{"id":"dc64e137-1f3e-4696-8536-4bd36936e66d","arxiv_id":"1908.08533","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"The i-TED demonstrator, a Compton imager with a motorized scatter-to-absorber distance, achieves 8 to 14 degree angular resolution for 662 keV gamma rays and a detection efficiency about 50 times higher than comparable scintillator-based Compton cameras.","lead":"This paper introduces the first working version of i-TED, a gamma-ray camera that combines neutron time-of-flight with Compton imaging to measure neutron capture reactions. It reports the camera's measured image sharpness and efficiency, and shows that an adjustable detector spacing can trade one for the other.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The factor-of-50 efficiency claim rests on an unstated event selection and an extrapolation from 165 mm to 30 mm; without a full-energy/single-interaction cut the comparison with other cameras is not yet meaningful.","rationale":"The central claim of the paper is that i-TED works as a high-efficiency Compton imager with dynamic control of the scatter-absorber spacing. The ARM-versus-df trend and the DEC trade-off are directly measured and internally consistent, so the basic demonstrator claim is plausible. The most load-bearing weakness is the efficiency comparison: the 'factor of ~50' is the headline quantitative advantage, but it depends on an efficiency definition that is never fully stated and on an extrapolation from 165 mm to 30 mm that is not derived. The reader's weakest-assumption correctly flags the underspecified event selection; my concern is slightly more specific, because even if the event selection were stated, the factor-of-50 claim would still need a like-for-like comparison with the cited cameras. I do not see a more serious internal inconsistency: the paper is honest about thresholds, missing channels, and the need for neutron-beam validation. The reported ARM and FOV are empirical and, within the stated limitations, support the imaging-capability claim. The missing selection and comparison details are addressable by re-analysis and should be required before the high-efficiency claim is accepted as quantitative, but they do not overturn the demonstration itself. Hence the reader's CONDITIONAL verdict stands unchanged.","tokens_in":16997,"tokens_out":6234,"duration_ms":64010,"concrete_test":"Re-analyze the stored list-mode data for the Sec. 4.2/4.3 runs under two selection rules: (i) events whose S+A add-back sum lies within, say, ±2σ of the 662-keV peak of Fig. 8, and (ii) events with exactly one reconstructed interaction cluster in each layer. Recompute the Fig. 10 efficiency curve, the ds=30 mm extrapolated value, and the Sec. 5 factor-of-50 comparison against Ref. [44] and [52] using identical source distance and energy-window conventions. If the factor drops below ~10, or the ARM values broaden by more than a few degrees, the high-efficiency and imaging-resolution claims are not established as stated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Sec. 5 claims that i-TED's efficiency is a factor of ~50 higher than other scintillator-based Compton cameras [44,52]. This is the quantitative support for calling i-TED 'high-efficiency.' The measurement in Sec. 4.2, however, is specified only by a 10 ns S-A coincidence window and a ~100 keV threshold (Sec. 4). The paper does not state whether the add-back sum was required to be in the 662-keV full-energy peak, whether exactly one interaction per layer was required, or how multi-interaction events were handled. If the denominator is any S-A coincidence above threshold, the quoted efficiency is an inclusive coincidence efficiency; many such events deposit only part of the 662 keV and would not reconstruct a correct Compton cone. Comparing that number with literature photopeak/imaging efficiencies would inflate the factor. In addition, Fig. 10 shows a right-hand axis obtained by extrapolating the 165 mm source measurement to ds=30 mm, and Sec. 5's factor appears to rely on that extrapolation; no comparison table with source distances, energy windows, or analysis cuts of Refs. [44,52] is given. The ARM measurements in Sec. 4.3 are less exposed, since they are internally consistent point-source values, but the same missing selection criteria leave open whether the reported 8-14 degrees include mis-reconstructed partial-energy events.