TECHNICAL WHITE PAPER
Water-assisted multi-frequency bioimpedance imaging for oncology research
A scientific and development framework for AquaBIT: a proposed non-ionising platform designed to investigate tissue electrical properties and their potential value in longitudinal cancer assessment.
Web edition
Version 1.0 · August 2026
Investigational concept
Not approved for clinical use
Executive summary
AquaBIT is an early-stage medical technology concept that combines multi-frequency bioimpedance measurements, a controlled water interface and computational reconstruction. Its oncology-first research hypothesis is that repeatable maps of electrical properties may provide complementary information about tissue state and change over time, without ionising radiation.
The scientific basis is established: biological tissues exhibit frequency-dependent conductivity and permittivity, and electrical impedance tomography has been studied for biomedical imaging. What remains unproven is whether the proposed AquaBIT architecture can deliver measurements that are sufficiently stable, spatially informative and clinically meaningful for its intended use. This white paper therefore separates physical principles from device-specific hypotheses and describes the evidence required to test them.
1. The clinical and technical problem
Cancer care often depends on repeated assessment across diagnosis, treatment and surveillance. Established modalities—including CT, MRI, ultrasound and nuclear medicine—are indispensable, but each has different constraints involving radiation, infrastructure, cost, access, operator dependence or acquisition time.
AquaBIT is not intended to replace these modalities. The development question is narrower: could a low-energy, repeatable electrical measurement provide useful complementary information between or alongside standard investigations? Any future clinical claim must be defined by intended use and supported by prospective evidence.
2. Scientific foundation
Cells, extracellular fluid and membranes influence how alternating electrical currents pass through tissue. Conductivity and permittivity vary with tissue composition and measurement frequency. These are established biophysical properties, but they are not disease-specific: inflammation, oedema, fibrosis, necrosis, motion, temperature and anatomy may all affect a signal.
Electrical impedance tomography applies small alternating currents through electrodes and estimates an internal conductivity distribution from boundary voltages. The inverse problem is mathematically ill-posed, meaning reconstruction depends on measurement quality, geometry, modelling assumptions and regularisation. AquaBIT’s research programme must quantify these dependencies rather than treat reconstruction as a direct photograph of tissue.
3. The AquaBIT concept—and why water
AquaBIT proposes a person-centred measurement environment in which the body is partially immersed while the patient’s head remains fully above water. The water is intended to act as a controllable electrical coupling medium between the body and a surrounding measurement array.
Controlled coupling
Water may reduce variability associated with many direct skin-electrode contacts and support repeatable array geometry. Conductivity, temperature and water quality would require active monitoring.
Multi-angle sampling
A surrounding array could acquire boundary measurements from multiple directions and frequencies, creating a richer dataset for reconstruction and longitudinal comparison.
Research repeatability
A standardised protocol may enable within-person comparison over time. This is a hypothesis to validate through phantom, volunteer and clinical feasibility studies.
4. Proposed system architecture
- Patient interface: an accessible, supervised immersion environment with the head fully above water, controlled entry and exit, and defined positioning.
- Measurement array: distributed electrodes designed to inject low-amplitude alternating currents and record boundary voltages across selected frequencies.
- Acquisition electronics: calibrated current sources, low-noise sensing, switching, isolation and continuous system-health checks.
- Environmental monitoring: water conductivity, temperature and quality measurements incorporated into acquisition controls and signal interpretation.
- Computational reconstruction: forward models and regularised inverse methods used to estimate spatial electrical-property distributions and uncertainty.
- Research interface: protocol guidance, quality indicators, longitudinal comparison tools and audit-ready data provenance.
5. Measurement and analysis pipeline
Protocol → calibration → multi-frequency acquisition → signal quality control → physics-based reconstruction → feature extraction → longitudinal comparison → research interpretation.
Every stage should preserve provenance: device configuration, calibration state, environmental variables, raw measurements, processing version and operator actions. Quality thresholds should be specified before analysis, with failed or uncertain acquisitions identified rather than silently corrected.
