{"id":"2048efbc-92cb-4869-a565-d74c0a104e78","arxiv_id":"2507.06108","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A miniaturized laser-driven Bessel-beam ultrasound device generates a column-shaped focal zone 152 um wide and 1.93 mm long, and activates rodent brain tissue at depth, though the transcranial advantage over Gaussian beams is only shown in simulation.","lead":"Researchers built a tiny 2.33 mm ultrasound emitter that uses a laser-generated Bessel beam to create an elongated, column-shaped focus for non-invasive brain stimulation. This lightweight design could give neuroscientists a wearable tool for precisely stimulating elongated brain volumes, though its transcranial advantage over conventional beams is so far only shown in simulation.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Central 'non-invasive intact-brain' claim rests on one unreported intact-skull mouse; all shown in vivo data used cranial windows or craniotomies.","rationale":"The reader's CONDITIONAL verdict correctly identifies the gap between the 'transcranial' framing and the experimental record. My stress-test pass confirms that the single most load-bearing assumption is that OBUS can non-invasively stimulate intact brains and outperform Gaussian beams transcranially. This assumption is not directly tested: every reported in vivo dataset was obtained through a cranial window or craniotomy, and the only intact-skull result is an unreported N=1 with lower success rate, as admitted in the Discussion. The simulation-only transcranial comparison is further undermined by the model's inability to reproduce the measured free-field focal dimensions. These are not internal logical contradictions that falsify the device—the concept is plausible and the preliminary data are suggestive—but they are exactly the kind of missing support that warrants a conditional rather than an unconditional endorsement. The concrete test proposed (intact-skull c-Fos in N≥3) would directly settle whether the central claim holds; a secondary hydrophone measurement through ex vivo skull would test the simulation-based efficiency claim. My read does not change the verdict: it remains CONDITIONAL.","tokens_in":18904,"tokens_out":4411,"duration_ms":49783,"concrete_test":"Run the Fig. 3 c-Fos protocol on at least three mice with fully intact skull (no cranial window), using the same 25% duty cycle and ~4.6 MPa peak pressure, and quantify Pearson's c-Fos/DAPI colocalization in the intended VTA at 1.5–2.5 mm depth; if significant activation is not consistently detected above control, the intact-brain non-invasive claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The Abstract and Discussion claim that OBUS can 'non-invasively activate neural circuits in intact brains,' but the reported multi-subject in vivo experiments do not test this. The Discussion explicitly states: 'the c-Fos data in Figure 3 were acquired using a cranial window rather than a fully transcranial method,' and that intact-skull c-Fos was obtained in only one mouse with lower success rate. The fMRI experiments (Methods, 'fMRI acquisition and processing') used a '5-mm diameter cranial window' with the skullcap removed, and the LFP experiments (Methods, 'Electrophysiological recording') required 'craniotomy.' Thus all three in vivo modalities—c-Fos, LFP, and BOLD—were measured through an opening in the skull; the intact-skull demonstration is an N=1 mentioned only as a lower-success-rate aside. Separately, the claimed transcranial advantage over Gaussian beams rests entirely on k-Wave simulations (Fig. 2, Table 2) with no experimental pressure measurement through skull. Confidence in that simulation is weakened by the free-field mismatch: Table 1 predicts for the 15° cone a lateral resolution of 0.33 mm and axial 4.49 mm, while the measured values are 152 μm and 1.93 mm (Fig. 1E,F)—an unexplained factor-of-two discrepancy in the same modeling pipeline used for the skull predictions. The central 'transcranial, non-invasive' claim is therefore load-bearing on data that are either absent or explicitly identified as less reliable.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a miniaturized optoacoustic device (OBUS) that generates a Bessel-beam ultrasound field for brain stimulation, with hydrophone-measured lateral and axial resolutions of 152 μm and 1.93 mm in free field. Using k-Wave simulations with an imported rat skull profile, the authors claim that OBUS outperforms a conventional Gaussian emitter in transcranial transmission efficiency and beam-shape preservation. In vivo experiments include c-Fos immunofluorescence after stimulation in mice, LFP recordings in mice, BOLD fMRI in rats, and safety estimates based on mechanical index and temperature measurements. The central claims are that OBUS achieves non-invasive transcranial stimulation of intact brains, superior transcranial beam maintenance, and precise volumetric control with 152 μm lateral resolution.","tokens_in":19219,"tokens_out":7018,"duration_ms":76518,"significance":"The