REVIEW 4 major objections 5 minor 7 references
Rotational ultrasound and photoacoustic tomography of the human body
T0 review · 4 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash
Pith's one-line read Rotating arc-shaped ultrasound arrays can produce 3D structural and vascular images of the human body in twenty seconds.
desk verdict A genuine engineering advance in synthetic-aperture ultrasound/PAT that deserves peer review, but the in vivo resolution claim goes beyond what the evidence supports. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central mechanism is the co-axial rotation of four 256-element arc-shaped quarter-ring arrays mounted on a hemispherical bowl, combined with a single-element ultrasound transducer positioned at the arrays' intersection. The co-axial mounting keeps the virtual point source fixed during azimuthal rotation, so every synthesized detection position can be treated as hearing the same source field; 90 degrees of continuous rotation creates the synthetic hemispherical aperture. The same aperture serves as the detection geometry for PAT, and delay-and-sum plus universal back-projection reconstruction forms 3D volumes from the combined detection positions.
What would settle it
Place a point-like ultrasound scatterer at a fixed off-center position in the field of view and compare the reconstructed point-spread function when the arrays rotate through 90 degrees with the point-spread function of a static, physically complete hemispherical array of matched element density; if the width or shape of the point-spread function changes beyond the reported few-percent uniformity, the synthetic aperture assumption fails. A second decisive test is to image a phantom through a layer of skull bone and measure whether resolution and contrast degrade relative to the skull-free case.
Extended reading notes
Core claim
The central discovery is that a rotating set of arc-shaped transducer arrays can synthesize a hemispherical detection surface for both ultrasound tomography and photoacoustic tomography. In RUST mode, a single-element transducer creates a virtual point source whose location is fixed by mounting it co-axially with the rotation motor; four 256-element quarter-ring arrays rotate through 90 degrees to collect scattered ultrasound from a panoramic set of directions. Because PAT uses the same hemispherical detection geometry, the same rotating arrays detect photoacoustic signals when the acoustic source is swapped for a diffused 1064 nm laser. The paper argues that this design delivers the coverage of a dense hemispherical array at lower cost and demonstrates it on human head, breast, hand, and foot with approximately 400 µm isotropic resolution, a 10 cm diameter field of view, and 10 s acquisition per modality. The head images of a hemicraniectomy patient, claimed here as the first known 3D dual-contrast images of that kind, show scalp and cortical boundaries in RUST and cortical vasculature in PAT.
Load-bearing premise
The load-bearing premise is that while the arc-shaped arrays rotate, the acoustic field produced by the single-element source stays effectively identical at every rotation angle, so signals collected from all synthesized detection positions can be treated as hearing the same source field; tissue heterogeneity, refraction, transducer misalignment, or motor wobble would violate this and blur the reconstructed image.
Editorial extensions
If this is right
- A single system can provide whole-organ 3D structural and angiographic volumes in two 10-second acquisitions without exogenous contrast agents or ionizing radiation.
- The cost of 3D panoramic ultrasound drops relative to physical hemispherical arrays because transmission uses one transducer and the 1024 detection channels are reused for both RUST and PAT.
- Adding ultrasound to an existing PAT system requires swapping a light diffuser for a transducer rather than adding hundreds of pulsers and switches.
- The geometry is flexible: fewer arcs or slower rotation reduce cost, while higher pulse repetition rates or more arrays can push acquisition time toward about 2 seconds.
- Demonstrated target applications include hemicraniectomy patient head imaging, whole-breast structural and vascular imaging, hand imaging, and diabetic foot ulcer assessment.
Reading between the lines
- If the synthetic-aperture assumption survives tissue heterogeneity, a lower-frequency version of RUST could be tested for transcranial imaging, but the co-axial consistency assumption must first be validated through skull bone.
- A next-level test would be co-registering RUS-PAT volumes with MRI or CT angiography in the same patients to measure whether RUST can estimate tumor volume and whether PAT can grade vascular density.
- The simultaneous side-illumination RUST and PAT variant, shown only on phantoms, could remove the 20-second total acquisition time and repositioning errors if it works in vivo.
