{"id":"483e0696641e432359580a5c03681df3739b96aede92da9f24b1be799d48f99d","title":"ICare-DT updates intensive-care forecasts from new patient data","argument":"In an arXiv paper, Yanke Li, Yahang Qi, Yumi Kim and Diego Paez-Granados present ICare-DT, a Transformer-based digital twin that updates a patient state hourly as new measurements arrive, without retraining. Trained on 140,473 eICU-CRD stays, it forecast vital signs crossing their training-set means; later forecasts outperformed the model’s own hour-6 forecasts across 12 test settings. Validation controls that withheld or swapped new hours traced the gain to each patient’s own measurements. Without refitting, ICare-DT retained gains in 8 of 9 settings on a second database, in an evaluation using a previously analyzed MIMIC-IV test split. The paper draws a boundary between forecasting and treatment-response prediction: its benchmark uses real records altered with simulated treatment doses and injected effects. Standard learners did not beat predicting zero effect for a dosing regime no patient had received; a separate learner recovered simulated effects when given the way those effects were generated. The authors present routine records as support for forecasts that improve with new data, and the benchmark as a testbed for learning how responses are generated. These results concern the paper’s datasets, forecast targets and simulated-response benchmark; they do not establish performance in clinical use.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The response-prediction benchmark replaces treatments in real records with simulated doses and injects responses of known size; it is not evidence of demonstrated treatment-response prediction in clinical use. The abstract reports forecasting vital signs crossing their training-set means and results on specified datasets and settings; it does not establish performance in clinical use.","speaker":{"name":"Li, Yanke, Qi, Yahang, Kim, Yumi, Paez-Granados, Diego","roleAtSource":"Article author"},"source":{"title":"ICare-DT: a digital twin for continuous forecasting of intensive care patients","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.09712","publishedAt":"2026-11-06","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors propose ICare-DT, a Transformer-based twin that updates a latent patient state hourly from new measurements without retraining.","excerpt":"We propose ICare-DT, a Transformer-based twin that updates a latent patient state hourly from new measurements without retraining.","url":"https://arxiv.org/abs/2611.09712","relationship":"Source-stated"},{"text":"On test data, later forecasts beat the twin’s own hour-6 forecasts in all 12 settings, with reported AUROC gains of 0.073 to 0.153.","excerpt":"On test data, later forecasts beat the twin's own hour-6 forecasts of the same windows in all 12 settings (+0.073 to +0.153 AUROC).","url":"https://arxiv.org/abs/2611.09712","relationship":"Source-stated"},{"text":"Validation controls withholding or swapping new hours traced the forecasting gain to the patient’s own measurements.","excerpt":"On validation data, controls that withheld or swapped the new hours traced the gain to the patient's own measurements.","url":"https://arxiv.org/abs/2611.09712","relationship":"Source-stated"},{"text":"Without refitting, the model retained gains in 8 of 9 settings on a second database, evaluated on a previously analyzed MIMIC-IV test split.","excerpt":"Without refitting, the model kept its gain on a second database (8 of 9 settings, +0.025 to +0.114 AUROC) in a prespecified evaluation of the previously analyzed MIMIC-IV test split.","url":"https://arxiv.org/abs/2611.09712","relationship":"Source-stated"},{"text":"For simulated treatment responses at an out-of-support query, no standard learner outperformed predicting zero effect; a separate learner recovered effects when told how they were generated.","excerpt":"No standard learner outperformed predicting zero effect at an out-of-support query. When told how the simulated responses were generated, a separate learner recovered the effects","url":"https://arxiv.org/abs/2611.09712","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.09712","checkedAt":"2026-11-09T06:52:06.919Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"What approaches, beyond being told how the simulated responses were generated, can predict effects for dosing regimes no patient has received?