{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp04_retrieval_first_paired20_20260802","cell_id":"invariant_mode_decomposition_design__medicine_healthcare","round_index":0,"assessments":[{"hypothesis_id":"H1","search_queries":["dynamic mode decomposition patient deterioration early warning vital signs hospital ward","eigenvalue stability analysis clinical deterioration vital signs ICU transfer","Royal College Physicians NEWS2 official standard vital signs deterioration","state space stability model inpatient deterioration multivariate vital signs"],"sources":[{"source_id":"H1-S1","title":"Tracking Progression of Patient State of Health in Critical Care Using Inferred Shared Dynamics in Physiological Time Series","publisher":"IEEE / PubMed Central","url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC4030703/","source_class":"PRIMARY_RESEARCH","claims_supported":["Physiological time series have been represented with locally coherent linear dynamic modes.","Time spent in particular inferred modes was associated with hospital mortality.","The authors identified combining the framework with broader clinical data to generate predictive alerts as future work."]},{"source_id":"H1-S2","title":"Dynamic early warning scores for predicting clinical deterioration in patients with respiratory disease","publisher":"PubMed, U.S. National Library of Medicine","url":"https://pubmed.ncbi.nlm.nih.gov/35953815/","source_class":"PRIMARY_RESEARCH","claims_supported":["A dynamic early-warning model used temporal features from clinical observations to predict ICU admission, death, and urgent intervention.","The study directly addresses the limitation that NEWS2 does not represent detailed temporal trends."]},{"source_id":"H1-S3","title":"National Early Warning Score (NEWS) 2","publisher":"Royal College of Physicians","url":"https://www.rcp.ac.uk/resources/national-early-warning-score-news-2/","source_class":"OFFICIAL_GUIDANCE","claims_supported":["NEWS2 aggregates six routinely measured physiological parameters into a deterioration score.","NEWS2 uses parameter-specific abnormality scores and trigger thresholds rather than an estimated patient-state operator or eigenvalue-stability test.","NEWS2 is formally endorsed for identifying acutely ill hospital patients in England."]}],"closest_analogue":"The inferred shared-dynamics critical-care model, supplemented by dynamic early-warning scores that use multivariable temporal features.","overlap":"The analogues model coupled longitudinal physiology, infer recurring patient-state dynamics, associate those dynamics with adverse outcomes, and contemplate or implement deterioration alerts beyond static single-variable thresholds.","remaining_difference":"The screened sources do not show a general-ward alert that estimates a local patient-state transition operator, classifies its eigenmodes by growth or damping, abstains on spectral/residual drift, and is evaluated at equal alerts per patient-day. The testable distinction is whether that explicit stability criterion yields at least two hours of additional warning without reducing positive predictive value relative to a temporal risk model or NEWS2.","classification":"POSSIBLE_DISTINCTION","disposition":"ADVANCE","rationale":"Close prior work establishes dynamic physiological modes and temporal deterioration prediction, but the shallow search did not expose an evaluated ward system whose alert variable is an eigenmode crossing a stability boundary. This is a specific methodological and operational distinction, not a claim of novelty from search failure."},{"hypothesis_id":"H2","search_queries":["hemodialysis intradialytic hypotension machine learning adaptive ultrafiltration biofeedback control trial","hemodialysis SVD modal analysis ultrafiltration hypotension patient-specific","KDIGO intradialytic hypotension ultrafiltration official conference conclusions","Fresenius Hemocontrol biofeedback ultrafiltration first party"],"sources":[{"source_id":"H2-S1","title":"A Novel Ultrafiltration Rate Feedback Controller for Use in Hemodialysis: First Clinical Experience: An Interventional Pilot Study","publisher":"Kidney360 / PubMed Central","url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC12483043/","source_class":"PRIMARY_RESEARCH","claims_supported":["An adaptive controller generated ultrafiltration-rate recommendations intended to steer relative blood volume toward a target trajectory.","The controller improved relative-blood-volume target attainment, although blood-pressure outcomes were not improved in the pilot.","The intervention was evaluated over repeated treatments within patients."]