{"id":"28d76775-ea44-46b3-8b0a-37e427126147","arxiv_id":"2606.29285","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":4.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"An OPM gradiometer MCG system with WMSPCA filtering and SQE scoring achieves 28.56 dB QRS SNR in unshielded environments.","lead":"Researchers built an optically pumped magnetometer system that records adult heart magnetic fields in ordinary rooms without any magnetic shielding. If the performance holds, it removes a major barrier to using magnetocardiography in everyday clinical settings.","discovery_kind":"new_application","skeptic_critique":{"model":"grok-4.3","headline":"Signals may be residual noise or WMSPCA artifacts rather than true cardiac fields (no independent validation shown)","rationale":"Reader's weakest assumption matches the load-bearing point exactly. Full-text inaccessibility prevents checking for the missing controls, so the UNVERDICTED stance is retained; the proposed test is the minimal concrete check that would resolve the authenticity question.","tokens_in":1725,"tokens_out":288,"duration_ms":19042,"concrete_test":"Re-acquire at one anterior site with simultaneous surface ECG; verify that MCG QRS onset and T-wave peak align with ECG R-peak within the known 10-30 ms electromagnetic delay and that the sign of the magnetic deflection matches the expected anterior-posterior component of the cardiac dipole. Mismatch > chance level falsifies genuineness.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The feasibility claim requires that post-gradiometer (31 dB CMRR) and WMSPCA-filtered traces at the 16 sequential thoracic sites contain genuine MCG morphology (QRS SNR 28.56 dB, T-wave fidelity). Sequential acquisition plus multiscale PCA can pass cardiac-like transients from imperfectly subtracted environmental fields or introduce synthetic features; nothing in the abstract (or accessible methods) rules this out via simultaneous ECG timing, field-map consistency with a current dipole, or shielded-MCG cross-check.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The manuscript presents an OPM-based magnetocardiography system designed for unshielded, room-temperature operation in a gradiometer configuration. It reports a common-mode rejection ratio of 31 dB and gradient sensitivity of 314 fT/cm/√Hz, with sequential recordings at 16 anterior thoracic locations processed by wavelet multiscale principal component analysis (WMSPCA) filtering and signal quality estimation, yielding a QRS SNR of 28.56 ± 5.61 dB and claiming consistent morphological fidelity sufficient to demonstrate feasibility of clinical-grade unshielded MCG.","tokens_in":1852,"tokens_out":376,"duration_ms":18855,"significance":"If the post-processed traces can be independently confirmed as cardiac magnetic fields, the reported sensitivity and SNR values would constitute a meaningful experimental step toward practical MCG deployment in ordinary clinical environments without magnetic shielding.","major_comments":[{"comment":"Abstract (final paragraph): The feasibility claim that the 16 sequentially acquired traces exhibit 'consistent morphological fidelity across the QRS complex and T-wave segments' is load-bearing for the central result, yet the manuscript supplies no simultaneous ECG reference, no shielded-MCG cross-check, and no quantitative test (e.g., dipole-field consistency or timing correlation) to establish that the reported 28.56 dB QRS SNR reflects genuine cardiac sources rather than residual environmental fields or WMSPCA-induced transients after the 31 dB CMRR subtraction.","section":"Abstract"}],"minor_comments":[{"comment":"The sensitivity is given as 314 fT/cm/√Hz; the notation should be rendered consistently as fT cm^{-1} Hz^{-1/2} throughout the text and figures.","section":"Abstract"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the detailed review and constructive criticism. The central concern is the strength of evidence supporting that the recorded signals originate from cardiac sources rather than residual noise. We address this point directly below and propose textual revisions to qualify our claims accordingly.","responses":[{"response":"We agree that the manuscript does not include simultaneous ECG, shielded-MCG comparison, or explicit quantitative tests such as dipole fitting or cross-correlation with reference timing. The current validation rests on (i) the 31 dB CMRR of the gradiometer configuration, (ii) the multi-location consistency of QRS and T-wave timing and polarity expected for anterior thoracic recordings, and (iii) the SQE scoring that penalizes non-cardiac morphologies. These elements support the reported SNR but do not constitute independent confirmation of cardiac origin. To address the concern, we will revise the abstract and discussion to remove the unqualified 'clinical-grade' feasibility statement, explicitly note the absence of reference-channel validation, and add a limitations paragraph describing the reliance on morphological consistency alone. No new experimental data can be added at this stage.","revision_made":"yes","referee_comment":"[Abstract] Abstract (final paragraph): The feasibility claim that the 16 sequentially acquired traces exhibit 'consistent morphological fidelity across the QRS complex and T-wave segments' is load-bearing for the central result, yet the manuscript supplies no simultaneous ECG reference, no shielded-MCG cross-check, and no quantitative test (e.g., dipole-field consistency or timing correlation) to establish that the reported 28.56 dB QRS SNR reflects genuine cardiac sources rather than residual environmental fields or WMSPCA-induced transients after the 31 dB CMRR subtraction."}],"tokens_in":1335,"tokens_out":365,"duration_ms":23077,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The main takeaway is that an OPM gradiometer setup reached 31 dB CMRR and 314 fT/cm/sqrt(Hz) sensitivity without shielding, then sequential measurements at 16 thoracic sites were cleaned with wavelet multiscale PCA and SQE scoring to reach 28.56 dB QRS SNR. The integration for unshielded adult MCG is the practical step forward.