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports the assembly and first laboratory characterization of the i-TED demonstrator, a Compton camera built from one 50x50x10 mm^3 LaCl3(Ce) monolithic scatter crystal and four 50x50x25 mm^3 absorber crystals read out by 320 SiPM channels. The distinctive feature is Dynamic Electronic Collimation (DEC), implemented with a motorized stage that changes the scatter-absorber distance d_f over a 50 mm range. The authors present measurements of the 662 keV detection efficiency versus d_f, the angular resolution measure (ARM) versus d_f, backprojected images of a 22Na point source, and a field-of-view study. They report an efficiency about a factor of 50 higher than other scintillator-based Compton cameras at similar energies, angular resolutions between about 8 and 14 degrees, and an angular field of view of about 2/3 of 2 pi. The intended application is background discrimination in neutron time-of-flight capture measurements.","tokens_in":17228,"tokens_out":3910,"duration_ms":43278,"significance":"If the reported performance is taken at face value, the paper demonstrates a working high-efficiency Compton imager based on large monolithic scintillators, with a remotely adjustable geometry that enables an efficiency-resolution trade-off. This is a useful step for the i-TED program and for scintillator-based Compton imaging more generally. The main strengths are that the efficiency and ARM values come from direct laboratory measurements with quoted statistical uncertainties, and the authors are honest about hardware limitations such as missing readout channels and the 100 keV threshold. However, the central quantitative claim, the factor-of-50 efficiency comparison with other cameras, currently rests on an incompletely specified event selection and on an extrapolation of the source distance that is not described. These issues are fixable but are load-bearing for the paper's main conclusion.","major_comments":[{"comment":"The efficiency measurement is not sufficiently specified. The text states only a 10 ns coincidence window and a ~100 keV threshold, but does not state whether events were required to deposit the full 662 keV energy in the add-back spectrum, whether exactly one interaction per layer was required, or how multi-interaction and partial-energy events were treated. If the denominator is any S-A coincidence above threshold, the quoted efficiency is an inclusive coincidence efficiency that is not directly comparable to photopeak or imaging efficiencies in the literature. Please define the efficiency precisely, including the energy window, interaction multiplicity selection, and background subtraction.","section":"Sec. 4.2, Fig. 10"},{"comment":"The factor-of-50 comparison with Refs. [44,52] relies on an extrapolation of the efficiency from a source distance of 165 mm to 30 mm, shown on the right-hand axis of Fig. 10, but the extrapolation procedure is not described and no justification is given for the assumed distance scaling. Moreover, no comparison table is provided listing source distance, energy window, and analysis cuts for i-TED and for the cited cameras. Please either provide a like-for-like comparison with all relevant conditions stated, or substantially soften the factor-of-50 claim.","section":"Sec. 4.2 and Sec. 5"},{"comment":"The angular resolution measurement is internally consistent but the event selection is again unspecified. The ARM is defined only as 'to the level of one standard deviation' without stating whether the distribution is of the angular distance between the true source direction and the backprojected Compton cone for all accepted events, and without stating whether partial-energy or multi-interaction events were included. This matters because including mis-reconstructed partial-energy events could degrade the ARM and make the quoted 8-14 degrees not representative of imaging quality for a point source. Please define the ARM distribution and the event selection used for it.","section":"Sec. 4.3, Fig. 11"}],"minor_comments":[{"comment":"The design curves in Fig. 1 use spatial resolutions of 1-3 mm, while Sec. 3 reports a depth-of-interaction uncertainty of about 5 mm. Please clarify whether the quoted DOI uncertainty is included in the δr values used in Eq. (1) and Fig. 1, or whether the figure represents an idealized limit.","section":"Sec. 2, Fig. 1"},{"comment":"The image resolution is estimated from projections over a ±20 mm selection around the maximum; please state explicitly how the projection width is converted to an angular or spatial resolution value and whether the result is a FWHM or another measure.","section":"Sec. 4.1"},{"comment":"The source activity is given as 210.4 kBq without an uncertainty; including the activity uncertainty and its contribution to the absolute efficiency uncertainty would improve the reproducibility of the efficiency numbers.","section":"Sec. 4.2"},{"comment":"The field-of-view estimate of 'about 2/3 of 2π' is not accompanied by a quantitative criterion, such as the region within which reconstructed positions agree with true positions within a given tolerance. Please specify how the number was obtained.","section":"Sec. 4.4, Fig. 13"},{"comment":"The acronym PHWT is used without definition in the discussion of future work; please spell it out at first use.","section":"Sec. 5"},{"comment":"The text contains several typographical errors, including 'repeteability', 'uncerainty', 'lineraity', and 'taken appart'; a careful proofreading pass would remove these.