device concept—an optically driven conical emitter producing a Bessel-like columnar ultrasound field in a 2.33-mm package—is novel and potentially useful for targeting elongated brain subregions. The free-field characterization is careful and includes independent hydrophone measurements, a laser-off control, and a contralateral-placement control for the LFP experiments. The simulation work is reproducible and based on an open-source toolkit. However, the load-bearing claims are not currently supported: the in vivo experiments used cranial windows or craniotomies rather than intact skulls, the transcranial advantage rests entirely on simulation without experimental validation, and the 152 μm lateral resolution is not demonstrated in tissue. If the claims were revised and the missing intact-skull data or a validated simulation pipeline were provided, this could be a significant technical contribution; in its present form the manuscript substantially overstates the evidence.","major_comments":[{"comment":"The central claim that OBUS \"non-invasively activate[s] neural circuits in intact brains\" is not supported by the reported in vivo experiments. The c-Fos experiment was performed through a cranial window (N=1), as acknowledged in the Discussion; the LFP experiments required craniotomy (Methods, \"Electrophysiological recording\"); and the fMRI experiments used a 5-mm cranial window with the skullcap removed (Methods, \"fMRI acquisition and processing\"). The only intact-skull result is an N=1 c-Fos mouse mentioned in the Discussion as having a lower success rate, but no data from that experiment are shown. The authors must either present intact-skull data or revise the abstract, title, and Discussion so that the claims match the actual experimental preparations.","section":"Abstract; Methods (In vivo stimulation on mice; Electrophysiological recording; fMRI acquisition); Discussion"},{"comment":"The claimed transcranial advantage over Gaussian beams rests entirely on k-Wave simulations (Fig. 2, Table 2), with no experimental pressure measurement after skull passage. Confidence in this simulation pipeline is weakened by an unexplained discrepancy with the free-field measurements: Table 1 predicts for the 15° cone a lateral resolution of 0.33 mm and an axial resolution of 4.49 mm, while the measured values are 152 μm and 1.93 mm (Fig. 1E,F). The skull simulations use a different central frequency (10 MHz) from the design simulations (15 MHz), and no validation of the 10-MHz simulation against the measured free-field profile is provided. Please provide an experimental through-skull measurement or a clear validation of the simulation pipeline, and discuss the factor-of-two discrepancy.","section":"Transcranial efficiency and VTA maintenance of OBUS after skull aberration; Table 1; Fig. 1E,F"},{"comment":"The 152 μm lateral resolution is not demonstrated in vivo. The Discussion states that in the c-Fos experiment the \"VTA identified by the c-Fos-positive region exhibited a lateral profile exceeding 1 mm\" at 1.2 mm depth, because the pressure was increased to engage the full OBUS beam. This means the actual lateral extent of cellular activation in tissue was greater than 1 mm, far exceeding 152 μm. Thus the paper's central claim of precise volumetric control with 152 μm lateral resolution in brain tissue is not validated by any in vivo experiment. Lateral quantification of c-Fos or spatially resolved electrophysiological mapping is needed to support the lateral-resolution claim in tissue.","section":"Discussion; Supplementary Fig. S5"},{"comment":"The LFP statistical analysis is vulnerable to selection bias: the paired t-tests were performed at 46.9 Hz, described as \"the frequency with the largest PSD increase at 4.1 MPa,\" without correction for multiple comparisons across frequency bins. With N=3 mice, the reported significance at 3.7 and 4.1 MPa rests on a post hoc choice of test frequency. The authors should either pre-specify frequency bands (e.g., beta and gamma ranges) or use cluster-based permutation testing. The qualitative spectrograms and control experiments support a modulation effect, but the p-values as reported are overoptimistic.","section":"Electrophysiological responses induced in vivo by OBUS; Fig. 4E"}],"minor_comments":[{"comment":"The relationship between Eq. (1), which defines DOF = R/tan(θ), and the axial resolution values in Table 1 should be clarified: for the 15° cone, R/tan(θ) ≈ 6.16 mm, but Table 1 reports an axial resolution of 4.49 mm. Please state whether the table entries are FWHM values of the axial pressure profile and how they relate to the DOF definition.","section":"Design of OBUS; Eq. (1); Table 1"},{"comment":"The statement that OBUS \"surpass[es] the Gaussian beam by 70%\" is ambiguous because 18.7% versus 11.0% could be read as a 70-percentage-point difference. Please report the improvement as a relative percentage (e.g., \"70% relative improvement\") and specify