Using Machine Learning to develop a short-form measure assessing 5 functions in patients with stroke

Being insulted by parents is the most severe early adverse experience of anxiety in adulthood

INTRODUCTION

Which adverse childhood experiences are causally associated with anxiety in adulthood is unclear. Sensitive Period Model states that early adverse experiences can influence anxiety levels in adulthood. However, it remains unknown which adverse experience is the most important early adverse factor for anxiety in adulthood, and whether early alleviation or aggravation of this adverse experience predicts a decrease or increase in anxiety levels in adulthood.

METHODS

A national cross-sectional survey was conducted from 20 June 2023 to 31 August 2023, with a total of 30,054 adults aged 18 years or older recruited on questionnaires that completed the Adverse Childhood Experiences Scale and Generalised Anxiety Disorder Scale. We used network analysis and computer simulation techniques to simulate aggravating and simulate alleviating interventions to observe changes in anxiety levels to identify early adverse experiences causally associated with anxiety levels in adulthood.

RESULTS

The results of both the network analysis and the nodeIdentifyR algorithm (NIRA) indicated that in the Adverse Childhood Experiences-Anxiety Network (ACE-GAD Network), being cursed and insulted (CUR) by parents in childhood was the most severe early adverse experience contributing to anxiety in adulthood. In particular, simulated aggravation interventions targeting CUR in young and middle-aged adults resulted in significantly higher levels of anxiety in adulthood.

CONCLUSIONS

This suggests that preventing verbal curses or insults in childhood can be an early preventive measure for future anxiety interventions. This study has important theoretical and practical implications, providing new insights for early prevention of anxiety and family education.

Test-Retest Reliability and Responsiveness of the Machine Learning-Based Short-Form of the Berg Balance Scale in Persons With Stroke

OBJECTIVE

To examine the test-retest reliability, responsiveness, and clinical utility of the machine learning-based short form of the Berg Balance Scale (BBS-ML) in persons with stroke.

DESIGN

Repeated-measures design.

SETTING

A department of rehabilitation in a medical center.

PARTICIPANTS

This study recruited 2 groups: 50 persons who were more than 6 months post-stroke to examine the test-retest reliability, and 52 persons who were within 3 months post-stroke to examine the responsiveness. Test-retest reliability was investigated by administering assessments twice at a 2-week interval. Responsiveness was investigated by gathering data at admission and discharge from the hospital.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURE

BBS-ML.

RESULTS

The BBS-ML exhibited excellent test-retest reliability (intraclass correlation coefficient=0.99), acceptable minimal random measurement error (minimal detectable change %=13.6%), and good responsiveness (Kazis' effect size and standardized response mean values≥1.34). On average, the participants completed the BBS-ML in around 6 minutes per administration.

CONCLUSIONS

Our findings indicate that the BBS-ML appears an efficient measure with excellent test-retest reliability and responsiveness. Moreover, the BBS-ML may be used as a substitute for the original BBS to monitor the progress of balance function in persons with stroke.

Developing an Accumulative Assessment System of Upper Extremity Motor Function in Patients With Stroke Using Deep Learning

OBJECTIVE

The Fugl-Meyer assessment for upper extremity (FMA-UE) is a measure for assessing upper extremity motor function in patients with stroke. However, the considerable administration time of the assessment decreases its feasibility. This study aimed to develop an accumulative assessment system of upper extremity motor function (AAS-UE) based on the FMA-UE to improve administrative efficiency while retaining sufficient psychometric properties.

METHODS

The study used secondary data from 3 previous studies having FMA-UE datasets, including 2 follow-up studies for subacute stroke individuals and 1 test-retest study for individuals with chronic stroke. The AAS-UE adopted deep learning algorithms to use patients' prior information (ie, the FMA-UE scores in previous assessments, time interval of adjacent assessments, and chronicity of stroke) to select a short and personalized item set for the following assessment items and reproduce their FMA-UE scores.

RESULTS

Our data included a total of 682 patients after stroke. The AAS-UE administered 10 different items for each patient. The AAS-UE demonstrated good concurrent validity (r = 0.97-0.99 with the FMA-UE), high test-retest reliability (intra-class correlation coefficient = 0.96), low random measurement error (percentage of minimal detectable change = 15.6%), good group-level responsiveness (standardized response mean = 0.65-1.07), and good individual-level responsiveness (30.5%-53.2% of patients showed significant improvement). These psychometric properties were comparable to those of the FMA-UE.

CONCLUSION

The AAS-UE uses an innovative assessment method, which makes good use of patients' prior information to achieve administrative efficiency with good psychometric properties.

IMPACT

This study demonstrates a new assessment method to improve administrative efficiency while retaining psychometric properties, especially individual-level responsiveness and random measurement error, by making good use of patients' basic information and medical records.

Developing a machine learning-based short form of the positive and negative syndrome scale

BACKGROUND AND HYPOTHESIS

The Positive and Negative Syndrome Scale (PANSS) consists of 30 items and takes up to 50 minutes to administer and score. Therefore, this study aimed to develop and validate a machine learning-based short form of the PANSS (PANSS-MLSF) that reproduces the PANSS scores. Moreover, the PANSS-MLSF estimated the removed-item scores.

