Biomedical Imaging

Biomedical Imaging

Terahertz biomedical imaging investigates how tissues, cells, biological liquids and carefully designed phantoms interact with radiation between the microwave and infrared domains. Water and other polar constituents strongly influence the dielectric response in this frequency range, creating measurable contrast while also limiting penetration. The work presented here is exploratory research on physical and morphological contrast. It does not describe a clinically validated diagnostic method, an approved medical device or a replacement for histopathology.

Studies associated with Jean-Paul Guillet and collaborating teams span time-domain spectroscopy, measurements in the 300-600 GHz range, scanning-point terahertz source microscopy and silicon-oriented near-field sensor concepts. The samples and spatial scales differ substantially across these experiments. Freshly excised tissue, prepared histological sections, cell-density variations, biological liquids and phantoms cannot be treated as equivalent evidence, even when they produce a detectable terahertz response.

Terahertz contrast in tissue and cellular samples

Biomedical measurements begin by identifying which physical quantity might carry useful information. Depending on the protocol, this may be a change in complex refractive index, absorption, reflected amplitude, phase, local field response or morphology. Hydration, sample thickness, temperature, preparation, time after excision and the distance between a near-field probe and the specimen can all influence the result. A change in terahertz signal may correlate with water content, cellular density or tissue structure without being specific to a single pathology.

Near-field microscopy addresses spatial resolution by bringing the source or probe into a tightly controlled local interaction with the sample. Published studies on unstained ductal carcinoma in situ and breast-cancer cell-density variations reported micrometric observations under their respective experimental conditions. Those results should be read as demonstrations on prepared ex vivo material, not as the general resolution of all terahertz imaging systems. Far-field THz-TDS and sub-terahertz measurements answer different questions by providing broadband or spectroscopic information over larger sample regions.

Experimental design, references and scientific limits

A credible biomedical study requires a defined sample cohort, controlled handling and an independent reference such as histology, optical microscopy or a known phantom composition. Repeatability must be assessed across positions and samples rather than inferred from a single high-contrast image. Small cohorts, tissue heterogeneity, preparation artefacts and instrument drift can all produce apparent differences that do not generalise beyond the study.

Claims therefore remain tied to the population, preparation and measurement geometry that were tested. Terms such as “contrast”, “differentiation under the study conditions” and “correlation with histology” are more accurate than an unqualified claim of cancer detection. Work on blood plasma, diabetic samples or tumour-related tissue should similarly distinguish a physical response from disease-specific classification. Future sensor integration may improve repeatability or access, but clinical readiness would require substantially broader validation, safety assessment and comparison with established diagnostic pathways.

Related biomedical terahertz research

  • Terahertz near-field microscopy of ductal carcinoma in situ (DCIS) of the breast — DOI
  • Label-Free Observation of Micrometric Inhomogeneity of Human Breast Cancer Cell Density Using Terahertz Near-Field Microscopy — DOI
  • Pilot study of freshly excised breast tissue response in the 300-600 GHz range — DOI
  • Terahertz refractive index-based morphological dilation for breast carcinoma delineation — DOI
  • NearSense: advances towards a silicon-based terahertz near-field imaging sensor for ex vivo breast tumour identification — DOI

Continue with the terahertz imaging expertise, the principles of THz time-domain spectroscopy, or the historical NearSense project.

Discuss an exploratory research question

Biomedical enquiries should identify the sample type, preparation, reference method, expected spatial scale and research purpose. Review collaboration pathways or use the contact page; do not include confidential or identifiable health data in an initial message.