the normalization used for the peak intensity ratio.","section":"Transcranial efficiency and VTA maintenance of OBUS after skull aberration; Table 2"},{"comment":"The text says an additional 0.6 mm was added to the c-Fos depth to account for the ultrasound gel gap, but the Methods and Fig. 3A describe a 0.5 mm gap. Please reconcile this discrepancy and define precisely how the gel gap was measured.","section":"Elongated stimulation volume delivery in vivo with OBUS; Supplementary Fig. S7"},{"comment":"The c-Fos results section reports stereotaxic coordinates AP: -0.5, ML: 1.5 for the motor cortex, while the Methods section lists AP: -1.5, ML: 2 for the somatosensory cortex under the same c-Fos protocol. Please specify which coordinate set was actually used for the c-Fos experiments.","section":"In vivo stimulation on mice via OBUS; Elongated stimulation volume delivery in vivo with OBUS"},{"comment":"The mechanical index is estimated from the negative-to-positive peak pressure ratio of the waveform in Fig. 1D rather than from a direct measurement of rarefactional pressure at the in vivo pressure levels (up to 5 MPa peak-to-peak). Please report the actual measured negative peak pressure and state whether any derating for skull attenuation was applied.","section":"Safety evaluation of in vivo OBUS neural modulation"}],"recommendation":"major_revision","confidential_remarks":"The abstract and title overstate the evidence relative to the experimental preparations described in the Methods; the Discussion's disclosure that the c-Fos data came from a cranial window is a step in the right direction, but the abstract still claims intact-brain non-invasive stimulation. The discrepancy between Table 1 and Fig. 1E,F suggests the simulation pipeline needs careful benchmarking before the skull predictions in Table 2 can be relied upon. I would encourage the editor to require either new intact-skull data (with the N=1 result reported properly) or a substantive revision of the claims, plus an experimental or otherwise clearly validated measurement of through-skull field properties."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe thing to know: this is a genuine technical advance—a 2.33 mm, 2.1 mg optoacoustic emitter that produces a Bessel-like ultrasound column with ~150 μm lateral and ~1.9 mm axial FWHM in free field. That is a real jump in miniaturization and beam shaping for optoacoustic neuromodulation, and the engineering story (conical CS-PDMS emitter, rounded tip to avoid thermal damage, 8:1 PDMS ratio) is coherent and reproducible from the methods. The in vivo c-Fos, LFP, and fMRI readouts are internally consistent and include sensible controls: laser-off for electrical interference, contralateral placement to rule out auditory confound, and a paw-stimulation positive control for fMRI. The citation pattern is fine; the heavy self-citation to their prior SOAP work is justified because OBUS is a direct geometric variation on that line.\n\nThe soft spot is the word \"transcranial.\" The abstract and Discussion say OBUS non-invasively activates neural circuits in intact brains, but all the multi-subject in vivo measurements were made through openings in the skull: the c-Fos data in Fig. 3 were acquired with a cranial window (the Discussion says so explicitly), the LFP recordings used a craniotomy, and the rat fMRI used a 5-mm cranial window. The only intact-skull result is a single mouse mentioned as a lower-success-rate aside. The claim that OBUS beats a Gaussian beam through the skull rests entirely on k-Wave simulations with an imported rat skull; there is no experimental pressure measurement after passage through a skull. The free-field characterization also has an unexplained factor-of-two discrepancy with the design simulation (Table 1 predicts 0.33 mm lateral and 4.49 mm axial for the 15° cone; measured values are 152 μm and 1.93 mm), and since the same pipeline was used for the skull simulations, that gap weakens confidence in the simulated transcranial advantage. Some of the difference is likely explained by the rounded-tip emitter and the different central frequency (15 MHz in design simulations vs ~10.6 MHz measured), but the paper does not close that loop. One more caution: the in vivo c-Fos lateral profile exceeded 1 mm, so the 152 μm lateral resolution is a free-field property, not yet a demonstrated stimulation resolution.