Prediction-Driven Decision Support for Patients With Mild Stroke: A Model Based on Machine Learning Algorithms

Background and Purpose: Treatment for mild stroke remains an open question. We aim to develop a decision support tool based on machine learning (ML) algorithms, called DAMS (Disability After Mild Stroke), to identify mild stroke patients who would be at high risk of post-stroke disability (PSD) if they only received medical therapy and, more importantly, to aid neurologists in making individual clinical decisions in emergency contexts.Methods: Ischemic stroke patients were prospectively recorded in the National Advanced Stroke Center of Nanjing First Hospital (China) between July 2016 and September 2020. The exclusion criteria were patients who received thrombolytic therapy, age <18 years, lack of 3-month modified Rankin Scale (mRS), disabled before the index stroke, with an admission National Institute of Health stroke scale (NIHSS) > 5. The primary outcome was PSD, corresponding to 3-month mRS ≥ 2. We developed five ML models and assessed the area under curve (AUC) of receiver operating characteristic, calibration curve, and decision curve analysis. The optimal ML model was selected to be DAMS. In addition, SHapley Additive exPlanations (SHAP) approach was introduced to rank the feature importance. Finally, rapid-DAMS (R-DAMS) was constructed for a more urgent situation based on DAMS.Results: A total of 1,905 mild stroke patients were enrolled in this study, and patients with PSD accounted for 23.4% (447). There was no difference in AUCs between the five models (ranged from 0.691 to 0.823). Although there was similar discriminative performance between ML models, the support vector machine model exhibited higher net benefit and better calibration (Brier score, 0.159, calibration slope, 0.935, calibration intercept, 0.035). Therefore, this model was selected for DAMS. In addition, SHAP approach showed that the most crucial feature was NIHSS on admission. Finally, R-DAMS was constructed and there was similar discriminative performance between R-DAMS and DAMS, but the former performed worse on calibration.Conclusions: DAMS and R-DAMS, as prediction-driven decision support tools, were designed to aid clinical decision-making for mild stroke patients in emergency contexts. In addition, even within a narrow range of baseline scores, NIHSS on admission is the strongest feature that contributed to the prediction.

Malnutrition could predict 3-month functional prognosis in mild stroke patients: Findings from a Nationwide Stroke Registry

Background: Previous studies showed the inconsistent effects of malnutrition contributing to poor prognosis of stroke. The aim of this study is to explore the effect of malnutrition on 3-month functional prognosis of stroke patients with different stroke severity assessed by the national institute of health stroke scale (NIHSS). Patients with first-ever stroke were consecutively enrolled in a nationwide, multicenter, and prospective registry from March 2007 to January 2008. Methods: Nutritional status was assessed at admission. Malnutrition was defined by any abnormality of 6 nutrition indicators, including body mass index (BMI), mid-upper arm circumference (MUAC), triceps skinfold thickness (TSF), haemoglobin, albumin, and prealbumin. Stroke patients were classified into mild (NIHSS<8) and severe (NIHSS≥8) groups. Multivariable logistic regression was performed to assess the risk of poor functional prognosis [ modified Rankin Scale (mRS) ≥3] and mortality at 3-month follow-up in mild or severe patients with malnutrition at admission. A total of 755 patients with first-ever stroke were enrolled in the study. Results: Multivariable analysis showed that malnutrition independently contributed to a higher risk of mRS 3-6 at 3-month for mild stroke patients [odds ratio (OR) 1.86, 95% confidence interval (CI) 1.04-3.34], but didn’t for severe stroke patients (OR 0.91, 95% CI 0.53-1.54) after adjusting for confounders including age, NIHSS, and infection et al. After adjusting for the potential confounders, malnutrition was still an independent risk factor for 3-month poor functional prognosis in mild stroke patients. Conclusion: Further investigation may be needed to illustrate the effects of improving nutritional status on stroke patients.

Participants’ Perspective of Engaging in a Gym-Based Health Service Delivered Secondary Stroke Prevention Program after TIA or Mild Stroke

People who have had a transient ischemic attack (TIA) or mild stroke have a high risk of recurrent stroke. Secondary prevention programs providing support for meeting physical activity recommendations may reduce this risk. Most evidence for the feasibility and effectiveness of secondary stroke prevention arises from programs developed and tested in research institute settings with limited evidence for the acceptability of programs in ‘real world’ community settings. This qualitative descriptive study explored perceptions of participation in a secondary stroke prevention program (delivered by a community-based multidisciplinary health service team within a community gym) by adults with TIA or mild stroke. Data gathered via phone-based semi-structured interviews midway through the program, and at the end of the program, were analyzed using constructivist grounded theory methods. A total of 51 interviews from 30 participants produced two concepts. The first concept, “What it offered me”, describes critical elements that shape participants’ experience of the program. The second concept, “What I got out of it” describes perceived benefits of program participation. Participants perceived that experiences with peers in a health professional-led group program, held within a community-based gym, supported their goal of changing behaviour. Including these elements during the development of health service strategies to reduce recurrent stroke risk may strengthen program acceptability and subsequent effectiveness.