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"ICare-DT updates intensive-care forecasts from new patient data","summary":"In an arXiv paper, Yanke Li, Yahang Qi, Yumi Kim and Diego Paez-Granados present ICare-DT, a Transformer-based digital twin that updates a patient state hourly as new measurements arrive, without retraining. Trained on 140,473 eICU-CRD stays, it forecast vital signs crossing their training-set means; later forecasts outperformed the model’s own hour-6 forecasts across 12 test settings. Validation controls that withheld or swapped new hours traced the gain to each patient’s own measurements. Without refitting, ICare-DT retained gains in 8 of 9 settings on a second database, in an evaluation using a previously analyzed MIMIC-IV test split. The paper draws a boundary between forecasting and treatment-response prediction: its benchmark uses real records altered with simulated treatment doses and injected effects. Standard learners did not beat predicting zero effect for a dosing regime no patient had received; a separate learner recovered simulated effects when given the way those effects were generated. The authors present routine records as support for forecasts that improve with new data, and the benchmark as a testbed for learning how responses are generated. These results concern the paper’s datasets, forecast targets and simulated-response benchmark; they do not establish performance in clinical use.","question":"What approaches, beyond being told how the simulated responses were generated, can predict effects for dosing regimes no patient has received?","claims":[{"text":"The authors propose ICare-DT, a Transformer-based twin that updates a latent patient state hourly from new measurements without retraining.","excerpt":"We propose ICare-DT, a Transformer-based twin that updates a latent patient state hourly from new measurements without retraining.","support":"source-stated"},{"text":"On test data, later forecasts beat the twin’s own hour-6 forecasts in all 12 settings, with reported AUROC gains of 0.073 to 0.153.","excerpt":"On test data, later forecasts beat the twin's own hour-6 forecasts of the same windows in all 12 settings (+0.073 to +0.153 AUROC).","support":"source-stated"},{"text":"Validation controls withholding or swapping new hours traced the forecasting gain to the patient’s own measurements.","excerpt":"On validation data, controls that withheld or swapped the new hours traced the gain to the patient's own measurements.","support":"source-stated"},{"text":"Without refitting, the model retained gains in 8 of 9 settings on a second database, evaluated on a previously analyzed MIMIC-IV test split.","excerpt":"Without refitting, the model kept its gain on a second database (8 of 9 settings, +0.025 to +0.114 AUROC) in a prespecified evaluation of the previously analyzed MIMIC-IV test split.","support":"source-stated"},{"text":"For simulated treatment responses at an out-of-support query, no standard learner outperformed predicting zero effect; a separate learner recovered effects when told how they were generated.","excerpt":"No standard learner outperformed predicting zero effect at an out-of-support query. When told how the simulated responses were generated, a separate learner recovered the effects","support":"source-stated"}],"limitations":["The response-prediction benchmark replaces treatments in real records with simulated doses and injects responses of known size; it is not evidence of demonstrated treatment-response prediction in clinical use.","The abstract reports forecasting vital signs crossing their training-set means and results on specified datasets and settings; it does not establish performance in clinical use."],"mediaId":"23a001e44f4d39ed9f754381a6b0200ed3454a8710616a98ca22f83fe6700991","correctionReason":null,"id":"483e0696641e432359580a5c03681df3739b96aede92da9f24b1be799d48f99d","publicationId":"a2bfe5f0-3a6e-42dc-8dfe-59f76028385d","version":1,"sourceUrl":"https://arxiv.org/abs/2611.09712","sourceTitle":"ICare-DT: a digital twin for continuous forecasting of intensive care