SpeakEEG: A low-cost word-based P300 brain-computer interface

Supporting Families of Children Who Use AAC

Access to Justice: An SLP’s Guide to Helping Persons with Complex Communication Needs Voice Their Case

Using a descriptive quantitative approach, this study examined the role of caregiver self-perceived comfort level (SPCL) in the success of augmentative and alternative communication (AAC) system implementation. The role of SPCL as a mediator variable affecting the relationship between the predictor variable (caregiver training resources) and the outcome variable (AAC use) was investigated. Related factors examined as covariates include caregiver demographics (gender, education level, and setting), the primary disability of the child or youth with complex communication needs, the type of AAC system (low-technology vs high-technology), and caregiver involvement in system selection. To examine the contribution of these variables to AAC use, the innovation decision model was used as a guide to propose a new framework describing how these factors are related as well as how they directly or indirectly impact AAC use. Data from a pre-administered Caregiver Training Survey was analyzed to answer the research question. There were 291 participants who completed the original survey. An inspection of the initial data revealed that 52.86% of the overall data was missing. Following Multiple Imputation using the fully conditional specification method, the data was analyzed including test of model fit, and multiple regression to test the direct and indirect effects of the independent variables on the dependent variable. For low-technology AAC, there was no statistically significant direct or indirect effect observed between the study variables. Thus, SPCL cannot be said to mediate the relationship between training resources and low-technology AAC use. For high-technology AAC, although SPCL cannot be said to mediate the relationship between training resource and AAC use, there were statistically significant direct and total effects of training resource on AAC use. The results suggest that SPCL does not mediate the relationship between caregiver training resources and AAC use for both low-technology and high-technology AAC. However, there were a few limitations that must be considered when interpreting the results. The significant amount of missing data meant that there was a significant inability to detect the studied effects using statistical analysis. This was further compounded with the fact that the survey was pre-administered and causal relationships between the variables in the study could not be established before a mediation analysis was conducted. Conducting a mediation analysis with observational data leaves room for multiple confounding variables. These limitations notwithstanding, the role of training resources in the communicative outcomes of children and youth with complex communication needs was established. The nature of this relationship should be further investigated, including the role of SPCL as a moderator that strengthens or weakens the observed relationship between training resources and AAC use.

Augmentative and alternative communication (AAC) technologies provide individuals who have complex communication needs with an effective means to communicate. Yet the effect of these technologies on long-term memory is unclear. In addition, little is known regarding the impact of learning modality on memory performance of individuals who rely on AAC. The aim of this study was to explore the effect of AAC technologies on the visual and auditory verbal long-term memory abilities of 12 young persons who relied on AAC and had intact cognitive abilities. Participants performed 2 verbal memory tasks, in which familiar words were visually or aurally (i.e., auditorily) presented. The words were either actively produced using the AAC system or not produced (merely read or heard; a production effect paradigm). Memory tests followed. A production benefit (higher recognition rates for produced than no-produced words) was documented in both the visual and the auditory tasks. These findings support the active production of words via the AAC system as a memory strategy. Such technique may be easily used in everyday situations as well as in educational contexts. The results showcase the cognitive benefits of AAC system usage and provide significant insights into rehabilitation.

