{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp09_archetype_breadth150_20260804","research_id":"eoa_inverse_innovation_exp09_light_prior_art_20260804","cell_id":"predictive_residual_processing__chemistry_materials","search_lanes":{"direct_problem_and_intervention":{"queries":["operando X-ray diffraction real-time analysis emerging phase anomaly detection residual forward model","beamline streaming XRD live analysis data rates decision making experiment"],"source_ids":["SRC1","SRC2","SRC4"],"no_result_note":null},"synonyms_and_historical_terms":{"queries":["time resolved XRD online phase identification real time autonomous experiment","operando diffraction sequential Rietveld refinement real time residual difference pattern automated"],"source_ids":["SRC1","SRC2","SRC3"],"no_result_note":null},"products_practices_and_standards":{"queries":["X-ray diffraction streaming data reduction peak finding real-time beamline software","XRD residual difference plot unexplained peaks automatic phase identification iterative residual peaks"],"source_ids":["SRC2","SRC3","SRC4"],"no_result_note":null},"component_combination":{"queries":["diffraction pattern model residual anomaly detection phase transition adaptive acquisition","real-time XRD uncertainty feature prioritization residual phase identification raw replay"],"source_ids":["SRC1","SRC2","SRC3","SRC4"],"no_result_note":null}},"sources":[{"source_id":"SRC1","title":"Adaptively driven X-ray diffraction guided by machine learning for autonomous phase identification","publisher":"Springer Nature, npj Computational Materials","url":"https://www.nature.com/articles/s41524-023-00984-y","source_type":"PRIMARY_RESEARCH","claims_supported":["In situ and operando XRD must balance rapid acquisition against signal quality when monitoring short-lived intermediate phases.","Small peaks can blend into noise, while slow scans can outlast a transient phase.","An uncertainty-aware classifier and class-activation maps can prioritize discriminating scan regions in-line and steer confirmatory measurements.","The demonstrated adaptive workflow detected a short-lived intermediate phase missed by conventional scans."]},{"source_id":"SRC2","title":"Spotlight: efficient automated global optimization in rietveld analysis of diffraction data","publisher":"Springer Nature, Scientific Reports","url":"https://www.nature.com/articles/s41598-025-92452-4","source_type":"PRIMARY_RESEARCH","claims_supported":["Analysis of tens or hundreds of time-resolved diffraction datasets is a recognized bottleneck.","Rietveld analysis already compares modeled and observed intensities and exposes a dataset-minus-model residual.","Spotlight automates parameter and phase-fraction optimization and updates a surrogate when new refinement results invalidate it.","Unknown-phase identification remains rate-limiting and may require an expert."]},{"source_id":"SRC3","title":"New analytical software for XRD simplifies identification of complex phase mixtures","publisher":"International Union of Crystallography, Journal of Applied Crystallography","url":"https://journals.iucr.org/j/issues/2001/06/00/me0147/","source_type":"TRADE_PROFESSIONAL","claims_supported":["X'Pert HighScore used auto-residue scoring to rank candidate phases primarily from unexplained profile features and peaks.","The software dynamically recalculated candidate scores as phases were identified and could run configurable phase-identification sequences in batch mode.","Residual-guided attention to unexplained diffraction features is therefore an older commercial practice, not a phrase-level novelty."]},{"source_id":"SRC4","title":"PDFstream Documentation","publisher":"PDFstream/xpdAcq project","url":"https://xpdacq.github.io/PDFstream/","source_type":"FIRST_PARTY_PRODUCT","claims_supported":["PDFstream is deployed for automatic processing of streaming X-ray scattering data at beamline 28-ID.","Its pipeline accepts streams of XRD images and emits processed XRD and pair-distribution-function data.","The documented system includes separate processing, visualization, serialization, beamline testing, and replay-analysis functions."]}],"problem_evidence":{"status":"SUPPORTED","finding":"The underlying problem is visible: time-resolved diffraction produces analysis bottlenecks, small or transient-phase peaks can be obscured by noise or disappear during slower measurements, and conventional post hoc or expert-intensive analysis can miss the useful decision window. The sources do not quantify how often this occurs specifically in rapid two-dimensional operando detector-frame review, but they directly demonstrate the technical failure mode and the value of in-line prioritization.","source_ids":["SRC1","SRC2","SRC4"]},"closest_prior_art":[{"name":"Adaptive XRD with XRD-AutoAnalyzer","source_ids":["SRC1"],"overlap":"Performs in-line phase prediction, quantifies uncertainty, prioritizes scientifically discriminating diffraction regions, and requests confirmatory measurements quickly enough to detect a short-lived intermediate phase.","remaining_difference":"It prioritizes scan regions using classifier confidence and class-activation-map differences rather than transmitting a signed, reconstructive observed-minus-forward-model residual with synchronized model provenance, independent raw audits, governed slower updates, and explicit full-frame fallback."