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper is a demonstrator report and the hardware effort is real, but the main quantitative claim (factor-of-50 efficiency) is currently not verifiable from the manuscript because the event selection and the distance extrapolation are unspecified. If the authors provide the missing analysis details and either substantiate or qualify the comparison, the paper would likely be acceptable. I see no indication of circularity or fabrication; the issue is incompleteness of reporting."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First the bottom line: this is a legitimate first-demonstrator paper. The i-TED group has actually built the device that was simulated in Ref. [24], and the measurements show the expected trade-off between efficiency and angular resolution as the scatter-absorber distance is varied. The DEC concept is not new here—it is in their patent and earlier paper—but this is the first experimental performance characterization, and that is the paper's contribution.\n\nWhat is done well: the efficiency-vs-d_f and ARM-vs-d_f data are direct lab measurements with a calibrated source, the fits are simple, and the text is unusually candid about the current limitations (missing readout channels, 100 keV threshold, simple backprojection, no neutron-beam data yet). The field-of-view measurement with the gantry is a nice extra. The central claim, that i-TED works as a Compton imager with a controllable efficiency-resolution trade-off, is supported by the data as presented.\n\nThe soft spots are real but fixable. The biggest one is that the paper never states the event selection used for the efficiency and ARM numbers. We know there is a 10 ns coincidence window and a ~100 keV threshold, but not whether the add-back energy was required to be in the 662-keV full-energy peak, or whether exactly one interaction per layer was required. If the efficiency is just any S–A coincidence above threshold, it includes partial-energy events that would not reconstruct a correct Compton cone, and the factor-of-~50 comparison with other scintillator cameras (Refs. [44,52]) is not meaningful without a consistent definition. The authors should provide a comparison table with source distances, energy windows, and analysis cuts.\n\nOne point in the stress-test note is not quite right: Fig. 10's right-hand axis is the efficiency at d_f = 30 mm, not an extrapolation to a source distance of 30 mm. The source was at 165 mm throughout. That is a minor misreading. The extrapolation concern that remains is simply that the fit is used to quote a number at 30 mm, but that is well within the measured range.\n\nThe ARM numbers (8–14°) are less exposed because they are internally consistent point-source measurements, but the same selection ambiguity means the reported values might include mis-reconstructed events. That is a paper-clarity issue, not a fatal one.\n\nWho is this for? Detector developers working on Compton cameras for neutron TOF, and anyone considering DEC for other applications. It is a niche but useful data point. It deserves peer review; my recommendation is to send it out, asking the authors to define the event selection and make the efficiency comparison apples-to-apples.","headline":"First i-TED demonstrator paper: solid experimental characterization of a tunable Compton imager; the efficiency claims need a clearer event selection to be fully trustworthy.","tokens_in":17810,"tokens_out":3928,"would_cite":true,"duration_ms":33882,"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":"The paper claims that the i-TED demonstrator, a Compton imager with a motorized scatter-absorber separation, achieves 8–14 degree angular resolution for 662 keV gamma rays with about 50 times the efficiency of comparable…","keywords":["Compton imaging","Dynamic Electronic Collimation","LaCl3(Ce) monolithic crystals","Silicon photomultiplier","position-sensitive detectors","neutron time-of-flight","angular resolution measure","gamma-ray detection efficiency"],"falsifier":"One decisive check is to rerun the $^{137}$Cs measurements with a gate on the $662$ keV full-energy peak in the add-back spectrum and with exactly one reconstructed interaction per layer, then recompute efficiency and ARM; if the efficiency no longer exceeds the cited cameras by a factor of about $50$, or if the ARM moves outside $8^\\circ$ to $14^\\circ$, the reported imaging performance is not representative of clean single-Compton events.","tokens_in":16776,"feed_emoji":"⚛️","tokens_out":7846,"duration_ms":74543,"temperature":0.7,"pith_summary":"This paper tries to establish that a detector originally designed for neutron-capture cross-section measurements