},{"source_id":"H2-S2","title":"Randomized Crossover Trial of Blood Volume Monitoring–Guided Ultrafiltration Biofeedback to Reduce Intradialytic Hypotensive Episodes with Hemodialysis","publisher":"Clinical Journal of the American Society of Nephrology / PubMed Central","url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC5672962/","source_class":"PRIMARY_RESEARCH","claims_supported":["Blood-volume biofeedback has already been used to automatically alter ultrafiltration rate during dialysis.","A randomized crossover design evaluated symptomatic intradialytic hypotension against clinical practice.","The controller acted from relative blood-volume measurements rather than a multivariable modal response map."]},{"source_id":"H2-S3","title":"Blood pressure and volume management in dialysis: conclusions from a Kidney Disease: Improving Global Outcomes Controversies Conference","publisher":"KDIGO","url":"https://kdigo.org/wp-content/uploads/2017/05/KDIGO-BP-Volume-in-Dialysis-FINAL.pdf","source_class":"OFFICIAL_GUIDANCE","claims_supported":["KDIGO identifies insufficient intravascular volume and inadequate cardiovascular compensation as major coupled contributors to intradialytic hypotension.","Symptomatic blood-pressure decline should prompt reassessment of ultrafiltration rate, treatment time, interdialytic weight gain, target weight, and medications.","KDIGO reports conflicting evidence for relative-blood-volume monitoring and emphasizes individualized rather than universal ultrafiltration thresholds."]}],"closest_analogue":"Adaptive relative-blood-volume-guided ultrafiltration feedback, including the interventional ultrafiltration-rate controller and randomized blood-volume biofeedback trial.","overlap":"Both approaches learn from repeated patient sessions, monitor coupled hemodynamic response during dialysis, and alter or recommend ultrafiltration settings to reduce hypotension while preserving fluid-removal goals.","remaining_difference":"Existing controllers found here steer one observed trajectory or apply a predefined biofeedback rule. The proposed testable difference is a patient-specific multivariable setting-to-physiology map decomposed with SVD, an outcome-sensitivity comparison across feasible controls, and explicit abstention when out-of-sample reconstruction fails. A direct comparison can determine whether this adds hypotension reduction without a clinically meaningful loss of delivered Kt/V.","classification":"POSSIBLE_DISTINCTION","disposition":"ADVANCE","rationale":"Adaptive dialysis control is established and materially narrows the concept, but the searched primary studies do not expose the proposed multivariable modal-control and residual-abstention combination. The distinction is concrete enough for comparative testing."},{"hypothesis_id":"H3","search_queries":["patient transfer network eigenvector centrality infection prevention targeting healthcare facilities","spectral centrality intervention allocation healthcare-associated infection facility transfer network","CDC patient transfer network infection prevention nursing homes regional official","targeting infection control patient sharing network centrality simulation"],"sources":[{"source_id":"H3-S1","title":"Influence of a patient transfer network of US inpatient facilities on the incidence of nosocomial infections","publisher":"Scientific Reports / PubMed Central","url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC5462812/","source_class":"PRIMARY_RESEARCH","claims_supported":["A nationwide inpatient-facility transfer network was linked to healthcare-associated infection incidence.","The study explicitly evaluated eigenvector-centrality targeting as a facility sensor-selection strategy.","Eigenvector centrality did not uniformly outperform simpler degree or random strategies, demonstrating that spectral rank is not automatically an effective intervention rule."]},{"source_id":"H3-S2","title":"Network Analysis of Intra-Hospital Transfers and Hospital-Onset Clostridium difficile Infection","publisher":"Infection Control & Hospital Epidemiology / PubMed Central","url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC7321830/","source_class":"PRIMARY_RESEARCH","claims_supported":["Eigenvector centrality was used to identify structurally important locations in a patient-transfer network.","The authors explicitly proposed targeting high-centrality, high-edge-weight locations for additional infection-prevention activity."]