Background and aims: Adults with acquired neurological disorders (stroke, Traumatic Brain Injury ...) develop their verbal communication and literacy capabilities as typical speakers and writers. They use these skills to participate academically, vocationally, recreationally, and socially. Depending upon their neurological condition, they gradually or suddenly lose their speech or language capabilities and are required to rely on Augmentative and Alternative Communication (AAC) systems to meet their communication needs. In addition to the loss of their spoken communication, the impact of their neurological condition on their participation patterns is potentially profound with reduced ability to care for themselves, a reduction or loss of employment, and usually a sudden or gradual restriction of their social networks. AAC is an umbrella term that encompasses the communication methods used to supplement or replace speech or writing for those with impairments in the production or comprehension of spoken or written language. During the past five decades, AAC technologies have been developed to compensate for these natural communication losses. Stroke is one of the main causes of disability in the world. About 20% of stroke patients experience aphasia, with 20-30% of these individuals exhibiting severe communication deficits for at least a portion of their recovery period. Augmentative and Alternative Communication (AAC) encompasses the communication methods used to supplement or replace speech or writing for those with impairments in the production of spoken or written language. Specifically designed Human Computer Interfaces (HCI), as an assistive technology, provides new channels of communication for patients with aphasia, dysarthria, and dyspraxia, when accompanied by movement impairments. In this workshop after stating a science review of the following types of issues: AAC acceptance (individually, culturally); AAC availability; AAC use patterns; AAC limitations are main issues, an AAC technology for post-stroke patients will be presented which developed by a knowledgebase company in Iran. The system allows patients with communication and motor impairment to state their intentions and feelings by a minimum movement in their body, or just by moving their eyes. Different sensors and switches are available to adopt based on the limited ability of the patients. For detecting eye movements, a novel wearable miniaturized system has been developed that is worn as a headband and detects eye movements based on processing electro-oculogram. A high performance graphical user interface has been developed to type letters and numbers in Persian language. The system also provides words prediction, text to speech conversion with natural voice, and sending/receiving messages in the mobile networks for a convenient communication experience. The developed system has been tested successfully by more than 20 patients with different disabilities, and is now commercially available.The proposed system can also help the severely disabled people with amyotrophic lateral sclerosis, quadriplegia, muscular dystrophy or cerebral palsy to communicate with others and mention their intentions, needs and feelings. This low-cost wearable device assures high level of comfort for the user without fatigue and do not need long time training. The system can also be adapted for the patients who can speak, but could not move their hands, to work with the computer and enjoy using the internet.

Wearable electroencephalography (EEG) devices for seizure detection accuracy and reliability are deep learning (DL) applications in the field of epilepsy diagnosis. In this study, we sought to increase the accuracy of seizure detection using advanced DL algorithms on the Children’s Hospital Boston - Massachusetts Institute of Technology (CHB-MIT) EEG database. First, a fully convolutional network (FCN) was trained and assessed using accuracy and recall/precision metrics, and the early stopping technique was used to avoid overfitting. To assess the performance, the FCN was evaluated in terms of various metrics, including accuracy, precision, recall, F1-score, and receiver operating characteristic (ROC)-area under the curve (AUC). In addition, two-dimensional (2D) convolutional neural networks (CNNs) and long short-term memory (LSTM) models were used to model the database, and their performance was thoroughly measured using different metrics, graphs, and confusion matrices. Using LSTM variants, such as standard LSTM, bidirectional LSTM, stacked LSTM, and LSTM attention mechanisms, hybrid convolutional LSTM (ConvLSTM) models were trained and compared. The comparison was conducted based on the training and validation accuracy and loss, as well as the graphs resulting from the precision–recall curves. Apart from DL approaches, EEG signal analysis using time–frequency techniques, such as wavelet transform and short-time Fourier transform, has also been investigated. These methods assisted in the analysis of the time–frequency features of EEG signals in combination with DL models. This study demonstrates that the performance of wearable EEG devices can be augmented using a combination of DL and seizure signal processing techniques. The FCN achieved an accuracy of 92%, with a recall for seizures of 33%, an F1-score of 0.03, and strong ROC-AUC results. The 2D CNN achieved 96% accuracy, a seizure recall of 70%, an F1-score of 0.12, and an ROC-AUC score of 78%. The baseline LSTM struggled with effectiveness at 53% accuracy with a seizure recall of 18%. In contrast, the LSTM model, which incorporated synthetic minority oversampling technique (SMOTE) balancing, was able to reach improvements of up to 89% accuracy, with a precision of 91%, a recall of 86%, an F1-score of 0.89, and a strong ROC curve performance. Among the models, the LSTM with SMOTE was the best performer, with 89% accuracy, 91% precision, 86% recall, and an F1-score of 0.89. These results provide evidence that applying techniques for data balancing in combination with certain DL network architectures significantly improves the detection of seizures using wearable EEG devices worn on the body. We believe that real-time monitoring and high-performance systems are feasible using optimized DL frameworks. The analysis of the performance of different models allows for the understanding of the possibilities of optimizing the architectures of DL algorithms for the modern diagnosis of epilepsy in real time. The source code used to carry out the experiments is publicly available at CHB-MIT EEG Dataset Python Scripts (https://www.kaggle.com/code/adnankust/adnaneeg1).