Rancang bangun alat monitoring fisioterapi pasien pasca stroke menggunakan sensor MPU6050 melalui media transmisi bluetooth

Stroke Transient Ischemic Attack (TIA ) adalah kondisi stroke ringan. Stroke ringan diakibatkan oleh berkurangnya suplai darah menuju otak. Kelumpuhan disebabkan oleh kurangnya aliran darah ke otak, sehingga sisi tubuh akan mengalami kelumpuhan terutama pada siku dan lutut. Fisioterapi pada pasien pasca stroke ringan dilakukan dengan metode terapi gerakan antara lain seperti penggunaan alat bantu fisioterapi pasien pasca stroke menggunakan sensor MPU6050 melalui media transmisi Bluetooth. Sistem rancangan ini membantu latihan pada pasien berbentuk sensor MPU6050 yang digunakan untuk mendeteksi sudut serta banyaknya gerakan rehabilitasi pasien pasca stroke. Sensor gyroscope menghasilkan deteksi sudut pada siku sebesar 150 dan sudut pada lutut sebesar 135 . Data yang didapatkan diproses oleh arduino nano menggunakan metode I2C. Penelitian ini bertujuan untuk merancang alat bantu fisioterapi pasien pasca stroke berbasis mikrokontroler guna memperoleh hasil yang maksimal karena bentuk alat ini yang portable. Stroke Transient Ischemic Attack (TIA) is a mild stroke condition that can be caused by reduced blood supply to the brain due to blocked blood vessels. Lack of blood flow to the brain can cause paralysis, usually only on one side of the body such as the elbows and knees. Physiotherapy in post-stroke patients is carried out using movement therapy methods, including elbow and knee flexion. Currently, patients come to the therapist or vice versa to perform therapeutic movements, so a telemonitoring system for post-stroke patient physiotherapy tools is made using the MPU6050 sensor which is connected to ESP 232 and the internet network and can be monitored by therapists and clinic admins where the therapy is done. This system helps monitor elbow and knew flexion exercises using the MPU6050 sensor which is used to detect the angle and number of rehabilitation movements of post-stroke patients. . From the results of testing the accuracy of this sensor is able to detect the angle of elbow movement from 00-1500 and knee flexion angle of 00-1350 by 99%.The telemonitoring application for physiotherapy tools for post-stroke patients based on Android is equipped with a therapist menu so that the number of movements and the average angle of movement of the knees and elbows can be monitored by the therapist, this tool can also be used anywhere and anytime due to the portable form of this tool.

Delays in Presentation Time Under the COVID-19 Epidemic in Patients With Transient Ischemic Attack and Mild Stroke: A Retrospective Study of Three Hospitals in a Japanese Prefecture

Background: Coronavirus Disease 2019 (COVID-19) has spread worldwide with collateral damage and therefore might affect the behavior of stroke patients with mild symptoms seeking medical attention.Methods: Patients with ischemic stroke who were admitted to hospitals within 7 days of onset were retrospectively registered. The clinical characteristics, including onset-to-door time (ODT), of patients with a transient ischemic attack (TIA)/mild stroke (National Institutes of Health Stroke Scale [NIHSS] score of ≤ 3 on admission) or moderate/severe stroke were compared between those admitted from April 2019 to March 2020 (pre-COVID-19 period) and from April to September 2020 (COVID-19 period). Multivariable regression analysis was performed to identify factors associated with the ODT.Results: Of 1,100 patients (732 men, median age, 73 years), 754 were admitted during the pre-COVID-19 period, and 346 were admitted during the COVID-19 period. The number and proportion of patients with TIA/minor stroke were 464 (61.5%) in the pre-COVID-19 period and 216 (62.4%) during the COVID-19 period. Among patients with TIA/mild stroke, the ODT was longer in patients admitted during the COVID-19 period compared with that of the pre-COVID-19 period (median 864 min vs. 508 min, p = 0.003). Multivariable analysis revealed the COVID-19 period of admission was associated with longer ODT (standardized partial regression coefficient 0.09, p = 0.003) after adjustment for age, sex, route of arrival, NIHSS score on admission, and the presence of hypertension, diabetes mellitus, and wake-up stroke. No significant change in the ODT was seen in patients with moderate/severe stroke.Conclusions: The COVID-19 epidemic might increase the ODT of patients with TIA/mild stroke.

Thrombolysis in Mild Stroke: A Comparative Analysis of the PRISMS and MaRISS Studies

Background and Purpose: Mild ischemic stroke patients enrolled in randomized controlled trials of thrombolysis may have a different symptom severity distribution than those treated in routine clinical practice. Methods: We compared the distribution of the National Institutes of Health Stroke Scale (NIHSS) scores, neurological symptoms/severity among patients enrolled in the PRISMS (Potential of r-tPA for Ischemic Strokes With Mild Symptoms) randomized controlled trial to those with NIHSS score ≤5 enrolled in the prospective MaRISS (Mild and Rapidly Improving Stroke Study) registry using global P values from χ 2 analyses. Results: Among 1736 participants in MaRISS, 972 (56%) were treated with alteplase and 764 (44%) were not. These participants were compared with 313 patients randomized in PRISMS. The median NIHSS scores were 3 (2–4) in MaRISS alteplase-treated, 1 (1–3) in MaRISS non–alteplase-treated, and 2 (1–3) in PRISMS. The percentage with an NIHSS score of 0 to 2 was 36.3%, 73.3%, and 65.2% in the 3 groups, respectively ( P <0.0001). The proportion of patients with a dominant neurological syndrome (≥1 NIHSS item score of ≥2) was higher in MaRISS alteplase-treated (32%) compared with MaRISS nonalteplase-treated (13.8%) and PRISMS (8.6%; P <0.0001). Conclusions: Patients randomized in PRISMS had comparable deficit and syndromic severity to patients not treated with alteplase in the MaRISS registry and lesser severity than patients treated with alteplase in MaRISS. The PRISMS trial cohort is representative of mild patients who do not receive alteplase in current broad clinical practice.