patients","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-06","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-09T05:07:20.937Z","createdAt":"2026-11-09T05:49:29.033Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"23a001e44f4d39ed9f754381a6b0200ed3454a8710616a98ca22f83fe6700991","kind":"article","url":"https://arxiv.org/abs/2611.09712","sourceUrl":"https://arxiv.org/abs/2611.09712","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-09T05:48:26.601Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.09712","checkedAt":"2026-11-09T06:52:06.919Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"23a001e44f4d39ed9f754381a6b0200ed3454a8710616a98ca22f83fe6700991","confidence":0.86,"provider":"typesafe","model":"jev-1.13.0","inputHash":"4fd5a5efb4a4701c37f3b2d2b79b224052e43cd61c59a48e2e3e4e17bf3e0750","decidedAt":"2026-11-09T05:48:28.388Z"},"provenance":{"textHash":"835a04c9ccd15cc45f333fed062ababa0b1263040d6cd3f55e9079e03e7b2f21","observationId":"df6c668aea9340a22caef6bfe669f8d24ea3a0369d6136473ed9bc46b85eb001","retrievedAt":"2026-11-09T05:48:26.315Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"bf0f4084-f6d2-4eea-8596-d49d38e13031"},"reviewer":{"model":"gpt-6-luna","runId":"6851c2c6-8d71-4caa-8607-8310f13ca0d0"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-09T05:49:29.033Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Li, Yanke","url":null,"kind":"unknown"},{"name":"Qi, Yahang","url":null,"kind":"unknown"},{"name":"Kim, Yumi","url":null,"kind":"unknown"},{"name":"Paez-Granados, Diego","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Li, Yanke","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Li, Yanke","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.09712","locator":"head/meta[name=author|citation_author]"},"assertionId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-09T06:52:06.919Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7","sourceUrl":"https://arxiv.org/abs/2611.09712","contentSha256":"1b4c57b2397c3919ca1a4bc05d40f9cd7c4c95ab0973ba6ced4efda4a784eb53","textHash":"835a04c9ccd15cc45f333fed062ababa0b1263040d6cd3f55e9079e03e7b2f21","observedAt":"2026-11-09T06:52:06.919Z"}},{"actor":{"name":"Qi, Yahang","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Qi, Yahang","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.09712","locator":"head/meta[name=author|citation_author]"},"assertionId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-09T06:52:06.919Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7","sourceUrl":"https://arxiv.org/abs/2611.09712","contentSha256":"1b4c57b2397c3919ca1a4bc05d40f9cd7c4c95ab0973ba6ced4efda4a784eb53","textHash":"835a04c9ccd15cc45f333fed062ababa0b1263040d6cd3f55e9079e03e7b2f21","observedAt":"2026-11-09T06:52:06.919Z"}},{"actor":{"name":"Kim, Yumi","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Kim, Yumi","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.09712","locator":"head/meta[name=author|citation_author]"},"assertionId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-09T06:52:06.919Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7","sourceUrl":"https://arxiv.org/abs/2611.09712","contentSha256":"1b4c57b2397c3919ca1a4bc05d40f9cd7c4c95ab0973ba6ced4efda4a784eb53","textHash":"835a04c9ccd15cc45f333fed062ababa0b1263040d6cd3f55e9079e03e7b2f21","observedAt":"2026-11-09T06:52:06.919Z"}},{"actor":{"name":"Paez-Granados, Diego","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Paez-Granados, Diego","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.09712","locator":"head/meta[name=author|citation_author]"},"assertionId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-09T06:52:06.919Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7","sourceUrl":"https://arxiv.org/abs/2611.09712","contentSha256":"1b4c57b2397c3919ca1a4bc05d40f9cd7c4c95ab0973ba6ced4efda4a784eb53","textHash":"835a04c9ccd15cc45f333fed062ababa0b1263040d6cd3f55e9079e03e7b2f21","observedAt":"2026-11-09T06:52:06.919Z"}}],"participationCoverage":{"total":4,"returned":4,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.09712","snapshotId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7","contentSha256":"1b4c57b2397c3919ca1a4bc05d40f9cd7c4c95ab0973ba6ced4efda4a784eb53","textHash":"835a04c9ccd15cc45f333fed062ababa0b1263040d6cd3f55e9079e03e7b2f21","observedAt":"2026-11-09T06:52:06.919Z","basis":"publisher-page-metadata","sourceLinks":44,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/483e0696641e432359580a5c03681df3739b96aede92da9f24b1be799d48f99d/source"}},"canonicalUrl":"https://escapemeridian.com/readings/icare-dt-updates-intensive-care-forecasts-from-new-patient-data-483e0696","formats":{"html":"/readings/icare-dt-updates-intensive-care-forecasts-from-new-patient-data-483e0696","markdown":"/readings/icare-dt-updates-intensive-care-forecasts-from-new-patient-data-483e0696.md","json":"/readings/icare-dt-updates-intensive-care-forecasts-from-new-patient-data-483e0696.json"},"attribution":"Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.","maintainedGuideLinks":[],"readerContext":{"version":"reader-context-v1","attribution":"Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.","terms":[{"id":"term-transformer","name":"Transformer