Deep Learning-Based Automatic Classification of Ischemic Stroke Subtype Using Diffusion-Weighted Images
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Background and Purpose Accurate classification of ischemic stroke subtype is important for effective secondary prevention of stroke. We used diffusion-weighted image (DWI) and atrial fibrillation (AF) data to train a deep learning algorithm to classify stroke subtype. Methods Model development was done in 2,988 patients with ischemic stroke from three centers by using U-net for infarct segmentation and EfficientNetV2 for subtype classification. Experienced neurologists (n=5) determined subtypes for external test datasets, while establishing a consensus for clinical trial datasets. Automatically segmented infarcts were fed into the model (DWI-only algorithm). Subsequently, another model was trained, with AF included as a categorical variable (DWI+AF algorithm). These models were tested: (1) internally against the opinion of the labeling experts, (2) against fresh external DWI data, and (3) against clinical trial dataset. Results In the training-and-validation datasets, the mean (±standard deviation) age was 68.0±12.5 (61.1% male). In internal testing, compared with the experts, the DWI-only and the DWI+AF algorithms respectively achieved moderate (65.3%) and near-strong (79.1%) agreement. In external testing, both algorithms again showed good agreements (59.3%–60.7% and 73.7%–74.0%, respectively). In the clinical trial dataset, compared with the expert consensus, percentage agreements and Cohen’s kappa were respectively 58.1% and 0.34 for the DWI-only vs. 72.9% and 0.57 for the DWI+AF algorithms. The corresponding values between experts were comparable (76.0% and 0.61) to the DWI+AF algorithm. Conclusion Our model trained on a large dataset of DWI (both with or without AF information) was able to classify ischemic stroke subtypes comparable to a consensus of stroke experts.
Abstract
Background and Purpose Accurate classification of ischemic stroke subtype is important for effective secondary prevention of stroke. We used diffusion-weighted image (DWI) and atrial fibrillation (AF) data to train a deep learning algorithm to classify stroke subtype. Methods Model development was done in 2,988 patients with ischemic stroke from three centers by using U-net for infarct segmentation and EfficientNetV2 for subtype classification. Experienced neurologists (n=5) determined subtypes for external test datasets, while establishing a consensus for clinical trial datasets. Automatically segmented infarcts were fed into the model (DWI-only algorithm). Subsequently, another model was trained, with AF included as a categorical variable (DWI+AF algorithm). These models were tested: (1) internally against the opinion of the labeling experts, (2) against fresh external DWI data, and (3) against clinical trial dataset. Results In the training-and-validation datasets, the mean (±standard deviation) age was 68.0±12.5 (61.1% male). In internal testing, compared with the experts, the DWI-only and the DWI+AF algorithms respectively achieved moderate (65.3%) and near-strong (79.1%) agreement. In external testing, both algorithms again showed good agreements (59.3%–60.7% and 73.7%–74.0%, respectively). In the clinical trial dataset, compared with the expert consensus, percentage agreements and Cohen’s kappa were respectively 58.1% and 0.34 for the DWI-only vs. 72.9% and 0.57 for the DWI+AF algorithms. The corresponding values between experts were comparable (76.0% and 0.61) to the DWI+AF algorithm. Conclusion Our model trained on a large dataset of DWI (both with or without AF information) was able to classify ischemic stroke subtypes comparable to a consensus of stroke experts.
참고문헌 (47)
- Albers GW, "Diffusion-weighted MRI for evaluation of acute stroke", Neurology, 51(3 Suppl 3), pp.S47~S49 (1998)
- Kang DW, "Association of ischemic lesion patterns on early diffusion-weighted imaging with TOAST stroke subtypes", Arch Neurol, 60, pp.1730~1734 (2003)
- Chin YY, "Prevalence, risk factors and secondary prevention of stroke recurrence in eight countries from South, East and Southeast Asia : a scoping review", Med J Malaysia, 73, pp.90~99 (2018)
- Flach C, "Risk and secondary prevention of stroke recurrence : a population-base cohort study", Stroke, 51, pp.2435~2444 (2020)
- Ryu WS, "White matter hyperintensity load on stroke recurrence and mortality at 1 year after ischemic stroke", Neurology, 93, pp.e578~e589 (2019)
- Adams HP Jr, "Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment", Stroke, 24, pp.35~41 (1993)
- Ko Y, "MRIbased algorithm for acute ischemic stroke subtype classification", J Stroke, 16, pp.161~172 (2014)
- Mamoli A, "An analysis of the costs of ischemic stroke in an Italian stroke unit", Neurology, 53, pp.112~116 (1999)
- Jung WS, "National trends in medical costs and prognosis of acute ischemic stroke patients in endovascular thrombectomy era : analysis using medical claim data in Korea", Neurointervention, 17, pp.152~160 (2022)
- Fang G, "Automated ischemic stroke subtyping based on machine learning approach", IEEE Access, 8, pp.18426~118432 (2020)
- Zhang S, "Interpretable CNN for ischemic stroke subtype classification with active model adaptation", BMC Med Inform Decis Mak, 22, pp.3~ (2022)
- Goldstein LB, "Improving the reliability of stroke subgroup classification using the Trial of ORG 10172 in Acute Stroke Treatment(TOAST)criteria", Stroke, 32, pp.1091~1098 (2001)
