AI-Generated Images for Speech Pathology-An Exploratory Application to Aphasia Assessment and Intervention Materials.

PURPOSE: Images are a core component of aphasia assessment and intervention that require significant resources to produce or source. Text-to-image generation is an Artificial Intelligence (AI) technology that has recently made significant advances and could be a source of low-cost, highly customizable images. The aim of this study was to explore the potential of AI image generation for use in aphasia by examining its efficiency and cost during generation of typical images. METHOD: Two hundred targets (80 nouns, 80 verbs, and 40 sentences) were selected at random from existing aphasia assessments and treatment software. A widely known image generator, DALL-E 2, was given text prompts for each target. The success rate, number of prompts required, and costs were summarized across target categories (noun/verb/sentence) and compared to frequency and imageability. RESULTS: Of 200 targets, 189 (94.5%) successfully conveyed the key concept. The process took a mean of 2.3 min per target at a cost of $0.31 in U.S. dollars each. However, there were aesthetic flaws in many successful images that could impact their utility. Noun images were generated with the highest efficiency and accuracy, followed by verbs, while sentences were more challenging, particularly those with unusual scenes. Patterns of flaws and errors in image generation are discussed. CONCLUSION: The ability to rapidly generate low-cost, high-quality images using AI is likely to be a major contribution to aphasia assessment and treatment going forward, particularly as advances in this technology continue.

Comprehensive Assessment of Reading in Aphasia (CARA) reading questionnaire-German version.

BACKGROUND: Reading comprehension is frequently impaired in persons with aphasia (PWA). For goal-setting and outcome measurement, speech and language therapists (SLTs) need to determine an individual's perspective of their reading difficulties and everyday reading activities. The Comprehensive Assessment of Reading in Aphasia (CARA) reading questionnaire provides a person-centred tool to find out the individual perception of reading functions, reading-related emotions and reading activities in PWA. It was developed and evaluated in English. So far, there is no equivalent instrument in German. AIMS: To translate and adapt the CARA reading questionnaire into German language and culture, to evaluate its practicability and acceptance, and to provide the first psychometric properties of the German version. METHODS & PROCEDURES: Based on translation and adaptation guidelines, we conducted two forward translations that were merged and then adapted. A back translation was prepared and compared with the original version. It was found to be semantically equivalent by one of the authors of the original version. We performed pilot testing with 12 PWA, and the pilot version was adapted according to the comments of these participants. We then collected data on self-reported perception of reading and on psychometric properties of the translated and adapted German version. A total of 22 German-speaking PWA completed the questionnaire at least five times during an intervention study. We analysed retest reliability with Spearman correlation, internal consistency with Cronbach's alpha, internal responsiveness with the standardized response mean, as well as the relationship between outcomes of the questionnaire and text comprehension measures using repeated measures correlations. OUTCOMES & RESULTS: Our data suggest good practicability and acceptance of the German version of the CARA reading questionnaire as well as appropriate validity, reliability and sensitivity to measure therapy-induced change. We found moderate correlations between outcomes of the questionnaire and text-level reading speed. CONCLUSIONS & IMPLICATIONS: The German version of the CARA reading questionnaire could be helpful in intervention planning and goal-setting with German-speaking PWA. By using the questionnaire, SLTs can find out about a person's individual perception of reading difficulties as well as individually relevant reading activities. The questionnaire provides a tool to measure change and is therefore valuable to demonstrate self-reported individual progress. As reading speed seems to be an indicator of personal perception of reading difficulty, it is important to consider reading speed in reading interventions and in reading comprehension assessments. WHAT THIS PAPER ADDS: What is already known on the subject Reading comprehension is frequently impaired in PWA. Reading preferences, the perception of difficulties and the impact on everyday life reading activities are specific to the individual and thus need to be known for goal-setting, intervention planning and monitoring of change. As part of a comprehensive assessment of reading, Morris et al. developed a person-centred English language questionnaire for this purpose. So far, there is no equivalent tool in German. What this paper adds to the existing knowledge In this study, we translated and adapted the questionnaire to German language and culture, and analysed its validity and reliability with German-speaking PWA. We demonstrated that the German version is accessible for German-speaking PWA, and that it has appropriate validity, reliability and sensitivity to measure self-reported change. Outcomes of the questionnaire correlate with text level reading speed. What are the potential or actual clinical implications of this work? The German version of the questionnaire could be a valuable self-reported outcome measure to assess individual perceptions of reading and to measure progress (as perceived by an individual) as a consequence of recovery or intervention in either clinical or research settings. As reading speed might be an indicator of everyday life reading as perceived by an individual, it should be considered in reading assessments and interventions.