architecture","explanation":"The neural-network design behind most current language models; it weighs how each part of the input relates to the others.","referenceUrl":"https://developers.google.com/machine-learning/glossary#transformer"},{"id":"term-benchmark","name":"Benchmark","explanation":"A defined set of tests used to compare systems. Results depend on the tasks and measures selected.","referenceUrl":"https://arxiv.org/abs/2211.09110"}],"entities":[{"id":"arxiv","name":"arXiv","kind":"organization","context":"arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.","referenceUrls":["https://arxiv.org/","https://info.arxiv.org/help/api/index.html"],"path":"/entities/arxiv","json":"/entities/arxiv.json","markdown":"/entities/arxiv.md"}]},"researchQuestions":[],"withdrawn":false,"references":{"schemaVersion":1,"target":{"type":"reading","id":"483e0696641e432359580a5c03681df3739b96aede92da9f24b1be799d48f99d","revision":"a2bfe5f0-3a6e-42dc-8dfe-59f76028385d","revisionKind":"publication-id","title":"ICare-DT updates intensive-care forecasts from new patient data","url":"https://escapemeridian.com/readings/icare-dt-updates-intensive-care-forecasts-from-new-patient-data-483e0696"},"source":{"type":"source","id":"https://arxiv.org/abs/2611.09712","revision":"835a04c9ccd15cc45f333fed062ababa0b1263040d6cd3f55e9079e03e7b2f21","revisionKind":"sha256-source-text","title":"ICare-DT: a digital twin for continuous forecasting of intensive care patients","url":"https://arxiv.org/abs/2611.09712"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter."},"revisions":[{"id":"483e0696641e432359580a5c03681df3739b96aede92da9f24b1be799d48f99d","title":"ICare-DT updates intensive-care forecasts from new patient data","argument":"In an arXiv paper, Yanke Li, Yahang Qi, Yumi Kim and Diego Paez-Granados present ICare-DT, a Transformer-based digital twin that updates a patient state hourly as new measurements arrive, without retraining. Trained on 140,473 eICU-CRD stays, it forecast vital signs crossing their training-set means; later forecasts outperformed the model’s own hour-6 forecasts across 12 test settings. Validation controls that withheld or swapped new hours traced the gain to each patient’s own measurements. Without refitting, ICare-DT retained gains in 8 of 9 settings on a second database, in an evaluation using a previously analyzed MIMIC-IV test split. The paper draws a boundary between forecasting and treatment-response prediction: its benchmark uses real records altered with simulated treatment doses and injected effects. Standard learners did not beat predicting zero effect for a dosing regime no patient had received; a separate learner recovered simulated effects when given the way those effects were generated. The authors present routine records as support for forecasts that improve with new data, and the benchmark as a testbed for learning how responses are generated. These results concern the paper’s datasets, forecast targets and simulated-response benchmark; they do not establish performance in clinical use.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The response-prediction benchmark replaces treatments in real records with simulated doses and injects responses of known size; it is not evidence of demonstrated treatment-response prediction in clinical use. The abstract reports forecasting vital signs crossing their training-set means and results on specified datasets and settings; it does not establish performance in clinical use.","speaker":{"name":"Li, Yanke, Qi, Yahang, Kim, Yumi, Paez-Granados, Diego","roleAtSource":"Article author"},"source":{"title":"ICare-DT: a digital twin for continuous forecasting of intensive care