure is suitable for…; Docetaxel Treatment of MDA-MB-231 Cells; Figure legends

  4. SuggestedThe microscope worked inside a containment level 3 incubator and inside an anaerobic workstation. It resolved Leishmania donovani promastigotes entering mouse macrophages, and the sporangia and hyphae of Neocallimastix gut fungi.

    BALB/c bone marrow-derived macrophages were infected at MOI 10 with late-log L. donovani promastigotes for 6 h, then imaged in place; an extracellular flagellum was visible during entry. Neocallimastix cultures in an anaerobic chamber at 39°C were imaged after settling overnight. Both are qualitative, representative images. · Broad Applicability of the LCI-OFM Across…; Imaging of Leismania-Infected BMDMs in a…; Live-Cell Imaging of Monocentric Neocallimastix; Figure legends

  5. SuggestedA lightweight offline timelapse GUI and an ImageJ preprocessing macro simplify setup and data handling. The authors also report that the system kept running from a power bank when mains power was lost.

    The LCI-GUI installs locally on Raspberry Pi OS, needs no internet after installation, sets duration and interval in h:m:s, and saves timestamped JPEGs. The macro converts RGB to 8-bit greyscale with 50% down-scaling to correct colour imbalance and vignetting. Operation from an Anker 537 (PowerCore 24K) was confirmed, but no metrics are given. · Essential modifications to the OpenFlexure…; A simple graphical user interface for timelapse…; LCI-GUI Installation

Methods, data and reproducibility

The OFM V7 hardware was re-engineered for incubators: the Raspberry Pi 4 and Sangaboard moved outside, a sealed base, a 35 mm dish insert, an O-ring hammock, an ASA chassis and resistor plugins. It was paired with an offline timelapse GUI and tested through polymer drift comparisons, infrared motor thermometry and a docetaxel phenotyping assay.

  • MDA-MB-231 breast cancer cells treated with 1 μM docetaxel or 0.01% DMSO vehicle, imaged for 48 h at 6-min intervals and analysed with Cellpose, TrackMate and CellPhe · human (MDA-MB-231 cell line) · n = N = 4 biological replicates
  • MDA-MB-231 cells imaged for 24 h on LCI-OFMs whose main body was printed in PLA, PLA+, PC-FR or ASA · human (MDA-MB-231 cell line) · n = 3.2×10^4 or 1×10^5 cells seeded per dish
  • MCF-7 and MDA-MB-231 images used to demonstrate the greyscale conversion and down-scaling macro · human (MCF-7 and MDA-MB-231 cell lines)
  • Infrared temperature readings of x, y and z stepper motors at baseline, 0, 1, 24 and 48 h, with no plugin or with 15, 22 or 27 Ω plugins
  • BALB/c bone marrow-derived macrophages infected with Leishmania donovani MHOM/ET/1967/HU3 (LV9) promastigotes, imaged in a containment level 3 lab · mouse (BALB/c); Leishmania donovani · n = 5×10^5 BMDMs per dish; MOI 10
  • Uncharacterised monocentric Neocallimastix anaerobic gut fungi grown on wheat straw, imaged inside an anaerobic chamber · Neocallimastix sp. (anaerobic gut fungus)
  • Printable 3D models for all LCI-OFM parts · 10.5281/zenodo.17532866
Methods & tools
OpenFlexure Microscope V7, Raspberry Pi 4 Model B with Sangaboard V5, Raspberry Pi Camera Module V2, LCI-GUI (custom timelapse application), ImageJ macro for greyscale conversion, Cellpose v3.1.1.2 (custom model), TrackMate, CellPhe toolkit, UMAP, Hikmicro Eco handheld infrared camera, Autodesk Inventor, Bambu Lab X1 and P1S 3D printers
Code
https://doi.org/10.5281/zenodo.21379245https://github.com/uoy-research/cellphe-pipeline-imageshttps://build.openflexure.org/
Software & environment
Cellpose v3.1.1.2 for the custom segmentation model; Autodesk Inventor for 3D modelling; Bambu Lab X1 and P1S printers with standard 0.2 mm slicer settings for PLA or ASA; LCI-GUI installed locally on Raspberry Pi OS and usable offline.
Experimental conditions
Acquisition and GUI run on a Raspberry Pi 4 Model B with Sangaboard V5, housed outside the incubator; also tested on an Anker 537 (PowerCore 24K) power bank. No hardware or runtimes are reported for the Cellpose, TrackMate or CellPhe analyses.
Reproducibility gaps
No deposit of the raw timelapse images is reported. The ImageJ macro and adaptation details sit in supplementary files without listed URLs. The Cellpose model link is split by a space in the extracted text, so its exact path should be checked against the source.
Limitations (4)
  • ASA gave the most stable timelapses but may cost too much for tight budgets. Where only cheaper filaments such as PLA are available, the authors suggest reinforcing PLA (acrylic spray, thicker walls or infill) or using PETG; neither option was tested. · stated by the authors
  • The LCI-GUI has no autofocus, tiling or stitching, or multi-ROI sampling. It also lacks colour-channel correction, so raw images show uneven colour and vignetting that need a post-hoc ImageJ greyscale conversion. · stated by the authors
  • Polymer stability and focal drift were judged visually from one timelapse per material, with no quantitative drift or vibration metric and no replication. The ranking of ASA over PC-FR therefore rests on qualitative evidence. · noted in this summary
  • There was no same-sample comparison with a commercial live-cell imager; agreement with the Phasefocus Livecyte relies on previously reported features. The containment level 3 and anaerobic uses are shown only as representative images, without quantification. · noted in this summary

Who should read:Read this if you want a low-cost, open-source microscope that runs multi-day label-free timelapses inside humid incubators, containment level 3 labs or anaerobic chambers, producing images that Cellpose, TrackMate and CellPhe can analyse.

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISMerces '20Zehrer '23Badon '21Tepperman '20Collins '19Tristan-Lan '19
76% semantic matchCell Biology
G. O. T. Merces, C. Kennedy, B. Lenoci+3 more·bioRxiv·May 28, 2020