1
|
Li Y, Liu Y, Hong Z, Wang Y, Lu X. Combining machine learning with radiomics features in predicting outcomes after mechanical thrombectomy in patients with acute ischemic stroke. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2022; 225:107093. [PMID: 36055039 DOI: 10.1016/j.cmpb.2022.107093] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 05/28/2022] [Revised: 08/25/2022] [Accepted: 08/25/2022] [Indexed: 06/15/2023]
Abstract
BACKGROUND AND OBJECTIVE Some patients with mechanical thrombectomy will have a poor prognosis. This study establishes a model for predicting the prognosis after mechanical thrombectomy in acute stroke based on diffusion-weighted imaging (DWI) omics characteristics. METHODS A total of 260 stroke patients receiving mechanical thrombectomy in our hospital were randomly divided into a training set (n = 182) and a test set (n = 78) in a 7:3 ratio. The regions of interest (ROI) of the imaging features of the DWI infarct area were extracted, and the minimum absolute contraction and selection operator regression model were used to screen the best radiomics features. A support vector machine classifier established the prediction model of the prognosis after mechanical thrombectomy of acute stroke based on the selected features. The prediction efficiency of the model was evaluated by the receiver operating characteristic (ROC) curve. RESULTS A total of 1936 radiomic features were extracted, and six features highly correlated with prognosis were screened after dimensionality reduction. Based on the DWI model, the ROC analysis showed that the area under the curve (AUC) for correct prediction in the training and test sets was 0.945 and 0.920, respectively. CONCLUSION The model based on the characteristics of radiomics and machine learning has high predictive efficiency for the prognosis of acute stroke after mechanical thrombectomy, which can be used to guide personalized clinical treatment.
Collapse
Affiliation(s)
- Yan Li
- Department of Neurosurgery, Cangzhou Central Hospital, Cangzhou Clinical Medical College of Hebei Medical University, Canzhou 061011, China.
| | - Yongchang Liu
- Department of Neurosurgery, Cangzhou Central Hospital, Cangzhou Clinical Medical College of Hebei Medical University, Canzhou 061011, China
| | - Zhen Hong
- Department of Neurosurgery, Cangzhou Central Hospital, Cangzhou Clinical Medical College of Hebei Medical University, Canzhou 061011, China
| | - Ying Wang
- Department of Neurosurgery, Cangzhou Central Hospital, Cangzhou Clinical Medical College of Hebei Medical University, Canzhou 061011, China
| | - Xiuling Lu
- Cangzhou Infectious Disease Hospital, Canzhou 061011, China
| |
Collapse
|
2
|
Trandafir C, Sandiramourty S, Laurent-Chabalier S, Ter Schiphorst A, Nguyen H, Wacongne A, Ricci JE, Pereira F, Thouvenot E, Renard D. Brain Infarction MRI Pattern in Stroke Patients with Intracardiac Thrombus. Cerebrovasc Dis 2021; 50:581-587. [PMID: 34139688 DOI: 10.1159/000515707] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/14/2020] [Accepted: 02/25/2021] [Indexed: 11/19/2022] Open
Abstract
BACKGROUND Acute infarction patterns have been described in cardioembolic stroke, mainly with atrial fibrillation (AF) or patent foramen ovale. We aimed to analyse acute infarction magnetic resonance imaging (MRI) characteristics in stroke patients with intracardiac thrombus (ICT) compared with stroke patients with AF. METHODS We performed a retrospective study analysing brain MRI scans of consecutive acute symptomatic cardioembolic infarction patients associated with ICT or AF who were recruited and registered in the stroke database between June 2018 and November 2019. Diffusion-weighted imaging performed within 1 week after symptom onset, intra-/extracranial vessel imaging, echocardiography, and ≥24-h ECG monitoring were required for inclusion. Baseline, biological, and echocardiography characteristics were assessed. Analysed MRI characteristics were infarction location (anterior/middle/posterior cerebral artery territory; anterior/posterior/mixed anterior-posterior circulation; multiterritorial infarction; brainstem; cerebellum; small cortical cerebellar infarctions [SCCIs] or non-SCCI; cortical/subcortical/cortico-subcortical), lesion number, subcortical lesion size (> or <15 mm), and total infarction volume. RESULTS We included 28 ICT and 94 AF patients presenting with acute stroke. ICT patients were younger (median age 66 vs. 81 years, p < 0.001), more frequently male (79 vs. 47%, p = 0.003), and smokers (39 vs. 17%, p = 0.013), had more frequent history of diabetes (36 vs. 18%, p = 0.049) and ischaemic heart disease (57 vs. 21%, p < 0.001), and had lower HDL cholesterol levels (0.39 vs. 0.53 g/L, p < 0.001). On MRI, SCCI was more frequent in the ICT group (25 vs. 5%, p = 0.006) in the absence of other differences in infarction localisation, number, size, or volume on MRI. On multivariate analysis, younger age (p < 0.001), history of ischaemic heart disease (p < 0.001), and low HDL cholesterol levels (p = 0.01) were significantly associated with ICT. Results approaching statistical significance were observed for SCCI (more frequent in the ICT group, p = 0.053) and non-SCCI (more frequent in the AF group, p = 0.053) on MRI. CONCLUSIONS ICT-related stroke is associated with acute SCCI presence on MRI. Clinical Trial Registration-URL: http://www.clinicaltrials.gov. Unique identifier: NCT04456309.
Collapse
Affiliation(s)
- Cassiana Trandafir
- Department of Neurology, CHU Nîmes, University of Montpellier, Nîmes, France
| | - Shridevi Sandiramourty
- Department of Radiology, Research Team EA 2992, CHU Nîmes, University of Montpellier, Nîmes, France
| | - Sabine Laurent-Chabalier
- Department of Biostatistics, Clinical Epidemiology, Public Health, and Innovation in Methodology, CHU Nîmes, University of Montpellier, Nîmes, France
| | | | - Hai Nguyen
- Department of Radiology, Research Team EA 2992, CHU Nîmes, University of Montpellier, Nîmes, France
| | - Anne Wacongne
- Department of Neurology, CHU Nîmes, University of Montpellier, Nîmes, France
| | - Jean-Etienne Ricci
- Department of Cardiology, CHU Nîmes, University of Montpellier, Nîmes, France
| | - Fabricio Pereira
- Department of Radiology, Research Team EA 2992, CHU Nîmes, University of Montpellier, Nîmes, France
| | - Eric Thouvenot
- Department of Neurology, CHU Nîmes, University of Montpellier, Nîmes, France.,Institut de Génomique Fonctionnelle, CNRS UMR5203, INSERM 1191, University of Montpellier, Montpellier, France
| | - Dimitri Renard
- Department of Neurology, CHU Nîmes, University of Montpellier, Nîmes, France
| |
Collapse
|