Multimodal Neural and Behavioral Data Predict Response to Rehabilitation in Chronic Poststroke Aphasia

Connected Speech Fluency in Poststroke and Progressive Aphasia: A Scoping Review of Quantitative Approaches and Features

PURPOSE

Speech fluency has important diagnostic implications for individuals with poststroke aphasia (PSA) as well as primary progressive aphasia (PPA), and quantitative assessment of connected speech has emerged as a widely used approach across both etiologies. The purpose of this review was to provide a clearer picture on the range, nature, and utility of individual quantitative speech/language measures and methods used to assess connected speech fluency in PSA and PPA, and to compare approaches across etiologies.

METHOD

We conducted a scoping review of literature published between 2012 and 2022 following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews guidelines. Forty-five studies were included in the review. Literature was charted and summarized by etiology and characteristics of included patient populations and method(s) used for derivation and analysis of speech/language features. For a subset of included articles, we also charted the individual quantitative speech/language features reported and the level of significance of reported results.

RESULTS

Results showed that similar methodological approaches have been used to quantify connected speech fluency in both PSA and PPA. Two hundred nine individual speech-language features were analyzed in total, with low levels of convergence across etiology on specific features but greater agreement on the most salient features. The most useful features for differentiating fluent from nonfluent aphasia in both PSA and PPA were features related to overall speech quantity, speech rate, or grammatical competence.

CONCLUSIONS

Data from this review demonstrate the feasibility and utility of quantitative approaches to index connected speech fluency in PSA and PPA. We identified emergent trends toward automated analysis methods and data-driven approaches, which offer promising avenues for clinical translation of quantitative approaches. There is a further need for improved consensus on which subset of individual features might be most clinically useful for assessment and monitoring of fluency.

SUPPLEMENTAL MATERIAL

https://doi.org/10.23641/asha.25537237.

Resting-state brain network connectivity is an independent predictor of responsiveness to language therapy in chronic post-stroke aphasia

Post-stroke aphasia recovery, especially in the chronic phase, is challenging to predict. Functional integrity of the brain and brain network topology have been suggested as biomarkers of language recovery. This study sought to investigate functional connectivity in four predefined brain networks (i.e., language, default mode, dorsal attention, and salience networks), in relation to aphasia severity and response to language therapy. Thirty patients with chronic post-stroke aphasia were recruited and received a treatment targeting word finding. Structural and functional brain scans were acquired at baseline and resting state functional connectivity for each network was calculated. Additionally, graph measures quantifying network properties were calculated for each network. These included global efficiency for all networks and average strength and clustering coefficient for the language network. Linear mixed effects models showed that mean functional connectivity in the default mode, dorsal attention, and salience networks as well as graph measures of all four networks are independent predictors of response to therapy. While greater mean functional connectivity and global efficiency of the dorsal attention and salience networks predicted greater treatment response, greater mean functional connectivity and global efficiency in the default mode network predicted poorer treatment response. Results for the language network were more nuanced with more efficient network configurations (as reflected in graph measures), but not mean functional connectivity, predicting greater treatment response. These findings highlight the prognostic value of resting-state functional connectivity in chronic treatment-induced aphasia recovery.

Early within therapy naming probes as a clinically-feasible predictor of anomia treatment response

This study investigated the relationship between early within-therapy probe naming performance and anomia therapy outcomes in individuals with aphasia. Thirty-four adults with chronic, post-stroke aphasia participated in the Aphasia Language Impairment and Functioning Therapy (Aphasia LIFT) programme, comprised of 48 h of comprehensive aphasia therapy. Sets of 30 treated and 30 untreated items identified at baseline were probed during impairment therapy which targeted word retrieval using a combined semantic feature analysis and phonological component analysis approach. Multiple regression models were computed to determine the relationship between baseline language and demographic variables, early within-therapy probe naming performance (measured after 3 h of impairment therapy) and anomia treatment outcomes. Early within-therapy probe naming performance emerged as the strongest predictor of anomia therapy gains at post-therapy and at 1-month follow-up. These findings have important clinical implications, as they suggest that an individual's performance after a brief period of anomia therapy may predict response to intervention. As such, early within-therapy probe naming may provide a quick and accessible tool for clinicians to identify potential response to anomia treatment.

