1
|
Martín Vicario C, Rodríguez Salas D, Maier A, Hock S, Kuramatsu J, Kallmuenzer B, Thamm F, Taubmann O, Ditt H, Schwab S, Dörfler A, Muehlen I. Uncertainty-aware deep learning for trustworthy prediction of long-term outcome after endovascular thrombectomy. Sci Rep 2024; 14:5544. [PMID: 38448445 PMCID: PMC10917742 DOI: 10.1038/s41598-024-55761-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/16/2023] [Accepted: 02/27/2024] [Indexed: 03/08/2024] Open
Abstract
Acute ischemic stroke (AIS) is a leading global cause of mortality and morbidity. Improving long-term outcome predictions after thrombectomy can enhance treatment quality by supporting clinical decision-making. With the advent of interpretable deep learning methods in recent years, it is now possible to develop trustworthy, high-performing prediction models. This study introduces an uncertainty-aware, graph deep learning model that predicts endovascular thrombectomy outcomes using clinical features and imaging biomarkers. The model targets long-term functional outcomes, defined by the three-month modified Rankin Score (mRS), and mortality rates. A sample of 220 AIS patients in the anterior circulation who underwent endovascular thrombectomy (EVT) was included, with 81 (37%) demonstrating good outcomes (mRS ≤ 2). The performance of the different algorithms evaluated was comparable, with the maximum validation under the curve (AUC) reaching 0.87 using graph convolutional networks (GCN) for mRS prediction and 0.86 using fully connected networks (FCN) for mortality prediction. Moderate performance was obtained at admission (AUC of 0.76 using GCN), which improved to 0.84 post-thrombectomy and to 0.89 a day after stroke. Reliable uncertainty prediction of the model could be demonstrated.
Collapse
Affiliation(s)
- Celia Martín Vicario
- Department of Neuroradiology, Friedrich-Alexander University of Erlangen-Nuremberg, University Hospital Erlangen, Erlangen, Germany.
- Pattern Recognition Lab, Friedrich Alexander University, Erlangen, Germany.
| | - Dalia Rodríguez Salas
- Department of Neuroradiology, Friedrich-Alexander University of Erlangen-Nuremberg, University Hospital Erlangen, Erlangen, Germany
- Pattern Recognition Lab, Friedrich Alexander University, Erlangen, Germany
| | - Andreas Maier
- Pattern Recognition Lab, Friedrich Alexander University, Erlangen, Germany
| | - Stefan Hock
- Department of Neuroradiology, Friedrich-Alexander University of Erlangen-Nuremberg, University Hospital Erlangen, Erlangen, Germany
| | - Joji Kuramatsu
- Department of Neurology, Friedrich-Alexander University of Erlangen-Nuremberg, University Hospital Erlangen, Erlangen, Germany
| | - Bernd Kallmuenzer
- Department of Neurology, Friedrich-Alexander University of Erlangen-Nuremberg, University Hospital Erlangen, Erlangen, Germany
| | | | | | | | - Stefan Schwab
- Department of Neurology, Friedrich-Alexander University of Erlangen-Nuremberg, University Hospital Erlangen, Erlangen, Germany
| | - Arnd Dörfler
- Department of Neuroradiology, Friedrich-Alexander University of Erlangen-Nuremberg, University Hospital Erlangen, Erlangen, Germany
| | - Iris Muehlen
- Department of Neuroradiology, Friedrich-Alexander University of Erlangen-Nuremberg, University Hospital Erlangen, Erlangen, Germany
| |
Collapse
|
2
|
Shen GC, Hang Y, Ma G, Lu SS, Wang C, Shi HB, Wu FY, Xu XQ, Liu S. Prognostic value of multiphase CT angiography: estimated infarct core volume in the patients with acute ischaemic stroke after mechanical thrombectomy. Clin Radiol 2023; 78:e815-e822. [PMID: 37607843 DOI: 10.1016/j.crad.2023.07.015] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/02/2022] [Revised: 07/15/2023] [Accepted: 07/26/2023] [Indexed: 08/24/2023]
Abstract
BACKGROUND AND PURPOSE Recent studies reported the feasibility of quantifying a reliable infarct core (IC) volume using multiphase computed tomography (mCTA) based on deep learning, however its prognostic value was not fully clarified. Therefore, we aimed to evaluate the prognostic value of mCTA-estimated IC volume in patients with acute ischemic stroke (AIS) after mechanical thrombectomy (MT). MATERIALS AND METHODS We retrospectively reviewed patients who underwent mCTA and MT for large vessel occlusion in middle cerebral artery and (or) internal carotid artery within 6 hours after symptom onset between January 2018 and November 2019. Patients were dichotomized into good (modified Rankin Scale [mRS] score, 0-2) and poor (mRS, 3-6) outcome groups. mCTA-estimated IC volume were generated based on a multi-scale three-dimensional convolutional neural network. Univariate, multivariate logistic regression and receiver operating characteristic (ROC) curve analyses were used to identify the independent variables, and evaluate their performances in predicting the clinical outcome. RESULTS Of 44 included patients, 27 (61.4%) patients achieved good outcome. National Institutes of Health Stroke Scale scores at admission [NIHSSpre] (odds ratio [OR], 1.191; 95%confidence interval [CI], 1.028-1.379; P=0.020) and mCTA-estimated IC volume (OR, 1.076; 95%CI, 1.016-1.140; P=0.013) were found to be independently associated with functional outcome in patients with AIS after MT. After integrating NIHSSpre and mCTA-estimated IC volume, optimal performance (area under the ROC curve, 0.874; 95%CI, 0.739-0.954) could be obtained in predicting the clinical outcome. CONCLUSIONS mCTA-estimated IC volume might be promising for predicting the prognosis, and assisting in making individualized treatment decision in patients with AIS.