Picnic, Accident or Cookies? A Systematic Approach to Guide the Selection of the Picture Definition Tasks in Linguistic Assessment.

OBJECTIVE: Picture description tasks are used to elicit language samples in individuals with aphasia and other cognitive disorders. However, task selection may affect the type of language variables elicited. In this study, our goal is to compare the strengths and the weaknesses of the two internationally used picture description tasks with a novel picture description task developed specifically for Turkish-speaking individuals. MATERIALS AND METHODS: In sum, 20 cognitively healthy individuals with a mean age of 52 ± 14/3 were included. Three different picture description tasks composed of single pictures were used: The Picnic Scene from the Western Aphasia Battery-Revised, the Cookie Theft picture from the Boston Diagnostic Aphasia Examination and the Accident Scene from the Turkish Language Assessment Test for Aphasia. All language samples were recorded using a digital voice recorder. Automated and semi-automated approaches were used for the systematic analysis of the language features that were classified into four levels: general language measures, the morphosyntactic level, the lexicosemantic level and the sentence level. RESULTS: Participants showed greater efficiency and produced the greatest number of subordinate clauses, derivational suffixes, passive voice suffixes and relative past tense construction during the Accident Scene description. On the lexicosemantic level, the Picnic Scene was particularly superior to the Cookie Theft Picture in eliciting nouns. Imperfective aspect use was less frequent in the Accident Scene description. CONCLUSION: All three tasks had limitations and advantages compared with each other. We recommend the development of novel picture description tasks that would be more effective in eliciting specific language features in Turkish-speaking individuals.

A Comparative Investigation of Automatic Speech Recognition Platforms for Aphasia Assessment Batteries.

The rehabilitation of aphasics is fundamentally based on the assessment of speech impairment. Developing methods for assessing speech impairment automatically is important due to the growing number of stroke cases each year. Traditionally, aphasia is assessed manually using one of the well-known assessment batteries, such as the Western Aphasia Battery (WAB), the Chinese Rehabilitation Research Center Aphasia Examination (CRRCAE), and the Boston Diagnostic Aphasia Examination (BDAE). In aphasia testing, a speech-language pathologist (SLP) administers multiple subtests to assess people with aphasia (PWA). The traditional assessment is a resource-intensive process that requires the presence of an SLP. Thus, automating the assessment of aphasia is essential. This paper evaluated and compared custom machine learning (ML) speech recognition algorithms against off-the-shelf platforms using healthy and aphasic speech datasets on the naming and repetition subtests of the aphasia battery. Convolutional neural networks (CNN) and linear discriminant analysis (LDA) are the customized ML algorithms, while Microsoft Azure and Google speech recognition are off-the-shelf platforms. The results of this study demonstrated that CNN-based speech recognition algorithms outperform LDA and off-the-shelf platforms. The ResNet-50 architecture of CNN yielded an accuracy of 99.64 ± 0.26% on the healthy dataset. Even though Microsoft Azure was not trained on the same healthy dataset, it still generated comparable results to the LDA and superior results to Google's speech recognition platform.

Automatic Assessment of Aphasic Speech Sensed by Audio Sensors for Classification into Aphasia Severity Levels to Recommend Speech Therapies.