patients","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.09712","publishedAt":"2026-11-06","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors propose ICare-DT, a Transformer-based twin that updates a latent patient state hourly from new measurements without retraining.","excerpt":"We propose ICare-DT, a Transformer-based twin that updates a latent patient state hourly from new measurements without retraining.","url":"https://arxiv.org/abs/2611.09712","relationship":"Source-stated"},{"text":"On test data, later forecasts beat the twin’s own hour-6 forecasts in all 12 settings, with reported AUROC gains of 0.073 to 0.153.","excerpt":"On test data, later forecasts beat the twin's own hour-6 forecasts of the same windows in all 12 settings (+0.073 to +0.153 AUROC).","url":"https://arxiv.org/abs/2611.09712","relationship":"Source-stated"},{"text":"Validation controls withholding or swapping new hours traced the forecasting gain to the patient’s own measurements.","excerpt":"On validation data, controls that withheld or swapped the new hours traced the gain to the patient's own measurements.","url":"https://arxiv.org/abs/2611.09712","relationship":"Source-stated"},{"text":"Without refitting, the model retained gains in 8 of 9 settings on a second database, evaluated on a previously analyzed MIMIC-IV test split.","excerpt":"Without refitting, the model kept its gain on a second database (8 of 9 settings, +0.025 to +0.114 AUROC) in a prespecified evaluation of the previously analyzed MIMIC-IV test split.","url":"https://arxiv.org/abs/2611.09712","relationship":"Source-stated"},{"text":"For simulated treatment responses at an out-of-support query, no standard learner outperformed predicting zero effect; a separate learner recovered effects when told how they were generated.","excerpt":"No standard learner outperformed predicting zero effect at an out-of-support query. When told how the simulated responses were generated, a separate learner recovered the effects","url":"https://arxiv.org/abs/2611.09712","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.09712","checkedAt":"2026-11-09T06:52:06.919Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"What approaches, beyond being told how the simulated responses were generated, can predict effects for dosing regimes no patient has received?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"ICare-DT updates intensive-care forecasts from new patient data","summary":"In an arXiv paper, Yanke Li, Yahang Qi, Yumi Kim and Diego Paez-Granados present ICare-DT, a Transformer-based digital twin that updates a patient state hourly as new measurements arrive, without retraining. Trained on 140,473 eICU-CRD stays, it forecast vital signs crossing their training-set means; later forecasts outperformed the model’s own hour-6 forecasts across 12 test settings. Validation controls that withheld or swapped new hours traced the gain to each patient’s own measurements. Without refitting, ICare-DT retained gains in 8 of 9 settings on a second database, in an evaluation using a previously analyzed MIMIC-IV test split. The paper draws a boundary between forecasting and treatment-response prediction: its benchmark uses real records altered with simulated treatment doses and injected effects. Standard learners did not beat predicting zero effect for a dosing regime no patient had received; a separate learner recovered simulated effects when given the way those effects were generated. The authors present routine records as support for forecasts that improve with new data, and the benchmark as a testbed for learning how responses are generated. These results concern the paper’s datasets, forecast targets and simulated-response benchmark; they do not establish performance in clinical use.","question":"What approaches, beyond being told how the simulated responses were generated, can predict effects for dosing regimes no patient has received?","claims":[{"text":"The authors propose ICare-DT, a Transformer-based twin that updates a latent patient state hourly from new measurements without retraining.","excerpt":"We propose ICare-DT, a Transformer-based twin that updates a latent patient state hourly from new measurements without retraining.","support":"source-stated"},{"text":"On test data, later forecasts beat the twin’s own hour-6 forecasts in all 12 settings, with reported AUROC gains of 0.073 to 