6. AI: bounded and auditable
Machine learning could support artefact detection, reconstruction acceleration, segmentation or longitudinal feature analysis. It should not be positioned as an autonomous diagnostic layer.
Any model would require a defined purpose, representative data, locked evaluation datasets, subgroup analysis, version control, drift monitoring and human oversight. Performance must be reported against clinically relevant comparators, not only internal accuracy metrics.
7. Oncology-first research hypothesis
The initial hypothesis is that frequency-dependent electrical-property patterns, measured consistently over time, may reflect tissue and treatment-related change. Potential research endpoints could include repeatability, localisation, correlation with standard imaging, association with pathology and sensitivity to longitudinal change.
These are research questions—not claims of tumour detection, staging or treatment response. Indications and endpoints must be prioritised with oncologists, radiologists, physicists, patients and regulators.
8. Safety and human factors
The proposed configuration keeps the patient’s head fully above water. Safety engineering would need to address electrical isolation, current limits, emergency stop and drainage, supervised access, falls, infection control, water quality, thermal comfort, anxiety, mobility limitations and contraindications.
Human-factors work should begin before clinical studies, including co-design with patients and staff. The system must accommodate diverse body sizes and abilities, provide clear communication throughout an examination, and enable immediate assisted exit.
9. Evidence and development programme
01 · Computational
Forward models, sensitivity analysis, uncertainty and simulation benchmarks.
02 · Phantom
Calibration, spatial sensitivity, repeatability and controlled perturbations.
03 · Prototype
Engineering verification, electrical safety and human-factors evaluation.
04 · Clinical feasibility
Protocol performance, tolerability and exploratory correlation.
05 · Validation
Prospective, multicentre testing against predefined endpoints.
The intended quality and risk framework is expected to consider ISO 13485, ISO 14971 and relevant IEC 60601 standards as the product definition matures. This statement describes a development direction; it does not claim certification or regulatory clearance.
10. Limitations and falsifiable questions
- Can the system achieve repeatable measurements across sessions, sites and operators?
- How do motion, anatomy, water conditions and positioning affect the signal?
- What spatial information is recoverable at clinically practical acquisition times?
- Are observed changes specific enough to add value alongside standard care?
- Which patient groups can use the system safely and comfortably?
- Does performance generalise across representative populations?
11. Conclusion
AquaBIT brings together established bioimpedance science and a new water-assisted measurement architecture. Its promise depends on disciplined engineering and transparent evidence—not analogy alone.
The immediate objective is to test whether controlled coupling, multi-frequency acquisition and rigorous reconstruction can produce repeatable, interpretable measurements. Only prospective validation can determine whether those measurements have clinical utility in oncology.
Selected scientific references
- Gabriel S, Lau RW, Gabriel C. The dielectric properties of biological tissues: I. Literature survey. Phys Med Biol. 1996.
- Gabriel C, Gabriel S, Corthout E. The dielectric properties of biological tissues: II. Measurements in the frequency range 10 Hz to 20 GHz. Phys Med Biol. 1996.
- Gabriel S, Lau RW, Gabriel C. The dielectric properties of biological tissues: III. Parametric models for the dielectric spectrum of tissues. Phys Med Biol. 1996.
- Holder DS, ed. Electrical impedance tomography: methods, history and applications. Institute of Physics Publishing.
- Bayford RH. Bioimpedance tomography (electrical impedance tomography). Annu Rev Biomed Eng. 2006.
- Borcea L. Electrical impedance tomography. Inverse Problems. 2002.
Help test the questions that matter
We welcome dialogue with clinical, engineering, academic and industry collaborators interested in rigorous feasibility and validation work.
Important: AquaBIT is an investigational concept under development. It is not approved or cleared for diagnosis, treatment, treatment selection or patient management. Information on this page is provided for scientific and collaboration purposes and may change as evidence develops.
