Machine learning based prediction of chronic shunt-dependent hydrocephalus after spontaneous subarachnoid hemorrhage
Chronic posthemorrhagic hydrocephalus often arises following spontaneous subarachnoid hemorrhage (SAH). Timely identification of patients predisposed to develop chronic shunt-dependent hydrocephalus may significantly enhance clinical outcomes. We performed an analysis of 510 SAH-patients treated at...
Saved in:
Published in | World neurosurgery |
---|---|
Main Authors | , , , , , , , , , , |
Format | Journal Article |
Language | English |
Published |
Elsevier Inc
12.09.2024
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Chronic posthemorrhagic hydrocephalus often arises following spontaneous subarachnoid hemorrhage (SAH). Timely identification of patients predisposed to develop chronic shunt-dependent hydrocephalus may significantly enhance clinical outcomes.
We performed an analysis of 510 SAH-patients treated at our institution between 2013 and 2018. Clinical and radiological variables, including age, sex, Hunt & Hess grade, Fisher-Score, external ventricular drainage placement, central nervous system infection, aneurysm characteristics, and treatment modalities, were evaluated. Supervised machine learning models, trained and compared using Python and scikit-learn, were employed to predict chronic shunt-dependent hydrocephalus. Model performance was rigorously assessed through repeated cross-validation. To facilitate transparency and collaboration, we publicly released the dataset and code on GitHub (https://github.com/RISCSoftware/shuntclf) and developed an interactive web application (https://huggingface.co/spaces/risc42/shuntclf).
Among the evaluated machine learning models, logistic regression exhibited superior performance, with an AUC-ROC of 0.819 and an AUC-PR of 0.482, along with the highest F1 score of 0.473. Although the balanced accuracy scores of the models were generally proximate, ranging from 0.735 to 0.764, logistic regression consistently outperform others in key metrics such as AUC-ROC and AUC-PR. Conversely, female gender and absence of aneurysm within the anterior communicating artery were associated with reduced shunt requirement likelihood.
Machine learning models, including logistic regression, demonstrate strong predictive capability for early chronic shunt-dependent hydrocephalus following spontaneous SAH, which may potentially contribute to more timely shunt placement interventions. This predictive capability is supported by our web interface, which simplifies the application of these models, aiding clinicians in efficiently determining the need for shunt placement. |
---|---|
AbstractList | Chronic posthemorrhagic hydrocephalus often arises following spontaneous subarachnoid hemorrhage (SAH). Timely identification of patients predisposed to develop chronic shunt-dependent hydrocephalus may significantly enhance clinical outcomes.
We performed an analysis of 510 SAH-patients treated at our institution between 2013 and 2018. Clinical and radiological variables, including age, sex, Hunt & Hess grade, Fisher-Score, external ventricular drainage placement, central nervous system infection, aneurysm characteristics, and treatment modalities, were evaluated. Supervised machine learning models, trained and compared using Python and scikit-learn, were employed to predict chronic shunt-dependent hydrocephalus. Model performance was rigorously assessed through repeated cross-validation. To facilitate transparency and collaboration, we publicly released the dataset and code on GitHub (https://github.com/RISCSoftware/shuntclf) and developed an interactive web application (https://huggingface.co/spaces/risc42/shuntclf).
Among the evaluated machine learning models, logistic regression exhibited superior performance, with an AUC-ROC of 0.819 and an AUC-PR of 0.482, along with the highest F1 score of 0.473. Although the balanced accuracy scores of the models were generally proximate, ranging from 0.735 to 0.764, logistic regression consistently outperform others in key metrics such as AUC-ROC and AUC-PR. Conversely, female gender and absence of aneurysm within the anterior communicating artery were associated with reduced shunt requirement likelihood.