- Because PAT here measures relative hemoglobin content rather than flow, combining RUS-PAT with Doppler ultrasound could add functional flow information without requiring microbubbles.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript describes a hybrid rotational ultrasound tomography (RUST) and photoacoustic tomography (PAT) system, termed RUS-PAT. The system uses a single-element ultrasound transmitter and four rotating arc-shaped arrays to synthesize a hemispherical detection aperture, and it can be switched to PAT mode by replacing the ultrasound source with an engineered diffuser and 1064-nm laser illumination. The authors validate the system with phantom experiments and simulations, reporting an approximately 400-micrometer isotropic resolution and a 10-cm field of view, and they present in vivo images of a hemicraniectomy patient's head, healthy breasts, hands, and feet, including a patient with diabetic foot ulcers. The central technical claim is that 90 degrees of continuous rotation of the arc arrays yields a synthetic hemispherical detection aperture equivalent to a physical hemispherical array, enabling 3D panoramic RUST while sharing the detection hardware with PAT.
Significance. If the resolution and synthetic-aperture equivalence claims hold, RUS-PAT would be a meaningful engineering advance: it offers a low-channel-count, potentially cost-effective alternative to dense hemispherical arrays for both ultrasound tomography and photoacoustic computed tomography, and it demonstrates large-field-of-view dual-contrast imaging at multiple human body sites. The manuscript has notable strengths, including explicit phantom validation, quantitative resolution-uniformity and sensitivity simulations, a safety analysis against ANSI and FDA/IEC limits, comparisons with a linear-array ultrasound probe and with matrix-array and linear-array Doppler systems, and a simultaneous RUST/PAT variant. As it stands, the paper is more convincing as a proof-of-concept engineering demonstration than as a quantitative clinical validation: the in vivo images are anatomically plausible, but they are not validated against an independent gold standard, and the resolution and sensitivity analyses that support the headline numbers are performed in homogeneous or near-homogeneous simulated media.
major comments (4)
- [Results, 'Phantom validation of RUS-PAT'; Supplementary Figs. 3-4, 7] The central claim of submillimeter isotropic resolution (approximately 400 micrometers) for the in vivo images is supported only by homogeneous-medium simulations (Supplementary Figs. 3-4) and a line-target phantom (Supplementary Fig. 7); no in vivo resolution measurement or gold-standard comparison is reported. Because the abstract and Discussion state this resolution as an achieved property of the system, the manuscript should either provide an in vivo resolution estimate from a known anatomical feature or fiducial, or add phantom measurements through tissue-mimicking layers with realistic sound-speed heterogeneity, and should explicitly qualify that the in vivo resolution is not directly measured.
- [Methods, '3D RUST design and construction'; 'Image reconstruction'] The synthetic-aperture reconstruction assumes a stationary object and a homogeneous acoustic medium: the Methods state that 'we can assume a consistent acoustic field distribution in the FOV while scanning the ultrasonic arrays.' In vivo, the 10-s acquisition time is long compared with cardiac, respiratory, and tremor motion at the 400-micrometer scale, and the breast protocol acquires PAT and RUST sequentially with a mechanical mode switch (Supplementary Fig. 2); no motion measurement, gating, or correction is reported. Likewise, the delay-and-sum reconstruction uses fixed straight-ray delays, which cannot compensate for refraction and aberration in heterogeneous tissues such as scalp/brain or breast fat/glandular tissue. The authors should quantify the sensitivity of the reconstructed resolution and anatomical fidelity to representative motion amplitudes and tissue sound-speed maps, for example by adding controlled motion to a phantom and by simulating with numerical phantoms that include spatially varying sound speed.
- [Results, 'RUS-PAT implementation'] The statement that 90 degrees of rotation of four arc arrays yields a synthetic hemispherical array 'equivalent to a physical hemispherical detection array' is not substantiated by a sampling or coverage analysis. Equivalence requires that the rotated positions sample the same element locations and angular diversity as a physical hemisphere, which depends on the elevation extent of the quarter-ring arrays and on the coherent combination of data acquired at different times. The manuscript should report the angular coverage and sampling density of the synthetic aperture and demonstrate, numerically or experimentally, that the resulting point-spread function matches that of the corresponding physical hemispherical array. Without this analysis, the claimed large-FOV isotropic resolution in RUST is not fully supported.