can also work as a high-efficiency gamma-ray imager, and that its distinguishing feature, Dynamic Electronic Collimation, an adjustable separation between the scatter and absorber layers, delivers a practical trade-off between counting efficiency and image resolution. The demonstrator reports angular resolutions between $8^\\circ$ and $14^\\circ$ for $662$ keV gamma rays, an efficiency about a factor of $50$ higher than other scintillator-based Compton cameras in the same energy range, and a field of view covering about two-thirds of $2\\pi$. If these figures hold in a neutron beam, the camera would let time-of-flight capture measurements tag the spatial origin of each gamma event and reject the neutron-induced background that limits present detectors, without heavy shielding.","feed_headline":"Adjustable Compton imager reaches 8-14 degree resolution","feed_subtitle":"Five-crystal demonstrator images 662 keV gamma rays with 50x the efficiency of comparable cameras.","key_machinery":"The load-bearing mechanism is Dynamic Electronic Collimation: a motorized stage changes the distance $d_f$ between the $50\\times50\\times10$ mm$^3$ scatter crystal and the four $50\\times50\\times25$ mm$^3$ absorber crystals, tuning the camera between high efficiency at small $d_f$ and better angular resolution at large $d_f$. The argument is carried by three analytic constraints: the Compton-angle uncertainty formula $\\delta\\theta$, which grows with energy and position uncertainty and with $1/\\sin\\theta$; the geometrical maximum-angle formula $\\theta_{\\mathrm{Max}}$ set by detector sizes and source distance; and the threshold-induced minimum angle, about $30^\\circ$ for a $100$ keV threshold at $662$ keV. Position reconstruction relies on a fitted analytic light-response model for the monolithic crystals, giving about $1$ mm fwhm position accuracy, and the images are formed with a list-mode backprojection that intersects each Compton cone with the source plane.","core_discovery":"On the paper's own terms, the discovery is that a Compton imager built from five large LaCl$_3$(Ce) monolithic crystals and 320 readout channels works both as a total-energy detector and as a gamma-ray imager. For $662$ keV photons, the measured efficiency falls smoothly as the scatter-absorber separation $d_f$ grows, while the angular resolution improves roughly linearly from about $14^\\circ$ at $d_f = 20$ mm to about $9^\\circ$ at $d_f = 50$ mm and then flattens at $8.5(3)^\\circ$. The sum-signal energy resolution is about $9\\%$ fwhm at $662$ keV, the intrinsic position resolution is about $1$ mm fwhm, and the depth-of-interaction uncertainty is about $5$ mm; these values feed the analytic angular-uncertainty relation and motivate the design choices. Backprojected images of point-like $^{22}$Na and $^{137}$Cs sources reconstruct the source location over roughly two-thirds of $2\\pi$. The paper's intended consequence is that the same detector can provide event-by-event gamma-ray direction information in neutron time-of-flight experiments, enabling spatial background discrimination.","pith_inferences":["If the measurements are repeated with a strict gate on the full-energy peak and exactly one interaction per layer, the factor-of-50 efficiency advantage will probably shrink; as written, the efficiency includes whatever event mixture the unstated selection admits.","The same adjustable-separation hardware generalizes outside neutron capture, for example to security or decommissioning surveys where a short $d_f$ gives a fast coarse image and a longer $d_f$ refines it.","Because the dominant angular-error term is the energy resolution, a modest spectroscopic improvement toward the $3.5\\%$ assumed in the design curves could bring the angular resolution below $8^\\circ$ without changing the crystals or electronics.","A maximum-likelihood reconstruction would likely recover the compressed peripheral field-of-view positions seen in the simple backprojection images, so the $2/3$ of $2\\pi$ estimate is a conservative floor for the hardware."],"forward_implications":["For neutron-capture time-of-flight measurements, the imaging capability can suppress background from neutron-capture gamma rays in the surroundings, because true events point back to the sample while background events point elsewhere.","An operator can tune $d_f$ for each neutron-energy window, using short separations where statistics are scarce and long separations where background rejection matters more.","At the measured $8^\\circ$ to $14^\\circ$ angular resolution, a field of view of about $2/3$ of $2\\pi$ is sufficient to monitor the main background directions around the sample.","Improving the crystal energy resolution and lowering the $100$ keV threshold would extend the usable Compton-angle range and improve angular resolution beyond the present values.","The factor-of-50 efficiency gain