Aphasia is a type of speech disorder that can cause speech defects in a person. Identifying the severity level of the aphasia patient is critical for the rehabilitation process. In this research, we identify ten aphasia severity levels motivated by specific speech therapies based on the presence or absence of identified characteristics in aphasic speech in order to give more specific treatment to the patient. In the aphasia severity level classification process, we experiment on different speech feature extraction techniques, lengths of input audio samples, and machine learning classifiers toward classification performance. Aphasic speech is required to be sensed by an audio sensor and then recorded and divided into audio frames and passed through an audio feature extractor before feeding into the machine learning classifier. According to the results, the mel frequency cepstral coefficient (MFCC) is the most suitable audio feature extraction method for the aphasic speech level classification process, as it outperformed the classification performance of all mel-spectrogram, chroma, and zero crossing rates by a large margin. Furthermore, the classification performance is higher when 20 s audio samples are used compared with 10 s chunks, even though the performance gap is narrow. Finally, the deep neural network approach resulted in the best classification performance, which was slightly better than both K-nearest neighbor (KNN) and random forest classifiers, and it was significantly better than decision tree algorithms. Therefore, the study shows that aphasia level classification can be completed with accuracy, precision, recall, and F1-score values of 0.99 using MFCC for 20 s audio samples using the deep neural network approach in order to recommend corresponding speech therapy for the identified level. A web application was developed for English-speaking aphasia patients to self-diagnose the severity level and engage in speech therapies.

Formal language assessment in low-educated persons with aphasia: can the lesion effect be distinguished from the education effect?

BACKGROUND: Language tests are important in the assessment and follow up of people with aphasia (PWA). However, language assessment in the low literacy population is still a challenge. OBJECTIVE: To investigate whether a formal evaluation of aphasia is able to distinguish the neurological effect from the effect of low educational level in people with post-stroke aphasia. METHODS: The sample consisted of a group of 30 aphasic subjects (AG) and a control group (CG) of 36 individuals, both with an educational level of 1-4 years. The Brazilian Montreal-Toulouse Language Assessment battery was applied to all subjects. RESULTS: There were statistically significant differences between the groups in 19 out of the 20 tasks analyzed. CONCLUSIONS: These results suggest that formal evaluation procedures are able to detect language disorders resulting from stroke, even in subjects with low educational level.

Formal language assessment in low-educated persons with aphasia: can the lesion effect be distinguished from the education effect? / Avaliação da linguagem em pacientes afásicos com baixa escolaridade: o impacto da lesão neurológica pode ser discriminado da baixa escolaridade?

ABSTRACT Background: Language tests are important in the assessment and follow up of people with aphasia (PWA). However, language assessment in the low literacy population is still a challenge. Objective: To investigate whether a formal evaluation of aphasia is able to distinguish the neurological effect from the effect of low educational level in people with post-stroke aphasia. Methods: The sample consisted of a group of 30 aphasic subjects (AG) and a control group (CG) of 36 individuals, both with an educational level of 1-4 years. The Brazilian Montreal-Toulouse Language Assessment battery was applied to all subjects. Results: There were statistically significant differences between the groups in 19 out of the 20 tasks analyzed. Conclusions: These results suggest that formal evaluation procedures are able to detect language disorders resulting from stroke, even in subjects with low educational level.

RESUMO Antecedentes: Os testes de linguagem são importantes para a avaliação e o acompanhamento de pacientes afásicos. Apesar disso, a avaliação de linguagem em indivíduos com baixa escolaridade ainda é um desafio. Objetivo: Investigar se a avaliação formal da afasia é capaz de diferenciar o efeito da lesão neurológica versus o efeito da baixa escolaridade em pacientes afásicos, acometidos por acidente vascular cerebral (AVC). Métodos: A amostra foi composta de um grupo de 30 sujeitos afásicos (AG) e um grupo controle (CG) de 36 indivíduos, todos com um a quatro anos de escolaridade. A Bateria Montreal-Toulouse de Avaliação da Linguagem foi administrada a todos os participantes. Resultados: Das 20 tarefas analisadas, 19 apresentaram diferenças significativas entre os grupos. Conclusões: Os resultados sugerem que procedimentos formais de avaliação são capazes de identificar as alterações linguísticas ocasionadas por um AVC, também em pacientes com baixa escolaridade.