},{"name":"Rietveld residual analysis with Spotlight optimization","source_ids":["SRC2"],"overlap":"Uses a physical diffraction model, observed-versus-modeled discrepancy, explicit residual output, automated parameter optimization, distributed compute, and a surrogate updated when new results invalidate it.","remaining_difference":"Spotlight targets reliable starting values and refinement optimization; it does not present a per-frame residual-only live attention channel or the proposed checksum, audit-sampling, protected-condition bypass, and raw-presentation fallback controls."},{"name":"X'Pert HighScore auto-residue scoring","source_ids":["SRC3"],"overlap":"Ranks candidate phases from unexplained profile features and peaks, dynamically recalculates candidates after phase assignments, and supports automated batch sequences.","remaining_difference":"It is search-match phase identification on diffraction profiles, not a synchronized predictive detector-image residual channel for live operando triage with uncertainty-weighted consequences, reconstructive context, independent raw audits, and governed fallback."},{"name":"PDFstream","source_ids":["SRC4"],"overlap":"Provides beamline-deployed streaming ingestion, automatic XRD processing, visualization, and replay capabilities that could host an offline shadow evaluation.","remaining_difference":"The retained documentation does not describe physical-model residual selection, residual-driven phase alerts, model-version handshakes, raw audit sampling, or residual-failure fallback."}],"prior_art_disposition":"ADJACENT_PRIOR_ART","contrastive_claim_remaining":"Relative to the retained prior art, the testable design contrast is the single governed coupling of a versioned physical diffraction-frame predictor, signed reconstructive residuals prioritized by uncertainty and scientific consequence, model/display checksum synchronization, slower reviewable correction, independent raw-frame auditing, and automatic restoration of full-frame presentation. Individual elements and several close combinations are established; only this conjunction remains distinct in the bounded search.","contrastive_claim_falsifier":"The contrast is falsified by a pre-existing system or publication that implements substantially this same governed conjunction for live or replayed time-resolved diffraction, or operationally if a frozen-model replay misses any protected event, violates reconstruction tolerance, fails provenance synchronization, or does not reduce measured review or compute burden versus the full-frame workflow after audit and fallback costs.","gates":{"adequate_source_search":{"status":"PASS","rationale":"Four opened sources from three publisher groups cover direct problem evidence, older residual terminology, commercial and beamline practices, adaptive XRD, Rietveld modeling, streaming pipelines, and component combinations. This is adequate for a coarse screen, though not an exhaustive patent or literature search.","source_ids":["SRC1","SRC2","SRC3","SRC4"]},"supported_problem":{"status":"PASS","rationale":"Primary research documents both the time-resolved diffraction analysis bottleneck and missed small or short-lived phase evidence, while beamline documentation confirms operational streaming-analysis practice.","source_ids":["SRC1","SRC2","SRC4"]},"distinct_testable_claim":{"status":"PASS","rationale":"Although residual-guided phase identification, model fitting, adaptive attention, streaming, and model updating are prior art, the proposed governed conjunction has a clear operational boundary and measurable falsifiers.","source_ids":["SRC1","SRC2","SRC3","SRC4"]},"bounded_next_test":{"status":"PASS","rationale":"A read-only replay of one completed run with blinded full-frame labels, a frozen model, predeclared thresholds, protected-event recall, reconstruction error, review time, compute cost, audit disagreement, synchronization failures, and fallback counts is narrow and decision-producing. PDFstream documents that replayable streaming analysis is operationally plausible.","source_ids":["SRC1","SRC2","SRC4"]},"no_obvious_safety_or_authority_stop":{"status":"PASS","rationale":"The authorized first step is offline and read-only, preserves authoritative raw data, forbids automatic instrument changes and scientific model adoption, assigns instrument authority to beamline staff, and has explicit halt criteria. No retained source reveals an obvious authority or safety prohibition on such a replay.","source_ids":["SRC4"]}},"screen_survival":true,"world_novelty_boundary":"This bounded four-source screen establishes neither world novelty nor patentability. It also cannot establish market size, expert acceptance, prevalence across facilities, realized value, or whether patents, proprietary beamline systems, unpublished workflows, standards, or literature outside the searched terminology already contain the remaining conjunction."}