STUDY DESIGN

The PANSS-MLSF was developed using an artificial neural network, and the removed-item scores were estimated using the eXtreme Gradient Boosting classifier algorithm. The reliability of the PANSS-MLSF was examined using Cronbach's alpha. The concurrent validity was examined by the association (Pearson's r) between the PANSS-MLSF and the PANSS. The convergent validity was examined by the association (Pearson's r) between the PANSS-MLSF and the Clinical Global Impression-Severity, Mini-Mental State Examination, and Lawton Instrumental Activities of Daily Living Scale. The agreement of the estimated removed-item scores with their original scores was examined using Cohen's kappa.

STUDY RESULTS

Our analysis included data from 573 patients with moderate severity. The two versions of the PANSS-MLSF comprised 15 items and 9 items were proposed. The PANSS-MLSF scores were similar to the PANSS scores (mean squared error=2.6-24.4 points). The reliability, concurrent validity, and convergent validity of the PANSS-MLSF were good. Moderate to good agreement between the estimated removed-item scores and the original item scores was found in 60% of the removed items.

CONCLUSION

The PANSS-MLSF offers a viable way to reduce PANSS administration time, maintain score comparability, uphold reliability and validity, and even estimate scores for the removed items.

Relevance and use of health policy, health systems and health services research for strengthening rehabilitation in real-life settings: methodological considerations

Research on health policy, systems, and services (HPSSR) has seen significant growth in recent decades and received increasing attention in the field of rehabilitation. This growth is driven by the imperative to effectively address real-life challenges in complex healthcare settings. A recent resolution on 'Strengthening rehabilitation in health systems' adopted by the World Health Assembly emphasizes the need to support societal health goals related to rehabilitation, particularly to promote high-quality rehabilitation research, including HPSSR. This conceptual paper, discussed with the participants in the 5th Cochrane Rehabilitation Methodological Meeting held in Milan on September 2023, outlines study designs at diverse levels at which HPSSR studies can be conducted: the macro, meso, and micro levels. It categorizes research questions into four types: those framed from the perspective of policies, healthcare delivery organizations or systems, defined patient or provider populations, and important data sources or research methods. Illustrative examples of appropriate methodologies are provided for each type of research question, demonstrating the potential of HPSSR in shaping policies, improving healthcare delivery, and addressing patient and provider perspectives. The paper concludes by discussing the applicability, usefulness, and implementation of HPSSR findings, and the importance of knowledge translation strategies, drawing insights from implementation science. The goal is to facilitate the integration of research findings into everyday clinical practice to bridge the gap between research and practice in rehabilitation.

Developing a Short-Form Buss-Warren Aggression Questionnaire Based on Machine Learning

For adolescents, high levels of aggression are often associated with suicide, physical injury, worsened academic performance, and crime. Therefore, there is a need for the early identification of and intervention for highly aggressive adolescents. The Buss-Warren Aggression Questionnaire (BWAQ) is one of the most widely used offensive measurement tools. It consists of 34 items, and the longer the scale, the more likely participants are to make an insufficient effort response (IER), which reduces the credibility of the results and increases the cost of implementation. This study aimed to develop a shorter BWAQ using machine learning (ML) techniques to reduce the frequency of IER and simultaneously decrease implementation costs. First, an initial version of the short-form questionnaire was created using stepwise regression and an ANOVA F-test. Then, a machine learning algorithm was used to create the optimal short-form questionnaire (BWAQ-ML). Finally, the reliability and validity of the optimal short-form questionnaire were tested using independent samples. The BWAQ-ML contains only four items, thirty items less than the BWAQ, and its AUC, accuracy, recall, precision, and F1 score are 0.85, 0.85, 0.89, 0.83, and 0.86, respectively. BWAQ-ML has a Cronbach's alpha of 0.84, a correlation with RPQ of 0.514, and a correlation with PTM of -0.042, suggesting good measurement performance. The BWAQ-ML can effectively measure individual aggression, and its smaller number of items improves the measurement efficiency for large samples and reduces the frequency of IER occurrence. It can be used as a convenient tool for early adolescent aggression identification and intervention.

Wearable Intelligent Machine Learning Rehabilitation Assessment for Stroke Patients Compared with Clinician Assessment

In order to solve the shortcomings of the current clinical scale assessment for stroke patients, such as excessive time consumption, strong subjectivity, and coarse grading, this study designed an intelligent rehabilitation assessment system based on wearable devices and a machine learning algorithm and explored the effectiveness of the system in assessing patients’ rehabilitation outcomes. The accuracy and effectiveness of the intelligent rehabilitation assessment system were verified by comparing the consistency and time between the designed intelligent rehabilitation assessment system scores and the clinical Fugl−Meyer assessment (FMA) scores. A total of 120 stroke patients from two hospitals participated as volunteers in the trial study, and statistical analyses of the two assessment methods were performed. The results showed that the R2 of the total score regression analysis for both methods was 0.9667, 95% CI 0.92−0.98, p < 0.001, and the mean of the deviation was 0.30, 95% CI 0.57−1.17. The percentages of deviations/relative deviations falling within the mean ± 1.96 SD of deviations/relative deviations were 92.50% and 95.83%, respectively. The mean time for system assessment was 35.00% less than that for clinician assessment, p < 0.05. Therefore, wearable intelligent machine learning rehabilitation assessment has a strong and significant correlation with clinician assessment, and the time spent is significantly reduced, which provides an accurate, objective, and effective solution for clinical rehabilitation assessment and remote rehabilitation without the presence of physicians.