Machine learning models, including logistic regression, demonstrate strong predictive capability for early chronic shunt-dependent hydrocephalus following spontaneous SAH, which may potentially contribute to more timely shunt placement interventions. This predictive capability is supported by our web interface, which simplifies the application of these models, aiding clinicians in efficiently determining the need for shunt placement. Chronic posthemorrhagic hydrocephalus often arises following spontaneous subarachnoid hemorrhage (SAH). Timely identification of patients predisposed to develop chronic shunt-dependent hydrocephalus may significantly enhance clinical outcomes.BACKGROUNDChronic posthemorrhagic hydrocephalus often arises following spontaneous subarachnoid hemorrhage (SAH). Timely identification of patients predisposed to develop chronic shunt-dependent hydrocephalus may significantly enhance clinical outcomes.We performed an analysis of 510 SAH-patients treated at our institution between 2013 and 2018. Clinical and radiological variables, including age, sex, Hunt & Hess grade, Fisher-Score, external ventricular drainage placement, central nervous system infection, aneurysm characteristics, and treatment modalities, were evaluated. Supervised machine learning models, trained and compared using Python and scikit-learn, were employed to predict chronic shunt-dependent hydrocephalus. Model performance was rigorously assessed through repeated cross-validation. To facilitate transparency and collaboration, we publicly released the dataset and code on GitHub (https://github.com/RISCSoftware/shuntclf) and developed an interactive web application (https://huggingface.co/spaces/risc42/shuntclf).METHODSWe performed an analysis of 510 SAH-patients treated at our institution between 2013 and 2018. Clinical and radiological variables, including age, sex, Hunt & Hess grade, Fisher-Score, external ventricular drainage placement, central nervous system infection, aneurysm characteristics, and treatment modalities, were evaluated. Supervised machine learning models, trained and compared using Python and scikit-learn, were employed to predict chronic shunt-dependent hydrocephalus. Model performance was rigorously assessed through repeated cross-validation. To facilitate transparency and collaboration, we publicly released the dataset and code on GitHub (https://github.com/RISCSoftware/shuntclf) and developed an interactive web application (https://huggingface.co/spaces/risc42/shuntclf).Among the evaluated machine learning models, logistic regression exhibited superior performance, with an AUC-ROC of 0.819 and an AUC-PR of 0.482, along with the highest F1 score of 0.473. Although the balanced accuracy scores of the models were generally proximate, ranging from 0.735 to 0.764, logistic regression consistently outperform others in key metrics such as AUC-ROC and AUC-PR. Conversely, female gender and absence of aneurysm within the anterior communicating artery were associated with reduced shunt requirement likelihood.RESULTSAmong the evaluated machine learning models, logistic regression