relative to other scintillator cameras keeps total measuring time feasible for low-mass or radioactive samples."],"supporting_citations":[{"why":"Defines the i-TED concept and the expected event-by-event background discrimination that the demonstrator is meant to realize.","marker":"[24]"},{"why":"Supplies the analytic Compton-angle uncertainty expression that motivates the efficiency-resolution trade-off.","marker":"[30]"},{"why":"Provides the analytic light-response function used to reconstruct interaction positions in monolithic crystals.","marker":"[31]"},{"why":"Establishes the position and depth-of-interaction resolutions for these LaCl3(Ce) crystals with SiPM readout.","marker":"[32]"},{"why":"Serves as a reference for angular resolution at a similar scatter-absorber separation.","marker":"[41]"},{"why":"Used as a high-resolution scintillator Compton camera baseline for the efficiency and angular-resolution comparison.","marker":"[44]"},{"why":"Provides the multilayer Compton camera performance used for the resolution comparison.","marker":"[45]"},{"why":"Supplies the fast list-mode backprojection algorithm used to form the reconstructed images.","marker":"[48]"},{"why":"Used as another scintillator-based Compton camera baseline for the factor-of-50 efficiency comparison.","marker":"[52]"}],"fun_headline_variants":["Adjustable Compton imager dials resolution from 14° to 8.5°","Five-crystal Compton imager resolves 8.5° at 662 keV","i-TED: gamma imager with adjustable angular resolution","Tunable Compton imaging for neutron capture measurements","LaCl3 imager: 8.5° resolution with efficiency trade-off"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The reported efficiency and angular resolution rest on an event selection that is only partially specified, since the paper does not state whether events must deposit the full 662 keV energy and contain exactly one interaction in each layer.","fun_headline_variants_meta":{"raw":{"variants":["Adjustable Compton imager dials resolution from 14° to 8.5°","Five-crystal Compton imager resolves 8.5° at 662 keV","i-TED: gamma imager with adjustable angular resolution","Tunable Compton imaging for neutron capture measurements","LaCl3 imager: 8.5° resolution with efficiency trade-off"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000735,"raw_usage":{"total_tokens":3278,"prompt_tokens":927,"completion_tokens":2351,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":543,"completion_tokens_details":{"reasoning_tokens":2255}},"tokens_in":543,"tokens_out":2351,"duration_ms":17999,"temperature":1.0,"reasoning_tokens":2255,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:41:28.024200+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"One decisive check is to rerun the $^{137}$Cs measurements with a gate on the $662$ keV full-energy peak in the add-back spectrum and with exactly one reconstructed interaction per layer, then recompute efficiency and ARM; if the efficiency no longer exceeds the cited cameras by a factor of about $50$, or if the ARM moves outside $8^\\circ$ to $14^\\circ$, the reported imaging performance is not representative of clean single-Compton events.","supporting_citations":[{"cited_title":"Walter, H","cited_arxiv_id":null,"evidence_quote":"Defines the i-TED concept and the expected event-by-event background discrimination that the demonstrator is meant to realize."},{"cited_title":"Lessons learnt from comptel for future telescopes","cited_arxiv_id":null,"evidence_quote":"Supplies the analytic Compton-angle uncertainty expression that motivates the efficiency-resolution trade-off."},{"cited_title":"narrow-FOV Si /CdTe semiconductor Compton camera","cited_arxiv_id":null,"evidence_quote":"Provides the analytic light-response function used to reconstruct interaction positions in monolithic crystals."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the position and depth-of-interaction resolutions for these LaCl3(Ce) crystals with SiPM readout."},{"cited_title":"Olleros, L","cited_arxiv_id":null,"evidence_quote":"Serves as a reference for angular resolution at a similar scatter-absorber separation."},{"cited_title":"Domingo Pardo, L","cited_arxiv_id":null,"evidence_quote":"Used as a high-resolution scintillator Compton camera baseline for the efficiency and angular-resolution comparison."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the multilayer Compton camera performance used for the resolution comparison."},{"cited_title":"Multiple scattering Compton camera with neutron activation for mate- rial inspection","cited_arxiv_id":null,"evidence_quote":"Supplies the fast list-mode backprojection algorithm used to form the reconstructed images."},{"cited_title":"Di Francesco, R","cited_arxiv_id":null,"evidence_quote":"Used as another scintillator-based Compton camera baseline for the factor-of-50 efficiency comparison."}],"review_version":1}