In January 2026, a team at IHU Méditerranée Infection in Marseille — one of the largest clinical infectious-disease institutes in Europe — published a review in Frontiers in Cellular and Infection Microbiology with a title that doubles as a thesis: “The return of a forgotten ally: tabletop scanning electron microscopy in the realm of bacteriology.” It is worth reading in full if you run a microbiology, clinical, or life-science lab and have ever wondered whether a benchtop SEM is a serious instrument or a teaching toy. The authors’ answer, backed by a decade of their own and others’ published work, is unambiguous — and it comes from people who use these instruments on real patient samples.
Here’s what the review documents, why it matters, and how we read it.
Why bacteriology forgot about the electron microscope
Electron microscopy and bacteriology grew up together; many of the first images of bacterial ultrastructure came from EM in the mid-twentieth century. Then, as the review recounts, EM “saw a decline in routine and clinical microbiological applications in the late 20th century.” The reasons will sound familiar to anyone who has worked around a conventional floor-standing SEM: extensive and sometimes hazardous sample preparation, high vacuum that punishes biological specimens, instruments that were “large, difficult to maintain, and required stable specialized environments,” and the rise of molecular methods that were faster and more sensitive for routine identification. The microscope moved to the core facility, and the microbiologist stopped thinking of it as a bench tool.
What changed, the authors argue, is the arrival of tabletop SEMs in the early 2000s — “compact, user-friendly instruments that could operate in standard laboratory environments.” Simpler prep, faster pumpdown, low-vacuum modes for delicate or uncoated specimens, and a footprint that fits next to the incubator rather than in a basement. The technique didn’t get better; it got reachable.
What the review says tabletop SEM can do in a clinical lab
Identify bacteria from a positive blood culture in about an hour
The headline result: working directly from positive blood cultures, SEM-based morphological classification reached genus-level identification with “96.6% accuracy,” with “turnaround time reduced to approximately one hour for twelve blood cultures.” The authors position this against MALDI-TOF MS, the current reference standard, and against Gram staining, over which SEM offered “superior morphological detail” while avoiding “inter-operator variability and false staining issues.” In oral samples, spiral-shaped Treponema were detected in a workflow “completed in approximately 15 minutes” including preparation.
Read antibiotic susceptibility from cell shape in one to two hours
This is the part that should make clinicians sit up. Standard culture-based susceptibility testing needs 16 hours or more. The review describes SEM-based approaches that watch bacterial morphology respond to an antibiotic:
- Gram-negative isolates exposed to imipenem: susceptible strains showed “swelling, elongation, and increased electron density”; resistant strains “maintained their baseline morphology.” A blind test of 58 clinical samples achieved “100% correct classification” within two hours.
- Gram-positive cocci and vancomycin: susceptible cells showed “reduced septation and increased cell size,” detectable “within an hour of antibiotic contact.”
- Colistin: susceptible strains showed inflation, fusion, and lysis with “a significant decrease in viable cells starting as early as 30 minutes,” versus “minimal changes even after 120 minutes” in resistant strains.
The same group showed that bactericidal and bacteriostatic activity could be distinguished within two hours — an answer that otherwise waits overnight. A phosphotungstic-acid staining approach, read through backscattered-electron intensity, let them tell live cells from dead ones across a wide range of Gram-positive and Gram-negative species.
See what culture and sequencing miss
Beyond identification and susceptibility, the review collects applications that play to SEM’s strengths as a direct observation technique: biofilm architecture on implants and in model systems (collapsed structures after treatment, adhesion and maturation stages, protective aggregates in synovial fluid); infective-endocarditis vegetations, where SEM plus EDS revealed species-specific bacterial distributions and “calcium-phosphate deposits resembling hydroxyapatite”; more than 200 urine samples imaged with EDS identifying crystals (“calcium oxalate was the most frequently detected crystal, followed by struvite”) alongside morphological classification of bacteria, yeasts, and host cells; gut-microbiome work that captured “over 40,000 micrographs from 40 stool samples”; and vaccine-development QC confirming microparticle integrity and size (mostly 1–5 µm).
The limitations the authors are honest about
A review that only cheered would be less persuasive. This one lists what still needs work: high-throughput image acquisition and shorter scan times for high-resolution images; “robust methodologies for imaging hydrated and partially hydrated specimens”; systematic evaluation “across diverse bacterial species, life-cycle stages, and ecological niches”; automated sample-preparation workflows; and, above all, image analysis, which they call a major bottleneck awaiting dedicated machine-learning models. Tabletop SEM, in their framing, must still “compete with other imaging techniques” on speed for routine diagnostics. Those are fair caveats — and they are engineering and workflow problems, not physics problems.
How we read it: category validation, from people with no instrument to sell
The studies in this review were run on tabletop instruments from Hitachi, JEOL, and Thermo Fisher — not on ours. We think that makes the paper more useful to a prospective buyer, not less. It is independent, clinical, peer-reviewed evidence that the benchtop category does serious microbiology: sample in, answer out, in a standard lab, in roughly the time a Gram stain used to take.
The review characterizes the category as offering roughly 10–20 nm resolution, magnification to about 100,000×, and 5–30 kV operation. For context, the SNE-Alpha specifies 5 nm resolution at 30 kV (secondary electrons) with magnification to 250,000×, and runs from 1 kV — the low end matters for beam-sensitive biological material, where gentler electrons mean less damage and more surface detail. Backscattered imaging comes standard, which is the signal the viability-staining work relies on; low-vacuum mode allows quick looks at uncoated specimens; and a factory-ready EDS port covers the crystal-identification and mineral-deposit work the review describes in urinalysis and endocarditis. Our life-sciences applications page covers preparation approaches for cells and tissues in more depth.
Resolution headroom is not the main point, though. The main point is the one the Marseille group makes: the microscope is back on the bench, and the bench is where the samples are.
Frequently asked questions
Can a benchtop SEM really image bacteria well enough to be useful?
Yes. Typical bacteria are 0.5–5 µm; resolving cell shape, septa, surface texture, and size changes is well within benchtop capability, as the review’s identification and susceptibility results show. Ultrastructure inside the cell remains the domain of TEM.
Do biological samples need coating?
For the best high-vacuum images, a thin conductive coat is standard after fixation and drying. Low-vacuum mode allows rapid uncoated observation when speed matters more than ultimate resolution, and the review notes that simpler preparation is a core reason tabletop SEM fits clinical workflows.
Is this only for hospitals?
No. The same direct-observation advantages apply to food and water microbiology, biofilm and antimicrobial research, probiotic and fermentation QC, and microbiome studies — anywhere a fast, image-based look at the organisms themselves answers a question that culture or sequencing answers slowly or indirectly.
Curious what your own samples look like?
If you’re evaluating a benchtop SEM for a microbiology or life-science lab, send us representative samples — cultures, biofilm coupons, filters — and we’ll image them live on an SNE-Alpha and talk through preparation, throughput, and what the instrument can and can’t do for your workflow. Get in touch.
Source: Zmerli O, Boukili M, Bellali S, Bou Khalil J. The return of a forgotten ally: tabletop scanning electron microscopy in the realm of bacteriology. Front Cell Infect Microbiol. 2026;15:1697696. doi:10.3389/fcimb.2025.1697696. Open access under a CC BY license; quotations are from the published article.