Integrating EEG and Machine Learning to Analyze Brain Changes during the Rehabilitation of Broca's Aphasia

The fusion of electroencephalography (EEG) with machine learning is transforming rehabilitation. Our study introduces a neural network model proficient in distinguishing pre- and post-rehabilitation states in patients with Broca's aphasia, based on brain connectivity metrics derived from EEG recordings during verbal and spatial working memory tasks. The Granger causality (GC), phase-locking value (PLV), weighted phase-lag index (wPLI), mutual information (MI), and complex Pearson correlation coefficient (CPCC) across the delta, theta, and low- and high-gamma bands were used (excluding GC, which spanned the entire frequency spectrum). Across eight participants, employing leave-one-out validation for each, we evaluated the intersubject prediction accuracy across all connectivity methods and frequency bands. GC, MI theta, and PLV low-gamma emerged as the top performers, achieving 89.4%, 85.8%, and 82.7% accuracy in classifying verbal working memory task data. Intriguingly, measures designed to eliminate volume conduction exhibited the poorest performance in predicting rehabilitation-induced brain changes. This observation, coupled with variations in model performance across frequency bands, implies that different connectivity measures capture distinct brain processes involved in rehabilitation. The results of this paper contribute to current knowledge by presenting a clear strategy of utilizing limited data to achieve valid and meaningful results of machine learning on post-stroke rehabilitation EEG data, and they show that the differences in classification accuracy likely reflect distinct brain processes underlying rehabilitation after stroke.

Insights gained over 60 years on factors shaping post-stroke aphasia recovery: A commentary on Vignolo (1964)

Aphasia is an acquired language disorder resulting from brain injury, including strokes which is the most common etiology, neurodegenerative diseases, tumors, traumatic brain injury, and resective surgery. Aphasia affects a significant portion of stroke survivors, with approximately one third experiencing its debilitating effects in the long term. Despite its challenges, there is growing evidence that recovery from aphasia is possible, even in the chronic phase of stroke. Sixty years ago, Vignolo (1964) outlined the primary challenges confronted by researchers in this field. These challenges encompassed the absence of an objective evaluation of language difficulties, the scarcity of evidence regarding spontaneous aphasia recovery, and the presence of numerous variables that could potentially influence the process of aphasia recovery. In this paper, we discuss the remarkable progress that has been made in the assessment of language and communication in aphasia as well as in understanding the factors influencing post-stroke aphasia recovery.

Cerebral Small Vessel Disease Burden: An Independent Biomarker for Anomia Treatment Responsiveness in Chronic Stroke Patients With Aphasia

OBJECTIVE

To determine whether MRI-based cerebral small vessel disease (cSVD) burden predicts treatment-induced aphasia recovery in chronic stroke patients above and beyond initial aphasia severity and stroke-lesion volume.

DESIGN

Retrospective. Four cSVD neuroimaging markers were rated using validated visual scales: white matter hyperintensities, enlarged perivascular spaces, lacunes, and global cortical atrophy. We also calculated a cSVD total score. We employed linear regression models to model treatment response as a function of cSVD burden. We also ran correlation analyses to determine the association among cSVD burden and pre-treatment linguistic and non-linguistic cognition.

SETTING

Research clinic.

PARTICIPANTS

The study includes data from 30 chronic stroke patients with aphasia who received treatment for word finding difficulties and completed additional pre-treatment neuroimaging and behavioral assessments (N=30).

INTERVENTIONS

120-minute sessions of anomia treatment 2 times per week for up to 12 weeks.

MAIN OUTCOME MEASURES

Change in accuracy on the treatment probes measured as a percentage (ie, change in accuracy percentage score=post-treatment accuracy percentage minus pre-treatment accuracy percentage).