0.153.","excerpt":"On test data, later forecasts beat the twin's own hour-6 forecasts of the same windows in all 12 settings (+0.073 to +0.153 AUROC).","support":"source-stated"},{"text":"Validation controls withholding or swapping new hours traced the forecasting gain to the patient’s own measurements.","excerpt":"On validation data, controls that withheld or swapped the new hours traced the gain to the patient's own measurements.","support":"source-stated"},{"text":"Without refitting, the model retained gains in 8 of 9 settings on a second database, evaluated on a previously analyzed MIMIC-IV test split.","excerpt":"Without refitting, the model kept its gain on a second database (8 of 9 settings, +0.025 to +0.114 AUROC) in a prespecified evaluation of the previously analyzed MIMIC-IV test split.","support":"source-stated"},{"text":"For simulated treatment responses at an out-of-support query, no standard learner outperformed predicting zero effect; a separate learner recovered effects when told how they were generated.","excerpt":"No standard learner outperformed predicting zero effect at an out-of-support query. When told how the simulated responses were generated, a separate learner recovered the effects","support":"source-stated"}],"limitations":["The response-prediction benchmark replaces treatments in real records with simulated doses and injects responses of known size; it is not evidence of demonstrated treatment-response prediction in clinical use.","The abstract reports forecasting vital signs crossing their training-set means and results on specified datasets and settings; it does not establish performance in clinical use."],"mediaId":"23a001e44f4d39ed9f754381a6b0200ed3454a8710616a98ca22f83fe6700991","correctionReason":null,"id":"483e0696641e432359580a5c03681df3739b96aede92da9f24b1be799d48f99d","publicationId":"a2bfe5f0-3a6e-42dc-8dfe-59f76028385d","version":1,"sourceUrl":"https://arxiv.org/abs/2611.09712","sourceTitle":"ICare-DT: a digital twin for continuous forecasting of intensive care patients","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-06","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-09T05:07:20.937Z","createdAt":"2026-11-09T05:49:29.033Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"23a001e44f4d39ed9f754381a6b0200ed3454a8710616a98ca22f83fe6700991","kind":"article","url":"https://arxiv.org/abs/2611.09712","sourceUrl":"https://arxiv.org/abs/2611.09712","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-09T05:48:26.601Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.09712","checkedAt":"2026-11-09T06:52:06.919Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"23a001e44f4d39ed9f754381a6b0200ed3454a8710616a98ca22f83fe6700991","confidence":0.86,"provider":"typesafe","model":"jev-1.13.0","inputHash":"4fd5a5efb4a4701c37f3b2d2b79b224052e43cd61c59a48e2e3e4e17bf3e0750","decidedAt":"2026-11-09T05:48:28.388Z"},"provenance":{"textHash":"835a04c9ccd15cc45f333fed062ababa0b1263040d6cd3f55e9079e03e7b2f21","observationId":"df6c668aea9340a22caef6bfe669f8d24ea3a0369d6136473ed9bc46b85eb001","retrievedAt":"2026-11-09T05:48:26.315Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"bf0f4084-f6d2-4eea-8596-d49d38e13031"},"reviewer":{"model":"gpt-6-luna","runId":"6851c2c6-8d71-4caa-8607-8310f13ca0d0"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-09T05:49:29.033Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Li, Yanke","url":null,"kind":"unknown"},{"name":"Qi, Yahang","url":null,"kind":"unknown"},{"name":"Kim, Yumi","url":null,"kind":"unknown"},{"name":"Paez-Granados, Diego","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Li, Yanke","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Li, Yanke","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.09712","locator":"head/meta[name=author|citation_author]"},"assertionId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-09T06:52:06.919Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7","sourceUrl":"https://arxiv.org/abs/2611.09712","contentSha256":"1b4c57b2397c3919ca1a4bc05d40f9cd7c4c95ab0973ba6ced4efda4a784eb53","textHash":"835a04c9ccd15cc45f333fed062ababa0b1263040d6cd3f55e9079e03e7b2f21","observedAt":"2026-11-09T06:52:06.919Z"}},{"actor":{"name":"Qi, Yahang","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Qi, Yahang","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.09712","locator":"head/meta[name