\n\nWho is this for? Neuromodulation engineers and optoacoustics people will get real value from the device design and characterization. The in vivo biology is preliminary. My recommendation: send it to peer review, not desk reject—there is enough novel engineering and reproducible characterization to warrant referee time. But reviewers should require either intact-skull experimental evidence (pressure measurement, or at minimum an intact-skull c-Fos with reasonable N) or a rewritten abstract that keeps \"transcranial\" out of the load-bearing claims. I would not cite it in my own work yet; I would wait for the follow-up.","headline":"Clever miniaturized optoacoustic Bessel-beam emitter with real engineering value, but the 'transcranial, non-invasive' headline claim is carried by simulation and a single unreported intact-skull mouse; all the multi-subject in vivo data were acquired through skull openings.","tokens_in":19780,"tokens_out":2946,"would_cite":false,"duration_ms":33856,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A miniature optically generated Bessel-beam ultrasound device creates a column-shaped acoustic focus and is reported to activate rodent brain circuits to depths of about 2 mm while preserving its shape through the skull.","keywords":["Bessel beam","optoacoustic","ultrasound neuromodulation","transcranial","volume of tissue activation","c-Fos","functional MRI","rodent brain"],"falsifier":"Measure the acoustic field of OBUS and of a Gaussian emitter of the same focal depth after propagation through a real rat or mouse skull using a calibrated hydrophone; if the post-skull axial and lateral FWHM changes and the peak-intensity ratio do not match the simulated 18.7% versus 11.0% efficiency and the 40.5% versus 566.7% axial change, the transcranial-advantage claim fails.","tokens_in":18706,"feed_emoji":"🧠","tokens_out":7806,"duration_ms":77536,"temperature":0.7,"pith_summary":"The paper sets out to resolve a trade-off in non-invasive brain stimulation: miniature devices tend to lose volumetric control, and high spatial resolution tends to come at the cost of transcranial penetration. It introduces a 2.33-mm, 2.1-mg optoacoustic emitter whose conical candle-soot-embedded PDMS surface turns nanosecond laser pulses into a zeroth-order Bessel beam of ultrasound, producing a column-shaped focus with a lateral full width at half maximum of 152 µm and an axial value of 1.93 mm. The central claim is that this column can stimulate elongated subregions such as ocular dominance columns to depths around 2 mm, and that it holds its shape and intensity through the skull better than a conventional Gaussian beam. Supporting evidence includes c-Fos activation to 2.2 mm depth in mouse cortex, local-field-potential increases in mice, BOLD responses in rats, and a mechanical index of 0.93 with sub-1-K temperature rise. If true, this would give researchers a wearable, non-genetic tool for mapping and modulating columnar brain circuits.","feed_headline":"A 2.1-mg ultrasound cone targets brain columns 2.2 mm deep","feed_subtitle":"A candle-soot-in-PDMS cone emits a 152-micron-wide sound column that could map columnar brain circuits without surgery.","key_machinery":"The central object is the OBUS device itself: a conical optoacoustic emitter whose curved, candle-soot-loaded PDMS surface acts as the acoustic source, launching a zeroth-order Bessel beam. The cone is the ultrasound analogue of an axicon lens, and its depth of focus is set by the relation $\\mathrm{DOF}=R/\\tan\\theta$, where $R$ is the cone radius (1.65 mm after optimization) and $\\theta$ the cone angle (15°). The optoacoustic conversion follows $p_0=\\Gamma\\mu_a F$, with the Grüneisen parameter $\\Gamma=\\beta K/(\\rho C_v)$ and bulk modulus $K=E/[3(1-2\\nu)]$, so tuning the PDMS stiffness through the base-to-curing-agent ratio (8:1 optimum) maximizes the emitted pressure. A polished rounded tip is the second key mechanism: it raises peak intensity inside the volume of tissue activation by 51.6% compared with a sharp cone, combining the elongated field of a Bessel beam with a stronger focal region.","core_discovery":"OBUS is reported as a miniaturized optoacoustic device that generates a Bessel beam of ultrasound for neuromodulation. A slightly rounded conical emitter is coated with candle soot and embedded in PDMS at an optimized 8:1 base-to-curing-agent ratio; under 2.2-ns, 1064-nm laser pulses it radiates a broad-band acoustic beam with a measured lateral FWHM of 152 µm and axial FWHM of 1.93 mm. Because the beam is a zeroth-order Bessel beam, it is non-diffracting and self-healing, which the authors argue is what preserves the elongated focus after skull passage. In acoustic-wave simulations with an imported rat skull profile, the device achieved 18.7% peak-intensity transcranial efficiency versus 11.0% for a Gaussian beam, and its axial resolution changed by 40.5% rather than 566.7%. In vivo, c-Fos staining marked activation to 2.2 mm depth in mouse cortex, LFP power rose in the 10–50 Hz band in mice, and fMRI BOLD responses appeared beneath the device in rats, with the measured mechanical index of 0.93 below the 1.9 safety threshold.","pith_inferences":["A natural next test is to place an ex vivo skull between OBUS and a hydrophone and compare the measured post-skull beam width and peak pressure with the simulated values; this would directly separate the transcranial-advantage claim from the in vivo activation results.","If the beam shape is as robust as simulated, an array of small OBUS cones, or a cone with an adjustable angle, could steer and reshape the column