RESULTS

Baseline cSVD burden predicted response to anomia treatment independently from demographic and stroke-related factors. Patients with lower cSVD burden exhibited enhanced rehabilitation response compared with those with higher cSVD burden (β=-6.816e-02, P=.019). cSVD burden was highly associated with nonverbal executive function at baseline (r=-0.49, P=.005): patients with lower cSVD burden exhibited higher performance on nonverbal executive function tasks compared with participants with higher cSVD burden. No association was observed among cSVD burden and performance on language tasks at the baseline.

CONCLUSIONS

cSVD, a marker of brain reserve and a robust risk factor for post-stroke dementia, may be used as a biomarker for distinguishing patients who are more likely to respond to anomia therapy from those who are less likely to do so and for individualizing treatment parameters (eg, targeting both linguistic and nonlinguistic cognition in severe cSVD).

Therapies Targeting Stroke Recovery

Stroke recovery therapeutics include many classes of intervention and numerous treatment targets. Stroke is a very heterogeneous disease. As such, stroke recovery therapeutics benefit from a personalized medicine approach that considers intersubject differences, such as in infarct location or stroke severity, when assigning treatment. Prediction of treatment responders can be improved by incorporating biological measures, such as neural injury and neural function, as the bedside behavioral phenotype has an incomplete relationship with the biological events underlying stroke recovery. Another ramification of high variability between patients is the need to examine effects of restorative therapies in relation to dose, time poststroke, and stroke severity in clinical trials. For example, enrollment across a wide time interval poststroke or in a population with a very broad range of deficits means high variance across patients in the biological state of the brain. The doses of rehabilitation therapy being studied are often low; it takes substantial practice to acquire a skill in the healthy brain; this is more, not less, pronounced after a stroke. Recognition and treatment of poststroke depression represents a major unmet need. These points are considered in the context of a review of recent advances in stroke recovery therapeutics.

Predicting Outcome in Patients with Brain Injury: Differences between Machine Learning versus Conventional Statistics

Defining reliable tools for early prediction of outcome is the main target for physicians to guide care decisions in patients with brain injury. The application of machine learning (ML) is rapidly increasing in this field of study, but with a poor translation to clinical practice. This is basically dependent on the uncertainty about the advantages of this novel technique with respect to traditional approaches. In this review we address the main differences between ML techniques and traditional statistics (such as logistic regression, LR) applied for predicting outcome in patients with stroke and traumatic brain injury (TBI). Thirteen papers directly addressing the different performance among ML and LR methods were included in this review. Basically, ML algorithms do not outperform traditional regression approaches for outcome prediction in brain injury. Better performance of specific ML algorithms (such as Artificial neural networks) was mainly described in the stroke domain, but the high heterogeneity in features extracted from low-dimensional clinical data reduces the enthusiasm for applying this powerful method in clinical practice. To better capture and predict the dynamic changes in patients with brain injury during intensive care courses ML algorithms should be extended to high-dimensional data extracted from neuroimaging (structural and fMRI), EEG and genetics.

Understanding, facilitating and predicting aphasia recovery after rehabilitation

Purpose: This paper reviews several studies whose aim was to understand the nature of language recovery in chronic aphasia and identify predictors of how people may recover their language functions after a brain injury.Method: Several studies that mostly draw from data collected within the Centre for Neurobiology of Language Recovery were reviewed and categorised in four aspects of language impairment and recovery in aphasia: (a) neural markers for language impairment and recovery, (b) language and cognitive markers for language impairment and recovery, (c) effective treatments and (d) predictive modelling of treatment-induced rehabilitation.Result: Language impairment and recovery in stroke-induced aphasia is multi-factorial, including patient-specific and treatment-specific factors. A combination of these factors may help us predict treatment responsiveness even before treatment begins.Conclusion: Continued work on this topic will lead to a better understanding of the mechanisms that underly language impairment and treatment-induced recovery in aphasia, and, consequently, use this information to predict each person's recovery profile trajectory and provide optimal prescriptions regarding the type and dosage of treatment.