=author|citation_author]"},"assertionId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-09T06:52:06.919Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7","sourceUrl":"https://arxiv.org/abs/2611.09712","contentSha256":"1b4c57b2397c3919ca1a4bc05d40f9cd7c4c95ab0973ba6ced4efda4a784eb53","textHash":"835a04c9ccd15cc45f333fed062ababa0b1263040d6cd3f55e9079e03e7b2f21","observedAt":"2026-11-09T06:52:06.919Z"}},{"actor":{"name":"Kim, Yumi","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Kim, Yumi","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.09712","locator":"head/meta[name=author|citation_author]"},"assertionId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-09T06:52:06.919Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7","sourceUrl":"https://arxiv.org/abs/2611.09712","contentSha256":"1b4c57b2397c3919ca1a4bc05d40f9cd7c4c95ab0973ba6ced4efda4a784eb53","textHash":"835a04c9ccd15cc45f333fed062ababa0b1263040d6cd3f55e9079e03e7b2f21","observedAt":"2026-11-09T06:52:06.919Z"}},{"actor":{"name":"Paez-Granados, Diego","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Paez-Granados, Diego","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.09712","locator":"head/meta[name=author|citation_author]"},"assertionId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-09T06:52:06.919Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7","sourceUrl":"https://arxiv.org/abs/2611.09712","contentSha256":"1b4c57b2397c3919ca1a4bc05d40f9cd7c4c95ab0973ba6ced4efda4a784eb53","textHash":"835a04c9ccd15cc45f333fed062ababa0b1263040d6cd3f55e9079e03e7b2f21","observedAt":"2026-11-09T06:52:06.919Z"}}],"participationCoverage":{"total":4,"returned":4,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.09712","snapshotId":"14f9e85ff34ecbb9f711fbe7020fb0b326ac08f7e867994eb43c6ea9203420a7","contentSha256":"1b4c57b2397c3919ca1a4bc05d40f9cd7c4c95ab0973ba6ced4efda4a784eb53","textHash":"835a04c9ccd15cc45f333fed062ababa0b1263040d6cd3f55e9079e03e7b2f21","observedAt":"2026-11-09T06:52:06.919Z","basis":"publisher-page-metadata","sourceLinks":44,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/483e0696641e432359580a5c03681df3739b96aede92da9f24b1be799d48f99d/source"}}}],"audits":[{"publicationId":"a2bfe5f0-3a6e-42dc-8dfe-59f76028385d","verdict":"uncertain","findings":[{"dimension":"attribution","status":"pass","explanation":"The paper and its claims are correctly attributed to the four listed authors on arXiv. Results are framed as what the paper reports, not as established fact.","excerpt":"Authors:Yanke Li, Yahang Qi, Yumi Kim, Diego Paez-Granados"},{"dimension":"summary","status":"uncertain","explanation":"The reading is mostly faithful but hard for a non-specialist to follow. Central terms such as digital twin, Transformer, eICU-CRD, MIMIC-IV, out-of-support and the AUROC metric are never explained; tooltips are needed. It also drops the condition that the separate learner recovered effects without seeing their true values, and only partly (R2 0.30 to 0.40 at one effect size). Without that, the result can read as trivial. Smallest repair: restore that condition and the partial R2. It also omits the refitted MIMIC-IV gradient-boosting results, including the acute kidney injury gain, and the prespecified nature of the evaluation.","excerpt":"a separate learner recovered the effects from the same records without seeing their true values (R2 0.30 to 0.40 at one effect size)"},{"dimension":"date","status":"pass","explanation":"The originalPublished date of 2026-11-06 matches the arXiv v1 submission date. It is not the later feed date.","excerpt":"[Submitted on 6 Nov 2026]"},{"dimension":"support","status":"pass","explanation":"Every claim excerpt is an exact substring of the abstract and supports its claim. The limitations correctly confine the results to the paper's datasets and its simulated benchmark, without implying clinical benefit. The question follows from the reported out-of-support failure and is not a disguised claim.","excerpt":"No standard learner outperformed predicting zero effect at an out-of-support query."},{"dimension":"media","status":"pass","explanation":"The selected media is the verified direct arXiv abstract page for this paper. It is the most direct way for a reader to reach the source.","excerpt":null}],"createdAt":"2026-11-09T05:49:29.033Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}