Deep Learning Approach Using Diffusion-Weighted Imaging to Estimate the Severity of Aphasia in Stroke Patients
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Background and Purpose This study aimed to investigate the applicability of deep learning (DL) model using diffusion-weighted imaging (DWI) data to predict the severity of aphasia at an early stage in acute stroke patients. Methods We retrospectively analyzed consecutive patients with aphasia caused by acute ischemic stroke in the left middle cerebral artery territory, who visited Asan Medical Center between 2011 and 2013. To implement the DL model to predict the severity of post-stroke aphasia, we designed a deep feed-forward network and utilized the lesion occupying ratio from DWI data and established clinical variables to estimate the aphasia quotient (AQ) score (range, 0 to 100) of the Korean version of the Western Aphasia Battery. To evaluate the performance of the DL model, we analyzed Cohen’s weighted kappa with linear weights for the categorized AQ score (0–25, very severe; 26–50, severe; 51–75, moderate; ≥76, mild) and Pearson’s correlation coefficient for continuous values. Results We identified 225 post-stroke aphasia patients, of whom 176 were included and analyzed. For the categorized AQ score, Cohen’s weighted kappa coefficient was 0.59 (95% confidence interval [CI], 0.42 to 0.76; P<0.001). For continuous AQ score, the correlation coefficient between true AQ scores and model-estimated values was 0.72 (95% CI, 0.55 to 0.83; P<0.001). Conclusions DL approaches using DWI data may be feasible and useful for estimating the severity of aphasia in the early stage of stroke.
Abstract
Background and Purpose This study aimed to investigate the applicability of deep learning (DL) model using diffusion-weighted imaging (DWI) data to predict the severity of aphasia at an early stage in acute stroke patients. Methods We retrospectively analyzed consecutive patients with aphasia caused by acute ischemic stroke in the left middle cerebral artery territory, who visited Asan Medical Center between 2011 and 2013. To implement the DL model to predict the severity of post-stroke aphasia, we designed a deep feed-forward network and utilized the lesion occupying ratio from DWI data and established clinical variables to estimate the aphasia quotient (AQ) score (range, 0 to 100) of the Korean version of the Western Aphasia Battery. To evaluate the performance of the DL model, we analyzed Cohen’s weighted kappa with linear weights for the categorized AQ score (0–25, very severe; 26–50, severe; 51–75, moderate; ≥76, mild) and Pearson’s correlation coefficient for continuous values. Results We identified 225 post-stroke aphasia patients, of whom 176 were included and analyzed. For the categorized AQ score, Cohen’s weighted kappa coefficient was 0.59 (95% confidence interval [CI], 0.42 to 0.76; P<0.001). For continuous AQ score, the correlation coefficient between true AQ scores and model-estimated values was 0.72 (95% CI, 0.55 to 0.83; P<0.001). Conclusions DL approaches using DWI data may be feasible and useful for estimating the severity of aphasia in the early stage of stroke.