},{"source_id":"H3-S3","title":"MDRO Prevention Strategies","publisher":"Centers for Disease Control and Prevention","url":"https://www.cdc.gov/healthcare-associated-infections/php/preventing-mdros/mdro-prevention-strategies.html","source_class":"OFFICIAL_GUIDANCE","claims_supported":["CDC guidance directs jurisdictions to risk-stratify facilities by their predicted role in regional MDRO spread.","The guidance prioritizes resource-intensive prevention activities at influential and highly connected facilities within patient-sharing networks.","CDC describes screening and infection-control resource allocation as constrained by available capacity and informed by mathematical modeling."]}],"closest_analogue":"CDC's network-informed MDRO prevention allocation, with the closest mathematical implementation being eigenvector-centrality targeting in patient-transfer infection studies.","overlap":"The prior art contains the same facility-transfer operator, structural centrality ranking, infection-prevention objective, regional allocation boundary, and targeting of scarce surveillance or prevention effort toward influential network nodes.","remaining_difference":"A prospective constrained allocation comparison using a dominant-eigenvector score, explicit equity floors, and secondary-facility introductions as the endpoint may remain unevaluated. That is an evaluation and governance refinement; it does not separate the core mechanism from existing network-centrality targeting.","classification":"OBVIOUS_COLLISION","disposition":"REJECT","rationale":"Both primary research and current CDC guidance already connect patient-sharing structure to facility risk stratification and targeted infection-prevention allocation; eigenvector centrality has also been tested directly. The proposed core causal lever therefore collides at this deliberately shallow screen."},{"hypothesis_id":"H4","search_queries":["radiotherapy respiratory motion reduced order model SVD deformation adaptive radiotherapy dose calculation","principal component analysis respiratory motion model radiotherapy deformation dose reconstruction","adaptive radiotherapy low rank motion model online dose calculation thoracic","AAPM respiratory motion management radiotherapy official task group"],"sources":[{"source_id":"H4-S1","title":"A patient-specific respiratory model of anatomical motion for radiation treatment planning","publisher":"Medical Physics / Weill Cornell Medicine","url":"https://vivo.weill.cornell.edu/display/pubid18196805","source_class":"PRIMARY_RESEARCH","claims_supported":["Patient-specific thoracic deformation fields were reduced with principal-component analysis.","Two retained components represented respiratory organ motion in the reported patients.","The reduced model predicted respiratory anatomy for potential improvement of radiotherapy dose calculation and was tested on data acquired days later."]},{"source_id":"H4-S2","title":"Optimizing principal component models for representing interfraction variation in lung cancer radiotherapy","publisher":"Medical Physics / PubMed Central","url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC2945744/","source_class":"PRIMARY_RESEARCH","claims_supported":["Principal components were used to model dominant correlated tumor and lung deformation during radiotherapy.","The work addresses adaptive radiotherapy and dose evaluation in deforming anatomy.","Model reconstruction error against observed anatomy was explicitly calculated."]},{"source_id":"H4-S3","title":"The Management of Respiratory Motion in Radiation Oncology","publisher":"American Association of Physicists in Medicine","url":"https://www.aapm.org/pubs/reports/detail.asp?docid=92","source_class":"STANDARD","claims_supported":["AAPM provides a clinical process guide for respiratory motion management in thoracic and abdominal radiotherapy.","Recognized decision mechanisms include respiratory gating, synchronized treatment, breath hold, and motion-encompassing methods.","AAPM recommends patient-specific motion measurement and quality assurance when respiratory-motion technology is used."]