\n\nThe work does a straightforward job of stating the hardware configuration and the denoising steps. Reporting specific sensitivity and rejection figures gives other groups something concrete to compare against when they try similar ambient setups. The choice to use sequential single-sensor scans instead of a full array also keeps the hardware simpler.\n\nThe soft spot is the missing validation. The abstract does not mention simultaneous ECG to lock timing, a shielded-room reference recording on the same subjects, or even a basic check that the 16-point field pattern matches a cardiac dipole. Sequential acquisition plus multiscale PCA can pass through or create cardiac-looking transients from imperfectly subtracted environmental fields, and nothing described rules that out. That leaves the central claim resting on an assumption that needs direct evidence.\n\nThis paper is for readers working on portable biomagnetic sensors or clinical translation of MCG. Someone who needs to know what noise floor is reachable in a normal room will find the numbers useful, while anyone who needs to trust the morphology will want the extra checks.\n\nI would send it to peer review. The experimental claims are specific enough for referees to ask for the validation data, and the topic is relevant enough that the paper is worth the time even if revisions are needed.","headline":"The paper reports concrete OPM gradiometer numbers in ambient conditions and a WMSPCA pipeline, but the abstract supplies no checks that the final traces are cardiac fields rather than residual noise or artifacts.","tokens_in":2345,"tokens_out":409,"would_cite":false,"duration_ms":27104,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"An OPM gradiometer with WMSPCA filtering records cardiac magnetic fields in unshielded rooms at 28.5 dB SNR.","keywords":["magnetocardiography","optically pumped magnetometers","unshielded recording","room-temperature MCG","gradiometer configuration","WMSPCA filtering","cardiac magnetic fields","signal quality estimation"],"falsifier":"A side-by-side comparison in which the unshielded waveforms fail to match timing and shape of simultaneous ECG traces or shielded MCG recordings at the same sites would show the signals are not cardiac fields.","tokens_in":2634,"feed_emoji":"🧲","tokens_out":662,"duration_ms":24787,"temperature":0.7,"pith_summary":"The paper establishes that room-temperature magnetocardiography is feasible in ordinary magnetic environments by using optically pumped magnetometers in a gradiometer arrangement. Sequential recordings at 16 thoracic sites, processed through wavelet multiscale principal component analysis, produce clear QRS and T-wave shapes with an average signal-to-noise ratio of 28.56 dB. A sympathetic reader would care because this removes the requirement for expensive shielded rooms that have confined MCG to specialized facilities. The work therefore points toward broader clinical access for magnetic heart mapping.","feed_headline":"Unshielded OPM gradiometer captures cardiac fields at 28 dB SNR","feed_subtitle":"WMSPCA filtering yields clinical-grade QRS and T-wave morphology at 16 chest sites in ordinary rooms.","key_machinery":"Gradiometer configuration of optically pumped magnetometers paired with wavelet multiscale principal component analysis filtering to suppress ambient noise while retaining cardiac field morphology.","core_discovery":"The OPM-based system in gradiometer configuration delivers a common-mode rejection ratio of 31 dB and gradient sensitivity of 314 fT/cm/sqrt(Hz). After WMSPCA filtering and signal quality estimation, the QRS complex reaches an SNR of 28.56 ± 5.61 dB across all 16 locations while preserving morphological fidelity in both QRS and T-wave segments, demonstrating that clinical-grade signals can be obtained without shielding.","pith_inferences":["If the signals are cardiac, the same processing chain could support source localization of heart currents without a shielded enclosure.","The framework lowers the infrastructure barrier enough that routine MCG might eventually run alongside standard ECG in ordinary clinics.","Faster acquisition hardware would be a direct next step to enable real-time rather than sequential mapping."],"forward_implications":["Cardiac magnetic signals become acquirable in ambient conditions without active or passive shielding.","Consistent morphological fidelity holds across QRS complex and T-wave at all 16 measurement locations.","The approach supports sequential multi-site mapping suitable for hospital and point-of-care use.","Gradient sensitivity of 314 fT/cm/sqrt(Hz) combined with 31 dB rejection enables usable recordings in real-world settings."],"fun_headline_variants":["Unshielded OPMs capture cardiac fields with 28 dB SNR","Gradiometer OPM-MCG yields 28 dB QRS SNR in ambient conditions","WMSPCA filtered OPM signals achieve 28 dB SNR across 16 locations","Room-temperature MCG at 31 dB CMRR without magnetic shielding"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The filtered signals at the 16 thoracic locations represent genuine cardiac magnetic fields rather than residual environmental noise or processing artifacts.","fun_headline_variants_meta":{"raw":{"variants":["Unshielded OPMs capture cardiac fields with 28 dB SNR","Gradiometer OPM-MCG yields 28 dB QRS SNR in ambient conditions","WMSPCA filtered OPM signals achieve 28 dB SNR across 16 locations","Room-temperature MCG at 31 dB CMRR without magnetic shielding"]},"model":"grok-4.3","cost_usd":0.007312,"raw_usage":{"total_tokens":3301,"prompt_tokens":699,"num_sources_used":0,"completion_tokens":84,"cost_in_usd_ticks":73115500,"prompt_tokens_details":{"text_tokens":699,"audio_tokens":0,"image_tokens":0,"cached_tokens":64},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":2518,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":699,"tokens_out":84,"duration_ms":19431,"temperature":1.0,"reasoning_tokens":2518,"cache_read_input_tokens":64,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-30T02:16:32.451030+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"A side-by-side comparison in which the unshielded waveforms fail to match timing and shape of simultaneous ECG traces or shielded MCG recordings at the same sites would show the signals are not cardiac fields.","supporting_citations":[],"review_version":1}