참고문헌 (36)
- Benjamin EJ, "Heart disease and stroke statistics-2019update : a report from the American Heart Association", Circulation, 139, pp.e56~e528 (2019)
- Ellis C, "Longitudinal trends in aphasia in the United States", NeuroRehabilitation, 27, pp.327~333 (2010)
- Ellis C, "The one-year attributable cost of poststroke aphasia", Stroke, 43, pp.1429~1431 (2012)
- Andreas Charidimou, "Why Is It Difficult to Predict Language Impairment and Outcome in Patients with Aphasia after Stroke?", Journal of Clinical Neurology, 10(2), pp.75~83 (2014), 대한신경과학회[DOI]
- Pedersen PM, "Aphasia in acute stroke: incidence, determinants, and recovery", Ann Neurol, 38, pp.659~666 (1995)
- Nouwens F, "Severity of aphasia and recovery after treatment in patients with stroke", Aphasiology, 28, pp.1168~1177 (2014)
- El Hachioui H, "Long-term prognosis of aphasia after stroke", J Neurol Neurosurg Psychiatry, 84, pp.310~315 (2013)
- Nouwens F, "Validation of a prediction model for long-term outcome of aphasia after stroke", BMC Neurol, 18, pp.170~ (2018)
- Payabvash S, "Predicting language improvement in acute stroke patients presenting with aphasia : a multivariate logistic model using location-weighted atlas-based analysis of admission CT perfusion scans", AJNR Am J Neuroradiol, 31, pp.1661~1668 (2010)
- 이은재, "Deep into the Brain: Artificial Intelligence in Stroke Imaging", 대한뇌졸중영문학회지, 19(3), pp.277~285 (2017), 대한뇌졸중학회
- Lee H, "Machine learning approach to identify stroke within 4.5 hours", Stroke, 51, pp.860~866 (2020)
- Kassner A, "Texture analysis : a review of neurologic MR imaging applications", AJNR Am J Neuroradiol, 31, pp.809~816 (2010)
- Wernick MN, "Machine learning in medical imaging", IEEE Signal Process Mag, 27, pp.25~38 (2010)
- Kim H, "Normative data on the Korean version of the Western Aphasia Battery", J Clin Exp Neuropsychol, 26, pp.1011~1020 (2004)
- Brott T, "Measurements of acute cerebral infarction : a clinical examination scale", Stroke, 20, pp.864~870 (1989)
- Forsting M, "MR Neuroimaging : Brain, Spine, Peripheral Nerves" (2017), Thieme
- Kim BJ, "Lesion location-based prediction of visual field improvement after cerebral infarction", PLoS One, 10, pp.e0143882~ (2015)
- Plowman E, "Post-stroke aphasia prognosis : a review of patient-related and stroke-related factors", J Eval Clin Pract, 18, pp.689~694 (2012)
- Srivastava N, "Dropout : a simple way to prevent neural networks from overfitting", J Mach Learn Res, 15, pp.1929~1958 (2014)
- Raita Y, "Emergency department triage prediction of clinical outcomes using machine learning models", Crit Care, 23, pp.64~ (2019)
- Chung MK, "Statistical and Computational Methods in Brain Image Analysis" (2013), CRC Press
- Martinez-Martin P, "Extensive validation study of the Parkinson’s disease composite scale", Eur J Neurol, 26, pp.1281~1288 (2019)
- Kertesz A, "Western Aphasia Battery-Revised (WAB-R): Examiner’s Manual" (2006), PsychCorp
- Andersen R, "Modern Methods for Robust Regression" (2007), Sage Publications
- Watila MM, "Factors predicting post-stroke aphasia recovery", J Neurol Sci, 352, pp.12~18 (2015)
- Laska AC, "Aphasia in acute stroke and relation to outcome", J Intern Med, 249, pp.413~422 (2001)
- Lazar RM, "Improvement in aphasia scores after stroke is well predicted by initial severity", Stroke, 41, pp.1485~1488 (2010)
- Glize B, "Language features in the acute phase of poststroke severe aphasia could predict the outcome", Eur J Phys Rehabil Med, 53, pp.249~255 (2017)
- Blom-Smink MR, "Predicting everyday verbal communicative ability after inpatient stroke rehabilitation in patients with moderate and severe aphasia at admission : validation of a prognostic model", Eur J Phys Rehabil Med, 55, pp.532~534 (2019)
- Godecke E, "Amount of therapy matters in very early aphasia rehabilitation after stroke : a clinical prognostic model", Semin Speech Lang, 34, pp.129~141 (2013)
- Liu X, "Advances in deep learning-based medical image analysis", Health Data Sci, 2021, pp.8786793~ (2021)
- "Abulia" (2019)[URL]
- Siegel JS, "The circuitry of abulia : insights from functional connectivity MRI", Neuroimage Clin, 6, pp.320~326 (2014)
- Labeyrie MA, "Diffusion lesion reversal after thrombolysis : a MR correlate of early neurological improvement", Stroke, 43, pp.2986~2991 (2012)
- 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)
- "Highway networks" (2015)[URL]