Assessment of language impairment and function.

This chapter is written for the qualified neurologist or related professional working with persons who have had a stroke or other sudden brain injury. It is critical that the presence of aphasia is detected, no matter how mild the presentation, and to support that assertion, this chapter highlights the plight of persons with latent aphasia. At the individual level, the impact of aphasia is devastating, with overwhelming evidence that aphasia negatively impacts psychosocial outcomes. At the global level, sensitive detection and accurate diagnosis of aphasia are critical for accurate characterization and quantification of the global burden of aphasia. The word "LANGUAGE" is leveraged as an acronym to create a useful and memorable checklist to guide navigation of aphasia screening and assessment: it begins with the definition of language (L), followed by the definition and diagnostic criteria for aphasia (A). Then language abilities and characteristics to be considered in assessment are presented: naming (N); grammar and syntax (G); unintelligible words, jargon, and paraphasias (U); auditory comprehension and repetition (A); graphemic abilities-reading and writing (G); and everyday communication and discourse (E). Recommendations for improving procedural adherence are provided, and a list of potential brief assessment measures are introduced.

Clinical assessment and screening of stroke patients with aphasia: a best practice implementation project.

INTRODUCTION AND AIMS: As a critical form of stroke damage, aphasia negatively impacts stroke patients' return to society. Speech and language intervention has been found to assist in optimizing poststroke aphasia patient outcomes; consequently, early identification and diagnosis are vital for poststroke aphasia to ensure that patients receive the rehabilitation they require. This project aimed to promote evidence-based practice (EBP) in the assessment and screening of stroke patients with aphasia and to improve the clinical outcomes of patients who suffer from poststroke aphasia in a large tertiary hospital. METHODS: The current evidence implementation project was conducted in the neurology and rehabilitation departments of a tertiary hospital in China. Six audit criteria were developed for the baseline and follow-up audits. The project used the JBI PACES software, as well as JBI's Getting Research into Practice audit and feedback tool, to foster evidence-based healthcare in practice. RESULTS: Although the performance of all evidence-based criteria during the baseline audit was poor, barriers were identified through baseline, and the project team carried out and implemented developed strategies following Getting Research into Practice resources. All the criteria improved from baseline after the follow-up cycle, with four out of six criteria achieving a compliance rate of 100%, and two evidence-based criteria recorded at 73 and 80% compliance, respectively. CONCLUSION: The current project successfully increased EBP for the assessment and screening of stroke patients with aphasia. Further studies are needed to ensure the project's long-term sustainability.

Performance of the vision, aphasia, neglect (VAN) assessment within a single large EMS system.

BACKGROUND: There is limited evidence on the performance of emergent large-vessel occlusion (LVO) stroke screening tools when used by emergency medical services (EMS) and emergency department (ED) providers. We assessed the validity and predictive value of the vision, aphasia, neglect (VAN) assessment when completed by EMS and in the ED among suspected stroke patients. METHODS: We conducted a retrospective study of VAN performed by EMS providers and VAN inferred from the National Institutes of Health Stroke Scale performed by ED nurses at a single hospital. We calculated sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of VAN by EMS and in the ED for LVO and a combined LVO and intracerebral hemorrhage (ICH) outcome. RESULTS: From January 2018 to June 2020, 1,547 eligible patients were identified. Sensitivity and specificity of ED VAN were similar for LVO (72% and 74%, respectively), whereas EMS VAN was more sensitive (84%) than specific (68%). PPVs were low for both EMS VAN (26%) and ED VAN (21%) to detect LVO. Due to several VAN-positive ICHs, PPVs were substantially higher for both EMS VAN (44%) and ED VAN (39%) to detect LVO or ICH. EMS and ED VAN had high NPVs (97% and 96%, respectively). CONCLUSIONS: Among suspected stroke patients, we found modest sensitivity and specificity of VAN to detect LVO for both EMS and ED providers. Moreover, the low PPV in our study suggests a significant number of patients with non-LVO ischemic stroke or ICH could be over-triaged with VAN.