- Buck BH, "Effect of implantable vs prolonged external electrocardiographic monitoring on atrial fibrillation detection in patients with ischemic stroke : the PER DIEM randomized clinical trial", JAMA, 325, pp.2160~2168 (2021)
- Sanna T, "Cryptogenic stroke and underlying atrial fibrillation", N Engl J Med, 370, pp.2478~2486 (2014)
- Kim BJ, "Case characteristics, hyperacute treatment, and outcome information from the clinical research center for stroke-fifth division registry in South Korea", J Stroke, 17, pp.38~53 (2015)
- Ryu WS, "Acute infarct segmentation on diffusion-weighted imaging using deep learning algorithm and RAPID MRI", J Stroke, 25, pp.425~429 (2023)
- Otsu N, "A threshold selection method from gray-level histograms", IEEE Trans Syst Man Cybern, 9, pp.62~66 (1979)
- Tan M, "EfficientNetV2: smaller models and faster training", Proceedings of Machine Learning Research. ICML 2021: Proceedings of the 38th International Conference on Machine Learning, pp.10096~10106 (2021), Journal of Machine Learning Research
- He K, "Deep residual learning for image recognition", Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.770~778 (2016), IEEE
- Tan M, "EfficientNet: rethinking model scaling for convolutional neural networks", Proceedings of Machine Learning Research. ICML 2019: Proceedings of the 36th International Conference on Machine Learning, pp.6105~6114 (2019)
- De Bruijn B, "User centred networked health care (volume 169)", pp.532~536 (2011), IOS Press
- Cuzick J, "A Wilcoxon-type test for trend", Stat Med, 4, pp.87~90 (1985)
- Garg R, "Automating ischemic stroke subtype classification using machine learning and natural language processing", J Stroke Cerebrovasc Dis, 28, pp.2045~2051 (2019)
- Kim BK, "Deep learning for prediction of mechanism in acute ischemic stroke using brain diffusion magnetic resonance image", J Neurocrit Care, 16, pp.85~93 (2023)
- Cheng B, "Stroke subtype classification by geometrical descriptors of lesion shape", PLoS One, 12, pp.e0185063~ (2017)
- Kang DW, "Early recurrent ischemic lesions on diffusion-weighted imaging in symptomatic intracranial atherosclerosis", Arch Neurol, 64, pp.50~54 (2007)
- Kleindorfer DO, "2021 guideline for the prevention of stroke in patients with stroke and transient ischemic attack : a guideline from the American Heart Association", American Stroke Association, 2021, pp.52, e364~e467
- Dawson J, "European Stroke Organisation(ESO)guideline on pharmacological interventions for long-term secondary prevention after ischaemic stroke or transient ischaemic attack", Eur Stroke J, 7, pp.I~II (2022)
- Benavente OR, "Blood-pressure targets in patients with recent lacunar stroke : the SPS3 randomised trial", Lancet, 382, pp.507~515 (2013)
- Jing J, "Dual antiplatelet therapy in transient ischemic attack and minor stroke with different infarction patterns : subgroup analysis of the CHANCE randomized clinical trial", JAMA Neurol, 75, pp.711~719 (2018)
- Gao Y, "Dual antiplatelet treatment up to 72 hours after ischemic stroke", N Engl J Med, 389, pp.2413~2424 (2023)
- Lee M, "Intensive medical management to prevent large and small artery atherothrombotic stroke : time to expand the horizon", JAMA, 326, pp.217~218 (2021)
- Kim JS, "Role of blood lipid levels and lipid-lowering therapy in stroke patients with different levels of cerebral artery diseases : reconsidering recent stroke guidelines", J Stroke, 23, pp.149~161 (2021)
- Sharma M, "Efficacy and harms of direct oral anticoagulants in the elderly for stroke prevention in atrial fibrillation and secondary prevention of venous thromboembolism : systematic review and meta-analysis", Circulation, 132, pp.194~204 (2015)
- Hart RG, "Embolic stroke of undetermined source : a systematic review and clinical update", Stroke, 48, pp.867~872 (2017)
- Poli S, "The ATTICUS randomized controlled trial-subgroup analyses", Stroke, 54(Suppl 1), pp.A31~ (2023)
- Hart RG, "Rivaroxaban for stroke prevention after embolic stroke of undetermined source", N Engl J Med, 378, pp.2191~2201 (2018)
- Diener HC, "Dabigatran for prevention of stroke after embolic stroke of undetermined source", N Engl J Med, 380, pp.1906~1917 (2019)
- Kamel H, "Machine learning prediction of stroke mechanism in embolic strokes of undetermined source", Stroke, 51, pp.e203~e210 (2020)
- Veltkamp R, "Characteristics of recurrent ischemic stroke after embolic stroke of undetermined source : secondary analysis of a randomized clinical trial", JAMA Neurol, 77, pp.1233~1240 (2020)
- Kim AS, "Controversies in stroke : should patients with embolic stroke of undetermined source undergo intensive heart rhythm monitoring with an implantable loop recorder?", Stroke, 53, pp.3243~3247 (2022)
- Hong KS, "Non-vitamin K antagonist oral anticoagulants in medical conditions at high risk of thromboembolism beyond atrial fibrillation", J Stroke, 21, pp.259~275 (2019)
- Liu Z, "Comparison between healthcare quality in primary stroke centers and comprehensive stroke centers for acute stroke patients : evidence from the Chinese Stroke Center Alliance", Lancet Reg Health West Pac, 38, pp.100863~ (2023)
- Kim J, "Nationwide acute stroke care quality and disparity in Korea: focusing on the type of healthcare facilities and the socioeconomic status of patients", Public Health Aff, 5, pp.e6~ (2021)
- Lee LJ, "Impact on stroke subtype diagnosis of early diffusion-weighted magnetic resonance imaging and magnetic resonance angiography", Stroke, 31, pp.1081~1089 (2000)