Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025;46(6):36.
doi: 10.1007/s10762-025-01055-7. Epub 2025 May 30.

Robotically Controlled Terahertz Probe for In Vivo Skin Evaluation: Imaging with the PicoBot

Affiliations

Robotically Controlled Terahertz Probe for In Vivo Skin Evaluation: Imaging with the PicoBot

Jacob J Young et al. J Infrared Millim Terahertz Waves. 2025.

Abstract

In this work, we demonstrate significant modifications to our robotically controlled terahertz (THz) sensing system, the "PicoBot," enabling it to perform in vivo imaging of skin rather than limiting it to single-point measurements. By integrating a robotic arm equipped with force-sensitive feedback control, we maintain consistent contact pressure between the probe and the skin surface throughout imaging. In conjunction with this hardware advancement, we introduce an accompanying image analysis pipeline that reduces noise and enhances repeatability across scans. These improvements allow for reliable intra- and inter-subject comparisons, a critical step toward the clinical utility of THz imaging. Our ultimate aim is to use THz imaging to detect skin cancer margins: this paper highlights progress towards this goal and skin evaluation in general.

Keywords: Imaging; In vivo; Skin cancer; Skin evaluation; Terahertz.

PubMed Disclaimer

Conflict of interest statement

Competing InterestsThe authors declare no competing interests.

Figures

Fig. 1
Fig. 1
Construction of the PicoBot imaging probe
Fig. 2
Fig. 2
A diagram showing how the Menlo Terasmart is incorporated into the PicoBot probe
Fig. 3
Fig. 3
a The XY stage, the Menlo THz reflection head and quartz window with the X and Y axes shown in green lines. b Raster scan path used to acquire an image. The stage moves in a continuous motion without stopping
Fig. 4
Fig. 4
Flowchart of the PicoBot imaging protocol for an in vivo measurement of skin. The image acquisition process itself is outlined with a dashed box
Fig. 5
Fig. 5
Side profile of the PicoBot probe. The base is attached to a robotic arm via a force sensor, this ensures accurate guidance of the PicoBot to the relevant area of skin, and consistent application of a mild force so the quartz imaging window makes good contact with the skin. The 3D camera used to guide the PicoBot can also be seen in the photo. The quartz imaging window of the PicoBot probe. The PicoBot imaging the right cheek of a volunteer
Fig. 6
Fig. 6
a Example of the THz beam path through a quartz window reflected from a quartz-sample boundary. b The measured THz waveform consists of 2 pulses, E1(τ,t) and Esample(τ,t) and are shown in the time domain
Fig. 7
Fig. 7
The change in the a amplitude and b phase of a measurement of air at a signal point over 90 min, with measurements made at a rate of 10Hz. c The change in amplitude and d of the corrected measurements of air made over 90 min, calculated using Eq. 8
Fig. 8
Fig. 8
The change in the a amplitude and b phase at 0.5THz of an image of air taken with PicoBot. c The change in amplitude and d of the image formed of the corrected measurements of air, calculated using Eq. 9. The standard deviation for each of the parameters plotted is given by σ
Fig. 9
Fig. 9
Images taken of the aluminium tape stuck to the quartz window of the PicoBot as shown in a. The images show the amplitude of the normalised corrected signal calculated using Eq. 6 at b 0.1THz, c 0.2THz, d 0.5THz, and e 0.8THz
Fig. 10
Fig. 10
An example of the method used to calculate the resolution in the X direction. a An image of tape at 0.5THz with a black line indicating the profile used to find the X resolution at 0.5THz. b The amplitude at 0.55THz as a function of X along the black line. This can then be differentiated and fit with a Gaussian peak c the width of the fitted Gaussian can then be used to find the resolution of the image at 0.5THz in the X direction
Fig. 11
Fig. 11
Resolution as a function of frequency in the X and Y direction a when the PicoBot setting dX=0.1mm is used and b when dX=1 mm is used, this setting is used during in vivo measurements
Fig. 12
Fig. 12
The THz image showing the amplitude of the corrected signal |Mc|. The THz image showing the phase of the corrected signal arg[Mc]. In both images a dashed line indicates the boundary at which moisturiser was applied. A photo of the region being measured taken using an endoscopic camera mounted inside the PicoBot, moisturiser has only been applied on one side of the line as indicated. |Mc| for an unmoisturised region and a moisturised region

References

    1. Dogra, A. et al. Towards autonomous robotic thz-based in vivo skin sensing: The picobot. Scientific Reports15 (2025). - PMC - PubMed
    1. Ashworth, P. C. et al. Terahertz pulsed spectroscopy of freshly excised human breast cancer. Opt. Express17, 12444–12454 (2009). https://opg.optica.org/oe/abstract.cfm?URI=oe-17-15-12444 - PubMed
    1. Reid, C. B. et al. Terahertz pulsed imaging of freshly excised human colonic tissues. Physics in Medicine and Biology56, 4333–4353 (2011). - PubMed
    1. Hernandez-Cardoso, G. G. et al. Terahertz imaging for early screening of diabetic foot syndrome: A proof of concept. Scientific Reports7 (2017). - PMC - PubMed
    1. Lamberg, J. et al. Wavefront-modified vector beams for thz cornea spectroscopy. Optics Express31, 40293 (2023). - PubMed

LinkOut - more resources