Assessment of aphasia: dialectal and cultural considerations in neurology.

INTRODUCTION: In the context of neurological disorders, language assessment is traditionally based on the use of the standard language of the community in which the assessment is performed. Dialectal variables constitute a "forgotten" topic. This review aims to provide resources and working guidelines to overcome this limitation. DEVELOPMENT: The study was conducted as follows: 1) review of basic concepts of dialectology (language, dialect, isogloss line, idiolect, register, "standard variant," bilingualism); 2) characterisation of Spanish and Catalan dialectologies and search for online resources for the development of neurological assessment tools; 3) ad hoc search on the PubMed database (keywords: "dialect," "aphasia," and "multilingual"). CONCLUSIONS: We provide online resources to promote adequate development of assessment tools in Spanish and Catalan, accounting for dialectal considerations. We propose a set of working guidelines for the introduction of dialectal variables in language tests to ensure their ultimate objective: to evaluate the real use of the language, and not cultural knowledge of the standard language.

Spoken Discourse Assessment and Analysis in Aphasia: An International Survey of Current Practices.

Purpose Spoken discourse analysis is commonly employed in the assessment and treatment of people living with aphasia, yet there is no standardization in assessment, analysis, or reporting procedures, thereby precluding comparison/meta-analyses of data and hindering replication of findings. An important first step is to identify current practices in collecting and analyzing spoken discourse in aphasia. Thus, this study surveyed current practices, with the goal of working toward standardizing spoken discourse assessment first in research settings with subsequent implementation into clinical settings. Method A mixed-methods (quantitative and qualitative) survey was publicized to researchers and clinicians around the globe who have collected and/or analyzed spoken discourse data in aphasia. The survey data were collected between September and November 2019. Results Of the 201 individuals who consented to participate, 189 completed all mandatory questions in the survey (with fewer completing nonmandatory response questions). The majority of respondents reported barriers to utilizing discourse including transcription, coding, and analysis. The most common barrier was time (e.g., lack of time). Respondents also indicated that there was a lack of, and a need for, psychometric properties and normative data for spoken discourse use in the assessment and treatment of persons with aphasia. Quantitative and qualitative results are described in detail. Conclusions The current survey study evaluated spoken discourse methods in aphasia across research and clinical settings. Findings from this study will be used to guide development of process standardization in spoken discourse and for the creation of a psychometric and normative property database. Supplemental Material https://doi.org/10.23641/asha.166395100.

Computer Adaptive Testing for the Assessment of Anomia Severity.

Anomia assessment is a fundamental component of clinical practice and research inquiries involving individuals with aphasia, and confrontation naming tasks are among the most commonly used tools for quantifying anomia severity. While currently available confrontation naming tests possess many ideal properties, they are ultimately limited by the overarching psychometric framework they were developed within. Here, we discuss the challenges inherent to confrontation naming tests and present a modern alternative to test development called item response theory (IRT). Key concepts of IRT approaches are reviewed in relation to their relevance to aphasiology, highlighting the ability of IRT to create flexible and efficient tests that yield precise measurements of anomia severity. Empirical evidence from our research group on the application of IRT methods to a commonly used confrontation naming test is discussed, along with future avenues for test development.

Performance Evaluation of Machine Learning Frameworks for Aphasia Assessment.

