Collimated ultrasound from an optical fibre

Add One

Ultrasound is a safe and low-cost medical imaging tool which allows us to probe inside the human body without the risks associated with ionising radiation. The concept is simple: a transducer sends a pulse of ultrasound into the tissue, where it reflects and scatters off various structures and back to the transducer. The time taken for these echoes to return depends on the depth of the structure, and so timing the returning ultrasound allows the position of the scattering structures to be inferred. This provides what is known as an ‘A-scan’ – information along a single line going into the tissue. 2D or 3D images are then formed from multiple A-scans acquired as the ultrasonic beam is scanned in one or two directions.

Unfortunately, the physics of ultrasound forces a trade-off between resolution and penetration depth, so that the highest resolution imaging isn’t possible from outside the body. This means that some applications need ultrasound probes that are small enough to be used inside the body. There’s a limit to how far high quality transducers can be miniaturised, which has encouraged research into ultrathin ultrasound probes that use fibre-optics both to generate an ultrasonic pulse and to detect the returning echoes.

An optical ultrasound detector can be made for creating a Febry-Perot interferometer at the end of a fibre. This involves two closely space reflective surfaces, arranged so that light of a specific wavelength will be reflected back due to constructive interference. As a sound wave hits the interferometer, the distance between the two reflectors changes slightly, losing constructive interference and leading to a drop in the signal. Monitoring the change in the signal over time allows detection of ultrasonic waves.

Generating an ultrasound pulse using a fibre is even easier. If certain types of absorbing material are placed at the end of a fibre, and a high intensity, short pulse of light is sent along it, thermal effects result in the generation of an acoustic pulse. However, unlike in a transducer, this pulse spreads out in all directions rather than being focused.

The drawback of using a fibre ultrasound probe is that this gives us only a single measurement. To reconstruct a 2D cross-sectional image (known as a B-scan), the fibres have to be scanned in 1 or 2 dimensions. It’s important to realise that each scan position doesn’t produce an A-scan. The ultrasound from the fibre spreads out in all directions and is reflected in all directions from the tissue. The only way to reconstruct an image is by shifting the fibres around, collecting the signal from each position, and then performing a mathematical reconstruction (similar to back-projection used in CT).

Crucially, because each point in the image depends on a large number of measurements going into the reconstruction algorithm, imprecision in the probe scanning (so that the true position of the measurement is unknown), or motion of the patient during the scan, results in severe artefacts. Given that the whole point of using a fibre is to make a small diameter probe, it’s unlikely that a very accurate scanning mechanism could be incorporated, limiting the real clinical applications of these systems.

A paper published by Aless et al.1 proposes a potential solution to this problem. The team designed a fibre probe that generates a collimated ‘pencil-beam’ of acoustic energy. This was possible by shaping the absorbing surface on the fibre to form something like an acoustic lens. So now, firing this fibre generates an independent A-scan along the collimated beam. The resolution is limited to the width of the beam – around 2 mm – which is fairly poor. However, unlike in the previous scheme, where each spatial position in the image depends on every positional measurement in some complex way, now each A-scan is captured independently, at a specific instance in time. This makes the whole thing much less susceptible to motion artefacts or inaccuracies in the 2D scanning.

The collimated fibre begins to outperform the non-collimated system for fibre positioning errors as small as 30 microns and produces good quality images from tissue. With the tolerances for fibre positioning accurate relaxed, there are now several scanning approaches that might become feasible, including the rotating fibre catheter-style scanner used in optical coherence tomography probes. This work therefore takes fibre-based ultrasound probes one-step closer to the clinical practice.


  1. Alles, Erwin J., et al. “Pencil beam all-optical ultrasound imaging.” Biomedical Optics Express 7.9 (2016): 3696-3704.

Leave a Comment

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>