- [Results, 'RUS-PAT of the breast in vivo'; Abstract] The abstract and introduction describe the two modalities as acquired 'quasi-simultaneously,' but the standard protocol in the in vivo studies is sequential: 10 s of PAT followed by a mechanical switch and then 10 s of RUST, for a total of 20 s per site, with the breast case explicitly involving two breath-holds. The claims of '10 s imaging time' and 'single breath-hold' apply to only one modality. The manuscript should state the total examination time and discuss the implications of the inter-modality interval for co-registration, or present the simultaneous-imaging variant (Supplementary Fig. 14) as the main protocol.
minor comments (5)
- [Supplementary Fig. 4 caption vs. main text] The Supplementary Fig. 4 caption reports maximum FWHM changes of 6.3%, 5.7%, and 6.3%, while the main text reports 6.4%, 5.8%, and 6.3%; these values should be reconciled.
- [Methods, '3D RUS-PAT'] There is a typo in the phrase 'arc-shaoed detection arrays,' which should read 'arc-shaped detection arrays.'
- [Methods, '3D RUST design and construction'] The term 'virtual point source' is used to describe the output of a spherically focused single-element transducer; a focused transducer produces a focal region rather than a true point source, so the manuscript should clarify what is meant by 'virtual point source' and how its location and size affect the assumed 'consistent acoustic field distribution.'
- [Figs. 3-6] The MAP images in Figs. 3-6 are displayed with normalized amplitude or dB scales but the colorbars lack explicit unit labels; adding unit labels and stating the log-compression threshold used for display would aid interpretability.
- [Supplementary Fig. 16] The comparison with 'state-of-the-art Doppler ultrasound' uses probes with different center frequencies (8 MHz and 15 MHz versus the RUS-PAT 2.25 MHz) and different physical contrast (blood flow versus hemoglobin content); the caption should state these differences explicitly so that the comparison is not read as a resolution-equivalence claim.
Circularity Check
No significant circularity: the RUST synthetic-aperture design is a geometric construction statement, and the resolution and FOV claims are supported by phantom measurements and external comparisons, not by fitted inputs or self-cited results.
full rationale
The paper's central claim—that rotating four arc-shaped arrays over 90 degrees synthesizes a hemispherical detection aperture—is a geometric construction statement, not a derived prediction that depends on its own conclusion. The Methods explicitly state the co-axial design keeps the virtual point source fixed and that the reconstruction assumes a consistent acoustic field in the FOV; this is an openly stated modeling assumption for delay-and-sum reconstruction, not a circular use of the target result. Its validity under tissue heterogeneity or subject motion is a legitimate correctness risk, but it is not a logical circularity. The submillimeter resolution and 10-cm FOV claims are supported by phantom experiments, numerical resolution-uniformity studies, and comparisons with a standard linear-array probe, so they are not fitted parameters renamed as predictions. Citations to the authors' prior PAT work (refs. 21, 22, 34) provide technical background for the shared PAT detection module, but the RUST concept, the hybrid integration, and the in vivo demonstrations do not reduce to those citations. No self-definitional, fitted-input, or self-citation-load-bearing step was identified, so the appropriate finding is no significant circularity.
Assumptions & free parameters
assumptions (4)
- domain assumption The focused single-element transducer produces a virtual point source that remains fixed during azimuthal rotation, giving a consistent acoustic field in the FOV.
- domain assumption Delay-and-sum and universal back-projection algorithms are valid for this synthetic hemispherical aperture under a constant-speed-of-sound assumption.
- domain assumption A 90-degree rotation of four arc arrays provides sampling equivalent to a physical hemispherical array.
- domain assumption Simulated point-source FWHM and SNR distributions are representative of in vivo imaging conditions.