exhibited superior performance, with an AUC-ROC of 0.819 and an AUC-PR of 0.482, along with the highest F1 score of 0.473. Although the balanced accuracy scores of the models were generally proximate, ranging from 0.735 to 0.764, logistic regression consistently outperform others in key metrics such as AUC-ROC and AUC-PR. Conversely, female gender and absence of aneurysm within the anterior communicating artery were associated with reduced shunt requirement likelihood.Machine learning models, including logistic regression, demonstrate strong predictive capability for early chronic shunt-dependent hydrocephalus following spontaneous SAH, which may potentially contribute to more timely shunt placement interventions. This predictive capability is supported by our web interface, which simplifies the application of these models, aiding clinicians in efficiently determining the need for shunt placement.CONCLUSIONMachine learning models, including logistic regression, demonstrate strong predictive capability for early chronic shunt-dependent hydrocephalus following spontaneous SAH, which may potentially contribute to more timely shunt placement interventions. This predictive capability is supported by our web interface, which simplifies the application of these models, aiding clinicians in efficiently determining the need for shunt placement. |
Author | Steindl, Markus Sonnberger, Michael Gollwitzer, Maria Aspalter, Stefan Hauser, Anna Sardi, Gracija Gmeiner, Matthias Rauch, Philip Gruber, Andreas Stroh, Nico Rossmann, Tobias |
Author_xml | – sequence: 1 givenname: Maria surname: Gollwitzer fullname: Gollwitzer, Maria organization: Department of Neurosurgery, Kepler University Hospital, Johannes Kepler University Linz, Austria – sequence: 2 givenname: Markus orcidid: 0009-0008-8088-2264 surname: Steindl fullname: Steindl, Markus organization: Johannes Kepler University Linz, Austria – sequence: 3 givenname: Nico surname: Stroh fullname: Stroh, Nico email: nico.stroh@kepleruniklinikum.at organization: Department of Neurosurgery, Kepler University Hospital, Johannes Kepler University Linz, Austria – sequence: 4 givenname: Anna surname: Hauser fullname: Hauser, Anna organization: Department of Neurosurgery, Kepler University Hospital, Johannes Kepler University Linz, Austria – sequence: 5 givenname: Gracija surname: Sardi fullname: Sardi, Gracija organization: Department of Neurosurgery, Kepler University Hospital, Johannes Kepler University Linz, Austria – sequence: 6 givenname: Tobias surname: Rossmann fullname: Rossmann, Tobias organization: Department of Neurosurgery, Kepler University Hospital, Johannes Kepler University Linz, Austria – sequence: 7 givenname: Stefan orcidid: 0000-0003-0782-0311 surname: Aspalter fullname: Aspalter, Stefan organization: Department of Neurosurgery, Kepler University Hospital, Johannes Kepler University Linz, Austria – sequence: 8 givenname: Philip surname: Rauch fullname: Rauch, Philip organization: Department of Neurosurgery, Kepler University Hospital, Johannes Kepler University Linz, Austria – sequence: 9 givenname: Michael orcidid: 0000-0003-2342-6365 