Speech assessment is an essential part of the rehabilitation procedure for patients with aphasia (PWA). It is a comprehensive and time-consuming process that aims to discriminate between healthy individuals and aphasic patients, determine the type of aphasia syndrome, and determine the patients' impairment severity levels (these are referred to here as aphasia assessment tasks). Hence, the automation of aphasia assessment tasks is essential. In this study, the performance of three automatic speech assessment models based on the speech dataset-type was investigated. Three types of datasets were used: healthy subjects' dataset, aphasic patients' dataset, and a combination of healthy and aphasic datasets. Two machine learning (ML)-based frameworks, classical machine learning (CML) and deep neural network (DNN), were considered in the design of the proposed speech assessment models. In this paper, the DNN-based framework was based on a convolutional neural network (CNN). Direct or indirect transformation of these models to achieve the aphasia assessment tasks was investigated. Comparative performance results for each of the speech assessment models showed that quadrature-based high-resolution time-frequency images with a CNN framework outperformed all the CML frameworks over the three dataset-types. The CNN-based framework reported an accuracy of 99.23 ± 0.003% with the healthy individuals' dataset and 67.78 ± 0.047% with the aphasic patients' dataset. Moreover, direct or transformed relationships between the proposed speech assessment models and the aphasia assessment tasks are attainable, given a suitable dataset-type, a reasonably sized dataset, and appropriate decision logic in the ML framework.

Standardizing Assessment of Spoken Discourse in Aphasia: A Working Group With Deliverables.

Purpose The heterogeneous nature of measures, methods, and analyses reported in the aphasia spoken discourse literature precludes comparison of outcomes across studies (e.g., meta-analyses) and inhibits replication. Furthermore, funding and time constraints significantly hinder collecting test-retest data on spoken discourse outcomes. This research note describes the development and structure of a working group, designed to address major gaps in the spoken discourse aphasia literature, including a lack of standardization in methodology, analysis, and reporting, as well as nominal data regarding the psychometric properties of spoken discourse outcomes. Method The initial initiatives for this working group are to (a) propose recommendations regarding standardization of spoken discourse collection, analysis, and reporting in aphasia, based on the results of an international survey and a systematic literature review and (b) create a database of test-retest spoken discourse data from individuals with and without aphasia. The survey of spoken discourse collection, analysis, and interpretation procedures was distributed to clinicians and researchers involved in aphasia assessment and rehabilitation from September to November 2019. We will publish survey results and recommend standards for collecting, analyzing, and reporting spoken discourse in aphasia. A multisite endeavor to collect test-retest spoken discourse data from individuals with and without aphasia will be initiated. This test-retest information will be contributed to a central site for transcription and analysis, and data will be subsequently openly curated. Conclusion The goal of the working group is to create recommendations for field-wide standards in methods, analysis, and reporting of spoken discourse outcomes, as has been done across other related disciplines (e.g., Consolidated Standards of Reporting Trials, Enhancing the Quality and Transparency of Health Research, Committee on Best Practice in Data Analysis and Sharing). Additionally, the creation of a database through our multisite collaboration will allow the identification of psychometrically sound outcome measures and norms that can be used by clinicians and researchers to assess spoken discourse abilities in aphasia.

The City Gesture Checklist: The development of a novel gesture assessment.