}],"closest_analogue":"The patient-specific PCA respiratory-deformation model for radiotherapy planning, together with later PCA models that quantify residual reconstruction error for adaptive radiotherapy.","overlap":"The analogue already derives a low-dimensional basis from patient-specific respiratory deformation, reconstructs moving anatomy, supports radiotherapy dose calculation or adaptive evaluation, validates across treatment-time changes, and measures reconstruction residuals.","remaining_difference":"The proposed fixed performance targets—real-time gating or replanning selection, 40% faster calculation, a 2% organ-at-risk dose budget, and automatic recommendation rejection—could require new implementation and validation. They do not create a clear structural distinction from the established reduced respiratory-motion surrogate and residual-check workflow.","classification":"OBVIOUS_COLLISION","disposition":"REJECT","rationale":"Low-rank respiratory deformation surrogates for radiotherapy, including patient-specific PCA modes and reconstruction-error checks, are longstanding direct analogues. The remaining differences are chiefly workflow integration and acceptance thresholds."},{"hypothesis_id":"H5","search_queries":["post stroke gait principal component modes personalized rehabilitation therapy prescription perturbation","patient-specific gait modes stroke rehabilitation robotic assistance synergy targeting trial","stroke gait motor modules muscle synergies targeted rehabilitation intervention","site:clinicaltrials.gov stroke muscle synergy targeted gait rehabilitation trial"],"sources":[{"source_id":"H5-S1","title":"Patient-specific functional electrical stimulation strategy based on muscle synergy and walking posture analysis for gait rehabilitation of stroke patients","publisher":"Journal of International Medical Research / PubMed Central","url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC8161899/","source_class":"PRIMARY_RESEARCH","claims_supported":["Patient-specific gait muscle synergies were measured before treatment.","Impaired synergies and joint-angle patterns were used to construct individualized multichannel stimulation profiles.","The resulting gait-rehabilitation protocol was clinically evaluated in two stroke patients."]},{"source_id":"H5-S2","title":"Synergy-based functional electrical stimulation for gait rehabilitation in chronic stroke: a pilot study","publisher":"medRxiv","url":"https://www.medrxiv.org/content/10.1101/2025.05.21.25328035v3","source_class":"PRIMARY_RESEARCH","claims_supported":["A personalized multichannel stimulation intervention was based on each participant's motor-coordination impairment.","The intervention was paired with high-intensity gait training and compared with high-intensity training alone over six weeks.","Gait speed, endurance, biomechanics, and muscle synergies were evaluated."]},{"source_id":"H5-S3","title":"Tracking Neural Synergies After Stroke","publisher":"ClinicalTrials.gov, U.S. National Library of Medicine","url":"https://clinicaltrials.gov/study/NCT04805866","source_class":"OFFICIAL_ORGANIZATION_DATA","claims_supported":["A registered observational study measures and tracks post-stroke neural muscle synergies during gait rehabilitation.","The study uses repeated high-density EMG assessments during postural and walking tasks.","The registry record does not describe assigning therapy according to perturbation-derived outcome sensitivity."]}],"closest_analogue":"Patient-specific muscle-synergy-guided multichannel functional electrical stimulation for post-stroke gait rehabilitation.","overlap":"The analogue empirically extracts coupled gait or muscle modes from an individual stroke survivor, identifies impaired modes, maps them to individualized stimulation, delivers repeated gait therapy, and measures functional and modal outcomes.","remaining_difference":"The bounded unresolved question is whether any controlled study has safely perturbed each feasible assistance, resistance, or cueing input, estimated patient-specific outcome sensitivity and cross-coupling for each measured gait mode, and then prescribed the highest-leverage mode rather than matching stimulation to impairment or a healthy reference pattern.","classification":"INDETERMINATE_RESEARCH_NEEDED","disposition":"REJECT","rationale":"The core patient-specific mode-to-therapy concept is already very close to published synergy-guided rehabilitation. The shallow screen cannot determine whether the explicit perturbation-and-sensitivity selection rule has also been implemented, so an independent critic should research the bounded question before any advancement."}],"nominated_ids":["H1","H2"],"replenishment_recommended":false}