surname: Sonnberger fullname: Sonnberger, Michael organization: Institute of Neuroradiology, Kepler University Hospital Linz, Austria – sequence: 10 givenname: Andreas surname: Gruber fullname: Gruber, Andreas organization: Department of Neurosurgery, Kepler University Hospital, Johannes Kepler University Linz, Austria – sequence: 11 givenname: Matthias surname: Gmeiner fullname: Gmeiner, Matthias organization: Department of Neurosurgery, Kepler University Hospital, Johannes Kepler University Linz, Austria |
BookMark | eNp9kEtLxDAQx4MouD6-gKccvbTm0W1S8CLiCxQveg7ZZLLN0k1q0ip-e7OseHQuMwz__zx-J-gwxAAIXVBSU0Lbq039FWCuGWFNTbqaNOIALagUspKi7Q7_6iU5Ruc5b0gJThsp-AJNL9r0PgAeQKfgwxqvdAaLxwTWm8nHgKPDpk8xeINzP4epsjBCsBAm3H_bFA2MvR7mjLWbIOE8xjDpALF08rzSqSwI0Vvcwzam1Os1nKEjp4cM57_5FL3f373dPlbPrw9PtzfPlaGCiwqoE61cMgtdyySVDbdSNEbzZSOpY7LTjAi26gyxxjlNORjrGDBqKHBCWn6KLvdzxxQ_ZsiT2vpsYBj25ylOSRnVMcKLlO2lJsWcEzg1Jr_V6VtRonaU1UbtKKsdZUU6VSgX0_XeBOWJTw9JZeMhmIIugZmUjf4_-w8oWYp2 |
ContentType | Journal Article |
Copyright | 2024 Copyright © 2024. Published by Elsevier Inc. |
Copyright_xml | – notice: 2024 – notice: Copyright © 2024. Published by Elsevier Inc. |
DBID | AAYXX CITATION 7X8 |
DOI | 10.1016/j.wneu.2024.09.047 |
DatabaseName | CrossRef MEDLINE - Academic |
DatabaseTitle | CrossRef MEDLINE - Academic |
DatabaseTitleList | MEDLINE - Academic |
DeliveryMethod | fulltext_linktorsrc |
EISSN | 1878-8769 |
ExternalDocumentID | 10_1016_j_wneu_2024_09_047 S1878875024015845 |
GroupedDBID | --- --K --M .1- .FO .~1 0R~ 1B1 1P~ 1~. 1~5 4.4 457 4G. 53G 5VS 7-5 71M 8P~ 9JM AACTN AAEDT AAEDW AAIKJ AAKOC AALRI AAOAW AAQFI AAQQT AAXKI AAXLA AAXUO ABBQC ABCQJ ABFNM ABFRF ABJNI ABLJU ABMAC ABMZM ABXDB ACDAQ ACGFS ACIUM ACRLP ADBBV ADEZE ADMUD AEBSH AEFWE AEKER AENEX AEVXI AFCTW AFKWA AFRHN AFTJW AFXIZ AGHFR AGUBO AGWIK AGYEJ AIEXJ AIKHN AITUG AJOXV AJRQY AJUYK AKRWK ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ ANZVX AXJTR BKOJK BLXMC BNPGV EBS EFJIC EJD EP3 F5P FDB FEDTE FIRID FNPLU FYGXN GBLVA HVGLF HZ~ J1W KOM M41 MO0 MOBAO N9A O-L O9- OAUVE OQ- OZT P-8 P-9 PC. PG~ Q38 RIG ROL SCC SDF SDP SEL SES SPCBC SSH SSN SSZ T5K Z5R ~G- AAYXX CITATION 7X8 |
ID | FETCH-LOGICAL-c1737-e1f76852de96281843d874ca35481f289a2072b9c0dcffa13ecdf2e21c1e30063 |
IEDL.DBID | .~1 |
ISSN | 1878-8750 1878-8769 |
IngestDate | Mon Sep 16 17:53:34 EDT 2024 Wed Sep 18 12:55:36 EDT 2024 Sat Sep 14 18:13:15 EDT 2024 |
IsDoiOpenAccess | false |
IsOpenAccess | true |
IsPeerReviewed | true |
IsScholarly | true |
Keywords | SVC PR hydrocephalus shunt-dependency CNS KNN SVM AUC bACC SMOTE SDASH NPV CSF PPV TPR AICA TNR ACom ACA OR ICA ROC MCA F1 VA AUC-ROC PCA outcome prediction SCA CHESS EVD PICA CV PCom subarachnoid hemorrhage Machine learning AUC-PR SAH BNI BA |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-c1737-e1f76852de96281843d874ca35481f289a2072b9c0dcffa13ecdf2e21c1e30063 |
Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
ORCID | 0009-0008-8088-2264 0000-0003-2342-6365 0000-0003-0782-0311 |
OpenAccessLink | https://doi.org/10.1016/j.wneu.2024.09.047 |
PQID | 3105489203 |
PQPubID | 23479 |