BACKGROUND: People with aphasia rely on gesture more than healthy controls to get their message across, but use a limited range of gesture types. Gesture therapy is thus a potential avenue of intervention for people with aphasia. However, currently no gesture assessment evaluates how they use gesture. Such a tool could inform therapy targets and measure outcomes. In gesture research, many different coding categories are used to describe gesture forms and functions. These coding methods are prohibitively time-consuming to use in clinical practice. There is therefore a need for a 'quick and dirty' method of assessing gesture use. AIMS: To investigate current practice among UK-based clinicians (speech and language therapists) in relation to gesture assessment and therapy, to synthesize gesture-coding frameworks used in aphasia research, to develop a gesture checklist based on the synthesized coding frameworks suitable for use in clinical practice, and to investigate the interrater reliability (IRR) of the checklist among experienced and unfamiliar users. METHODS & PROCEDURES: The research team synthesized seven gesture-coding frameworks and trialled three resulting prototype checklists at a co-design workshop with 20 clinicians. Attending clinicians were also consulted about their current clinical gesture practice using a questionnaire. A final City Gesture Checklist (CGC) was developed based upon outcomes and feedback from the workshop. The IRR of the CGC was evaluated between the research team and 11 further clinicians within a second workshop. Both groups used the CGC to count gestures in video clips of people with aphasia talking to a conversation partner. MAIN CONTRIBUTION: A total of 18 workshop attendees completed the current practice questionnaire. Of these, 10 reported assessing gesture informally and five also used formal assessment. Gesture-coding synthesis highlighted six main categories of gesture form. Clinicians at the co-design workshop provided feedback on prototype checklists regarding the relevance and usability of the gesture categories, layout, use of images and instructions. A final version of the CGC was created incorporating their recommendations. The IRR for the CGC was moderate between both the researchers and clinicians. CONCLUSIONS & IMPLICATIONS: The CGC can be used to assess the types of gesture that people with aphasia produce. The IRR was moderate amongst both experienced users and new users who had received no training. Future research directions include investigating how to improve IRR, evaluating intra-rater reliability and sensitivity to change, and exploring use of the CGC in clinical practice. What this paper adds What is already known on the subject People with aphasia rely on gesture more than healthy speakers, yet use a more limited range of gesture types. Gesture therapy is used by clinicians with the aim of helping people with aphasia to compensate for their language impairment and/or to facilitate speech. What this paper adds to existing knowledge This study explores current gesture assessment practice among UK-based clinicians and synthesizes the coding categories used in the literature about gesture research in aphasia. It describes the development of a novel outcome measure, the CGC, and preliminary testing of its IRR. What are the potential or actual clinical implications of this work? This 'quick and dirty' tool enables clinicians to analyse and record the types of gesture produced by people with aphasia without the need for gesture coding. Preliminary findings suggest that clinicians can use it with a fair degree of reliability by following the checklist's written instructions.

An Efficient Deep Learning Based Method for Speech Assessment of Mandarin-Speaking Aphasic Patients.

Speech assessment is an important part of the rehabilitation process for patients with aphasia (PWA). Mandarin speech lucidity features such as articulation, fluency, and tone influence the meaning of the spoken utterance and overall speech clarity. Automatic assessment of these features is important for an efficient assessment of the aphasic speech. Hence, in this paper, a standardized automatic speech lucidity assessment method for Mandarin-speaking aphasic patients using a machine learning based technique is presented. The proposed assessment method adopts the Chinese Rehabilitation Research Center Aphasia Examination (CRRCAE) standard as a guideline. Quadrature based high-resolution time-frequency images with a convolutional neural network (CNN) are utilized to develop a method that can map the relationship between the severity level of aphasic patients' speech and the three speech lucidity features. The results show a linear relationship with statistically significant correlations between the normalized true-class output activations (TCOA) of the CNN model and patients' articulation, fluency, and tone scores, i.e., 0.71 (p < 0.001), 0.60 (p < 0.001) and 0.58 (p < 0.001), respectively. The linearity of the proposed Mandarin aphasic speech assessment method and its significant correlation with the speech severity levels show the efficacy of the method in predicting the severity of impaired Mandarin speech. The outcome of this research envisages assisting speech-language pathologists in Mandarin-speech impairment assessment and promoting early support discharge; hence could alleviate the stress that the healthcare system is currently experiencing in China nationwide. The framework of the proposed Mandarin aphasic speech assessment method can be readily extended to other languages.

The assessment of severe lexical disorders in Italian individuals with aphasia.