Collapse
Affiliation(s)
- G-C Shen
- Department of Radiology, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China
| | - Y Hang
- Department of Interventional Radiology, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China
| | - G Ma
- Department of Radiology, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China
| | - S-S Lu
- Department of Radiology, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China
| | - C Wang
- Human Phenome Institute, Fudan University, Shanghai, China
| | - H-B Shi
- Department of Interventional Radiology, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China
| | - F-Y Wu
- Department of Radiology, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China
| | - X-Q Xu
- Department of Radiology, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China.
| | - S Liu
- Department of Interventional Radiology, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China.
| |
Collapse
|
3
|
Shen X, Liao J, Jiang Y, Xu Y, Liu M, Zhang X, Dong N, Yu L, Chen Q, Fang Q. Elevated NT-proBNP levels are associated with CTP ischemic volume and 90-day functional outcomes in acute ischemic stroke: a retrospective cohort study. BMC Cardiovasc Disord 2022; 22:431. [PMID: 36180827 PMCID: PMC9524121 DOI: 10.1186/s12872-022-02861-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/02/2022] [Accepted: 09/14/2022] [Indexed: 11/21/2022] Open
Abstract
Objective To investigate the impact of N-terminal pro-B-type natriuretic peptide (NT-proBNP) on CTP infarct core volume and poor 90-day functional outcomes in acute ischemic stroke (AIS). Methods A total of 403 hospitalized patients with AIS in the Stroke Center of the First Hospital Affiliated to Soochow University were enrolled from March 2018 to January 2021. The association between NT-proBNP and clinical outcomes in acute ischemic patients was assessed by logistic regression and adjusted for confounding factors. Also, subgroup analyses were conducted based on treatment decisions. Results NT-proBNP was positively correlated with CTP ischemic volume (p < 0.001), infarct core volume (p < 0.001), and ischemic penumbra volume (p < 0.001). Univariate analysis showed that the influence of NT-proBNP and functional outcomes were statistically significant in model 1 (p = 0.002). This phenomenon was persistent after adjusted for age, sex, and body mass index in model 2 (p = 0.011), adjusted for SBP, current smoking, family history of stroke, hypertension, and diabetes mellitus in model 3 (p < 0.001), and adjusted for TnI, D-dimer, PLT, Cr, TC, TG, HDL-C, treatment decisions, and NIHSS score in model 4 (p = 0.027). A high NT-proBNP was associated with a high 90-days mRS score among the total population, IV rt-PA, and standardized treatment groups, but not in IV rt-PA + EVT, EVT, and EVT/IV rt-PA + EVT groups. Conclusion Elevated NT-proBNP levels reveal large CTP infarct core volume and poor 90-day functional outcome in AIS. NT-pro BNP is an independent risk factor for functional outcomes.
Collapse
Affiliation(s)
- Xiaozhu Shen
- Department of Neurology, The First Affiliated Hospital of Soochow University, Suzhou, 215000, China.,Department of Geriatrics, Lianyungang Second People's Hospital, Lianyungang, China
| | - Juan Liao
- Department of Neurology, The First Affiliated Hospital of Soochow University, Suzhou, 215000, China
| | - Yi Jiang
- Department of Geriatrics, Lianyungang Second People's Hospital, Lianyungang, China
| | - Yiwen Xu
- Department of Geriatrics, Lianyungang Second People's Hospital, Lianyungang, China
| | - Mengqian Liu
- Department of Geriatrics, Lianyungang Second People's Hospital, Lianyungang, China
| | - Xianxian Zhang
- Department of Neurology, The First Affiliated Hospital of Soochow University, Suzhou, 215000, China. .,Department of Neurology, Yancheng Third People's Hospital, Yancheng, China.
| | - Nan Dong
- Department of Neurology, The First Affiliated Hospital of Soochow University, Suzhou, 215000, China.,Department of Neurology, Suzhou Industrial Park Xinghai Hospital, Suzhou, China
| | - Liqiang Yu
- Department of Neurology, The First Affiliated Hospital of Soochow University, Suzhou, 215000, China
| | - Qingmei Chen
- Department of Neurology, The First Affiliated Hospital of Soochow University, Suzhou, 215000, China
| | - Qi Fang
- Department of Neurology, The First Affiliated Hospital of Soochow University, Suzhou, 215000, China.
| |
Collapse
|