in three dimensions, extending the approach to arbitrary subregion geometries rather than fixed vertical columns.","The 152-µm lateral resolution comes at a 10.6-MHz center frequency; at this frequency skull attenuation is severe, so translating the device to human-scale skulls would likely require lower frequencies, where the lateral-resolution advantage will shrink—a quantitative trade the paper does not address.","The self-healing property of Bessel beams might be exploited to stimulate through heterogeneities such as blood vessels or ventricles, but that resilience is so far supported only by the skull simulation and by prior Bessel-beam physics, not by tissue measurements in this study."],"forward_implications":["Columnar subregions about 0.4–0.9 mm wide and 2–3 mm long, such as ocular dominance columns, become targetable with a single 2.33-mm emitter rather than a multielement array.","Because the device weighs 2.1 mg, the approach is compatible with wearable and freely moving animal experiments, which are difficult with rigid piezoelectric arrays.","If the skull-aberration simulations hold, OBUS could deliver a well-defined deep-brain volume of tissue activation at targets like the rat thalamus at 4.8 mm while sparing tissue along the propagation path from Gaussian-like off-target exposure.","OBUS offers a non-genetic alternative to optogenetics for circuit-mapping studies, avoiding viral delivery and long incubation while reaching depths of about 2 mm that blue-light optogenetics cannot.","Combining a Bessel-beam-shaped volume of tissue activation with fMRI readouts gives a way to map the hemodynamic footprint of a precisely shaped stimulated region directly beneath the device."],"supporting_citations":[{"why":"Provides the soft optoacoustic pad method, the 15-MHz/200%-bandwidth simulation starting point, and prior ultrahigh-precision focusing that OBUS extends to Bessel beams.","marker":"(38)"},{"why":"Demonstrates self-healing and scattering resilience of acoustic Bessel beams, the property that motivates using a Bessel beam for transcranial shape preservation.","marker":"(31)"},{"why":"Compares Bessel and Gaussian beams and is the basis for the statement that standard Bessel beams have lower localized intensity, which the rounded-tip design is meant to overcome.","marker":"(48)"},{"why":"Gives the dimensions of human ocular dominance columns, defining the elongated subregion geometry that OBUS is designed to target volumetrically.","marker":"(5)"},{"why":"Supplies the 1.9 mechanical-index safety threshold used to judge OBUS safe at a measured MI of 0.93.","marker":"(42)"},{"why":"Establishes the maximum 0.35-mm lateral resolution requirement for mouse brain mapping that guided the cone-angle choice.","marker":"(39)"},{"why":"Shows that the PDMS base-to-curing-agent ratio changes Young's modulus, which the optimization of optoacoustic conversion efficiency relies on.","marker":"(40)"}],"fun_headline_variants":["Optoacoustic Bessel beam sculpts 152-µm sound columns in brain","Mini device shoots self-healing ultrasound columns 2.2 mm deep","Soot-coated cone emits Bessel ultrasound for precision brain targeting","2.33-mm optoacoustic probe makes volumetric ultrasound columns","Bessel ultrasound passes skull better than Gaussian beam"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The claim that OBUS keeps its beam shape and transmission efficiency through the skull rests entirely on simulations with an imported rat skull profile, not on measured acoustic fields after skull passage, and the in vivo demonstrations of non-invasive activation were mostly obtained through a cranial window or craniotomy, with only one intact-skull c-Fos mouse reported and at a lower success rate.","fun_headline_variants_meta":{"raw":{"variants":["Optoacoustic Bessel beam sculpts 152-µm sound columns in brain","Mini device shoots self-healing ultrasound columns 2.2 mm deep","Soot-coated cone emits Bessel ultrasound for precision brain targeting","2.33-mm optoacoustic probe makes volumetric ultrasound columns","Bessel ultrasound passes skull better than Gaussian beam"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000677,"raw_usage":{"total_tokens":3108,"prompt_tokens":1002,"completion_tokens":2106,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":618,"completion_tokens_details":{"reasoning_tokens":2015}},"tokens_in":618,"tokens_out":2106,"duration_ms":14770,"temperature":1.0,"reasoning_tokens":2015,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T19:10:33.040262+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the acoustic field of OBUS and of a Gaussian emitter of the same focal depth after propagation through a real rat or mouse skull using a calibrated hydrophone; if the post-skull axial and lateral FWHM changes and the peak-intensity ratio do not match the simulated 18.7% versus 11.0% efficiency and the 40.5% versus 566.7% axial change, the transcranial-advantage claim fails.","supporting_citations":[],"review_version":1}