Is a higher spatial resolution always better in mass spectrometry imaging?

Higher spatial resolution is not necessarily better; it is subject to trade-offs with sampling time, ion flux and data volume. The higher the resolution, the smaller the area covered by a single point; consequently, more points must be acquired to cover the same sample, which takes longer, and the signal at each point may also be weaker. The resolution should be selected based on the scale of the structure to be analysed and the molecular signal strength, rather than blindly pursuing the highest possible resolution.

What exactly does resolution refer to

The spatial resolution of mass spectrometry imaging refers to the smallest distinguishable adjacent feature size, which is determined by the size of the probe spot or spray spot and the step distance. New Source’s MSI DPI ranges from 20 to 200 μm, whilst MSI LDPI ranges from 2 to 3 μm, covering different scales from tissue compartments to macromolecular aggregates respectively.

The Triple Cost of High Resolution

Firstly, there is the issue of time. The total imaging time is approximately equal to the number of points multiplied by the time per point. A tenfold increase in resolution results in approximately a hundredfold increase in the number of points across the area; if the time per point remains unchanged, the total imaging time also increases by approximately a hundredfold. Although LDPI offers high resolution, the high sampling rate of TOF (e.g. 30,000 Hz) can be utilised to reduce the time per point, thereby alleviating some of this conflict.

Secondly, there is the issue of signal. As the area of each point decreases, the amount of material present at that location is reduced, leading to a drop in the number of ions and a lower signal-to-noise ratio; substances with weak signals may become ‘invisible’. In such cases, more sensitive detection or post-electronisation amplification is required, though there remains a lower limit.

Thirdly, there is the issue of data volume. The surge in the number of data points places strain on storage and image reconstruction, placing higher demands on computers and workflows, whilst also increasing the cumulative risk of errors and calibration deviations.

How to choose an ‘adequate’ resolution

Consider the structure being imaged: organelles and nanomedicine carriers require sub-micrometre to several micrometre resolution (LDPI class); tissue compartments and tumour boundaries can be imaged with tens of micrometres (DPI class); whilst for large-scale distributions, a few hundred micrometres is sufficient. Selecting a resolution that is ‘just sufficient’ ensures reliable images can be obtained within a reasonable timeframe.

Can resolution be improved during post-processing?

No. Resolution is determined by the spot size and step size at the time of acquisition; post-processing interpolation merely enlarges pixels without adding true detail, and cannot make a low-resolution image clearer.

Does low resolution necessarily mean the distribution cannot be seen clearly?

Not necessarily. If the structure of interest is relatively large in scale (such as organ compartmentalisation or macroscopic layering of materials), a lower resolution may be sufficient to characterise the distribution; pursuing excessively high resolution may be a waste of time and result in weaker signals.

What else should be considered when selecting a resolution?

You should also consider the molecular signal intensity, sample stability (whether it can withstand prolonged scanning), and the actual size of the target structure. These three factors collectively determine a ‘sufficient’ resolution, rather than simply aiming for the lowest possible numerical value.