Tests and batteries used in the evaluation of language impairments are overly complex and often ineffective (too difficult) in the assessment of post-stroke patients affected by severe aphasia (global aphasia). The present study reports details on the construction and standardization of a new Italian battery of tasks, specifically designed to assess severe lexical disorders in acquired aphasia (Battery for the Assessment of Severe Acquired Lexical Damage in Italian, BASALDI). The battery is composed of a common set of 64 stimuli (concrete nouns), belonging to both living and non-living categories, and consists of four lexical tasks assessing picture naming, repetition, reading aloud, and oral comprehension. The item selection was based on word frequency, word length, and phonological-articulatory complexity, namely the presence of continuant vs. plosive phones, a variable that may interact with word production in case of severe language damage. Standardization (naming agreement) of a new set of 64 colored images and normative data on Italian healthy subjects pooled across homogenous subgroups for age, gender, and education are reported. Finally, for the four tasks, percentile ranks and z-scores were calculated from a pool of 92 left brain-damaged patients affected by aphasia of different types and severity. The battery allows a fine investigation of lexical disorders, being suitable for diagnostic assessment of mild-to-moderate and severe aphasic lexical deficits, detection of changes over time, and possible dissociations between tasks.

Empirical Evaluation of Computer-Adaptive Alternate Short Forms for the Assessment of Anomia Severity.

Purpose The purpose of this study was to verify the equivalence of 2 alternate test forms with nonoverlapping content generated by an item response theory (IRT)-based computer-adaptive test (CAT). The Philadelphia Naming Test (PNT; Roach, Schwartz, Martin, Grewal, & Brecher, 1996)was utilized as an item bank in a prospective, independent sample of persons with aphasia. Method Two alternate CAT short forms of the PNT were administered to a sample of 25 persons with aphasia who were at least 6 months postonset and received no treatment for 2 weeks before or during the study. The 1st session included administration of a 30-item PNT-CAT, and the 2nd session, conducted approximately 2 weeks later, included a variable-length PNT-CAT that excluded items administered in the 1st session and terminated when the modeled precision of the ability estimate was equal to or greater than the value obtained in the 1st session. The ability estimates were analyzed in a Bayesian framework. Results The 2 test versions correlated highly (r = .89) and obtained means and standard deviations that were not credibly different from one another. The correlation and error variance between the 2 test versions were well predicted by the IRT measurement model. Discussion The results suggest that IRT-based CAT alternate forms may be productively used in the assessment of anomia. IRT methods offer advantages for the efficient and sensitive measurement of change over time. Future work should consider the potential impact of differential item functioning due to person factors and intervention-specific effects, as well as expanding the item bank to maximize the clinical utility of the test. Supplemental Material https://doi.org/10.23641/asha.11368040.

Vision, Aphasia, Neglect Assessment for Large Vessel Occlusion Stroke.

BACKGROUND AND PURPOSE: Vision, Aphasia, Neglect (VAN) is a large vessel occlusion (LVO) screening tool that was initially tested in a small study where emergency department (ED) nurses were trained to perform VAN assessment on stroke code patients. We aimed to validate the VAN assessment in a larger inpatient dataset. METHODS: We utilized a large dataset and used National Institute of Health Stroke Scale (NIHSS) performed by physicians to extrapolate VAN. VAN was compared to NIHSS greater than or equal to 6 and established prehospital LVO screening tools including Rapid Arterial Occlusion Evaluation scale (RACE), Field Assessment Stroke Triage for Emergency Destination (FAST-ED), and Cincinnati Pre-hospital Stroke Scale (CPSS). Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, and area under receiver operating characteristics curve was calculated to estimate the predictive value of LVO. RESULTS: VAN was comparable in sensitivity (79% versus 80%) and NPV (88% versus 87%) to NIHSS greater than or equal to 6. It was superior in specificity (69% versus 57%), PPV (53% versus 46%) and accuracy to NIHSS greater than or equal to 6 (72% versus 64%) with significant receiver operating curve (.74 versus .69, P = .02). VAN also had comparable area under the curve when compared to RACE, FAST-ED, and CPSS however slightly lower accuracy (69%-73%) compared to RACE (76%), FAST-ED (77%), and CPSS (75%). VAN had the highest NPV among all screening assessments (88%). CONCLUSIONS: VAN is a simple screening tool that can identify LVOs with adequate accuracy in hospital setting. Future studies need to be conducted in prehospital setting to validate its utility to detect LVOs in the field.