Cite this review
Pith. "Pith review of Rotational ultrasound and photoacoustic tomography of the human body." pith.science (2026). https://pith.science/paper/4OXMVFE2
@misc{pith2026250416036,
author = {Pith},
title = {Pith review of: Rotational ultrasound and photoacoustic tomography of the human body},
year = {2026},
howpublished = {\url{https://pith.science/paper/4OXMVFE2}},
note = {Machine review of arXiv:2504.16036}
}
read the original abstract
Imaging the human body's morphological and angiographic information is essential for diagnosing, monitoring, and treating medical conditions. Ultrasonography performs the morphological assessment of the soft tissue based on acoustic impedance variations, whereas photoacoustic tomography (PAT) can visualize blood vessels based on intrinsic hemoglobin absorption. Three-dimensional (3D) panoramic imaging of the vasculature is generally not practical in conventional ultrasonography with limited field-of-view (FOV) probes, and PAT does not provide sufficient scattering-based soft tissue morphological contrast. Complementing each other, fast panoramic rotational ultrasound tomography (RUST) and PAT are integrated for hybrid rotational ultrasound and photoacoustic tomography (RUS-PAT), which obtains 3D ultrasound structural and PAT angiographic images of the human body quasi-simultaneously. The RUST functionality is achieved in a cost-effective manner using a single-element ultrasonic transducer for ultrasound transmission and rotating arc-shaped arrays for 3D panoramic detection. RUST is superior to conventional ultrasonography, which either has a limited FOV with a linear array or is high-cost with a hemispherical array that requires both transmission and receiving. By switching the acoustic source to a light source, the system is conveniently converted to PAT mode to acquire angiographic images in the same region. Using RUS-PAT, we have successfully imaged the human head, breast, hand, and foot with a 10 cm diameter FOV, submillimeter isotropic resolution, and 10 s imaging time for each modality. The 3D RUS-PAT is a powerful tool for high-speed, 3D, dual-contrast imaging of the human body with potential for rapid clinical translation.
Reference graph
Works this paper leans on
-
[3]
The transducer array used in our RUS-PAT system was designed with a 13 cm radius
System adaption for different target sizes. The transducer array used in our RUS-PAT system was designed with a 13 cm radius. This size was selected to accommodate a wide range of head sizes and breast sizes typically encountered in clinical practice. The 13 cm radius offers sufficient coverage for most individuals, allowing adequate imaging of the target...
-
[5]
In the case of breast imaging, the dark gap observed in the image (Fig
Image quality and artifacts. In the case of breast imaging, the dark gap observed in the image (Fig. 4a) is due to the relatively weak scattered ultrasound signals from the corresponding breast tissue. In the case of extremity imaging, we observed striation patterns on the surface of the foot in the RUST images (Fig. 6b). These patterns can be attributed ...
-
[7]
Potential for future clinical application. The current study introduces an accessible and adaptable implementation of RUS-PAT that integrates 3D ultrasound tomography with photoacoustic tomography. While the system is in an early stage, its modular design offers flexibility for future optimization toward different clinical needs. In breast imaging, the cu...
work page 2010
-
[11]
Kruizinga, P. et al. Compressive 3D ultrasound imaging using a single sensor. Science advances 3, e1701423 (2017). 12. Christiansen, T. L. et al. 3-D imaging using row–column-addressed arrays with integrated apodization—part II: Transducer fabrication and experimental results. IEEE transactions on ultrasonics, ferroelectrics, and frequency control 62, 959...
work page Pith review arXiv 2017
-
[21]
Na, S. et al. Massively parallel functional photoacoustic computed tomography of the human brain. Nat Biomed Eng 1–9 (2021) doi:10.1038/s41551-021-00735-8. 22. Lin, L. et al. High-speed three-dimensional photoacoustic computed tomography for preclinical research and clinical translation. Nat Commun 12, 882 (2021). 23. Kim, J. et al. Programmable Real-time...
-
[31]
Transcranial photoacoustic computed tomography of human brain function
Tang, Y., Qian, X., Lee, D. J., Zhou, Q. & Yao, J. From Light to Sound: Photoacoustic and Ultrasound Imaging in Fundamental Research of Alzheimer’s Disease. OBM Neurobiol 4, 10.21926/obm.neurobiol.2002056 (2020). 32. Zhang, Y. et al. Transcranial photoacoustic computed tomography of human brain function. arXiv preprint arXiv:2206.00248 (2022). 33. Karella...
work page Pith review arXiv 2020
-
[42]
Xu, M. & Wang, L. V. Universal back-projection algorithm for photoacoustic computed tomography. Phys. Rev. E 71, 016706 (2005). 43. Li, L. et al. Single-impulse panoramic photoacoustic computed tomography of small-animal whole-body dynamics at high spatiotemporal resolution. Nat Biomed Eng 1, 1–11 (2017). Data availability The data that support the findin...
work page 2005
Reviewed August 16, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.