ParticipantIDs | proquest_miscellaneous_3105489203 crossref_primary_10_1016_j_wneu_2024_09_047 elsevier_sciencedirect_doi_10_1016_j_wneu_2024_09_047 |
PublicationCentury | 2000 |
PublicationDate | 20240912 |
PublicationDateYYYYMMDD | 2024-09-12 |
PublicationDate_xml | – month: 09 year: 2024 text: 20240912 day: 12 |
PublicationDecade | 2020 |
PublicationTitle | World neurosurgery |
PublicationYear | 2024 |
Publisher | Elsevier Inc |
Publisher_xml | – name: Elsevier Inc |
SSID | ssj0000314873 |
Score | 2.4465077 |
Snippet | Chronic posthemorrhagic hydrocephalus often arises following spontaneous subarachnoid hemorrhage (SAH). Timely identification of patients predisposed to... |
SourceID | proquest crossref elsevier |
SourceType | Aggregation Database Publisher |
SubjectTerms | hydrocephalus Machine learning outcome prediction shunt-dependency subarachnoid hemorrhage |
Title | Machine learning based prediction of chronic shunt-dependent hydrocephalus after spontaneous subarachnoid hemorrhage |
URI | https://dx.doi.org/10.1016/j.wneu.2024.09.047 https://www.proquest.com/docview/3105489203/abstract/ |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1LS8NAEF5KvXgRRcX6KCt4k7XZzftYiqUq7UULvS3JPkxF0tImiBd_uzN5CIp48JaEbEi-3cw3s_Mi5IoL6wTKeEwBmzMPCJCBHRIxbUJrbeBrJ8Tc4eksmMy9-4W_6JBRmwuDYZWN7K9leiWtmyuDBs3BerkcPPIII-F8LNLFgUYx0dwDMoI1ffPBv_ZZsDx7VDma8X6GA5rcmTrM6y03JZiJoi53il1WfuenH5K6op_xPtlr9EY6rF_tgHRMfkiKaRUKaWjT--GZIilput6g-wUhpytLVV3_lm6zMi9Y2_W2oNm7RvZaZ8lruaVVs3CKAbOgLpoVXNmW6InAnfelphlG5G4ykD5HZD6-fRpNWNNGgSkeAvaGW7ApfKFNHGDtJ8_VUeipxAVjhVswuBLhhCKNlaOVtQl3jdJWGMEVNy6qMMekm69yc0Io6Aup74NNCQeeSJ0kFW6QaOtrUFxS1-mR6xY8ua6rZcg2jOxFItQSoZZOLAHqHvFbfOW3OZcgzv8cd9lOhoSfAT0cNTASdFX4qFg47uk_n31GdvGMVU0izkm32JTmAtSOIu1X66pPdoZ3D5PZJ7Wx2Pw |
link.rule.ids | 315,786,790,4521,24144,27957,27958,45620,45714 |
linkProvider | Elsevier |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1LS8NAEF6qHvQiior1uYI3WZvdZJP2KGKp2vaiQm9Lsg9TkbS0CeLF3-5MHoIiHryFhA3Jt8nMNzuz8xFyzoXzQm0DpsGbswAcIIM4pMuMjZxzoTRehHuHR-Nw8BTcTeSkRa6bvTBYVlnb_sqml9a6PtOp0ezMp9POA-9iJZzEJl0c3KhcIWtI51G_4fKDfy20YH_2bplpxgEMR9SbZ6o6r7fMFhAniqrfKcqs_O6gfpjq0v_0t8hmTRzpVfVs26Rlsx2Sj8paSEtr8Ydnil7J0PkC8y-IOZ05qqsGuHSZFlnOGtnbnKbvBt3XPI1fiyUt1cIpVswCX7QzOLMsMBWBS-9TQ1MsyV2kYH52yVP_5vF6wGodBaZ5BOBb7iCokMLYXojNnwLfdKNAxz5EK9xBxBULLxJJT3tGOxdz32rjhBVcc-sjh9kjq9kss_uEAmFIpISgEg4CkXhxIvwwNk4aYC6J77XJRQOemlftMlRTR_aiEGqFUCuvpwDqNpENvurbpCuw53-OO2smQ8HfgCmOChgFZBVeqic8_-Cf9z4l64PH0VANb8f3h2QDr7BSMeKIrOaLwh4DB8mTk_Ib-wTVhtqO |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Machine+learning+based+prediction+of+chronic+shunt-dependent+hydrocephalus+after+spontaneous+subarachnoid+hemorrhage&rft.jtitle=World+neurosurgery&rft.au=Gollwitzer%2C+Maria&rft.au=Steindl%2C+Markus&rft.au=Stroh%2C+Nico&rft.au=Hauser%2C+Anna&rft.date=2024-09-12&rft.issn=1878-8750&rft_id=info:doi/10.1016%2Fj.wneu.2024.09.047&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_wneu_2024_09_047 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1878-8750&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1878-8750&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1878-8750&client=summon |