BackgroundAugmentative and alternative communication (AAC) is a therapeutic approach and modality of expression for patients with limited or no expressive language. Speech-language pathologists and phoniatricians need to be competent in AAC to treat patients with complex communication needs. For knowledge acquisition and enhancement in AAC, a significant number of e-learning tools are available. To improve e-learning in AAC, it is essential to understand the attributes of these tools, such as formats, content areas, learning styles, or learning goals. However, these structures have yet to be investigated.ObjectiveWith this study, we aimed to (1) explore free access AAC e-learning tools that are appropriate for students and professionals of phoniatrics and speech-language pathology; (2) gain insight into formats, content areas, learning styles, and learning goals; and (3) investigate structural differences within and between basic and advanced learner level.MethodsIn 2023, we conducted a systematic web-based search with defined search terms in PubMed, peDOCS, Google Scholar, Google, the Apple App Store, and the Google Play Store in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 guidelines and piloting a protocol for data abstraction and validation. Inclusion criteria were free access, a mandatory minimum AAC content, and the use of the English or the German language. Social networks, video-sharing platforms, blogs, and forums were excluded. We analyzed formats (websites, online courses, apps, and podcasts), content areas (types of AAC, diagnostics, therapy, and other content areas), learning styles (visual, auditory, and audio-visual), and learning goals (receptive and performative) within and between basic and advanced level tools.ResultsWe identified 131 tools, of which 57 (43.5%) were basic level and 74 (56.5%) were advanced level. Of these 131 tools, 105 (80.2%) were websites, 21 (16%) were online courses, 3 (2.3%) were apps and 2 (1.5%) were podcasts. Only 12 out of 74 (16.2%) tools for advanced learners offered performative tasks. For basic learners no such tasks could be identified. For learning style, all basic tools and most of the advanced level tools were “visual (text)” (57/57, 100% basic vs 66/74, 89.2% advanced). In terms of content, advanced level tools pertained more often to “diagnostics” (28/57, 49.1% basic vs 65/74, 87.8% advanced) and “therapy” (17/57, 29.8% basic vs 64/74, 86.5% advanced). Advanced level courses were more likely online courses (2/57, 3.5% basic vs 19/74, 25.7% advanced) and more often showed audio-visual learning styles compared with basic level tools (5/57, 8.8% basic vs 27/74, 36.5% advanced).ConclusionsOur study showed that free-access AAC tools for phoniatrics and speech-language pathology varied in formats, content areas, learning styles, and learning goals. Furthermore, we found differences within and between learner levels. Thus, we established a basis for future research in e-learning in AAC.

Augmentative and alternative communication (AAC) in pediatric cochlear implant recipients with complex needs: A scoping review.

Augmentative and alternative communication (AAC) technologies can help bridge the gap between severe speech impairment with poor intelligibility and true language potential. By using AAC, preliterate children with intact receptive language skills can rapidly learn to produce novel, grammatically correct graphic symbol sentences. Two case studies from a larger five-year investigation designed to teach preliterate preschool children to produce sentences using a graphic symbol-based AAC application are presented. The first case is a four-year-old, presenting with intact receptive language skills and speech impairment, resulting in 26% intelligibility with context and 6% without context on the Index of Augmented Speech Comprehensibility in Children, I-ASCC. The second case is a five-year-old with Down syndrome and speech intelligibility of 19% with context and 13% without context on the I-ASCC. The AAC Generative Language Intervention (AAC-GLI), supporting production of sentences using a graphic symbol-based AAC application is presented, along with pre- and postintervention language analysis of aided utterances. Participants received a total of 21 to 28 play-based aided language intervention sessions over three to four months. Both boys demonstrated significant progress and growth in their semantic and grammatical development using graphic symbols, as evidenced through the use of accurate word order, relevant symbol use, utterance length, and range of vocabulary across word classes.

ABSTRACTWe report on how functional capacity of communication and the use of augmentative and alternative communication (AAC) technology changed in persons with amyotrophic lateral sclerosis (ALS) and progressive dysarthria during a period of two years. The initial sample comprised 30 participants, nine of them survived the whole two years. The data was gathered in 157 communication assessments by a speech-language pathologist. Functional capacity of communication was evaluated with combined severity scales of speech and upper and lower limb function. Communication strategies used were gathered based on the participants’ own reports, and their use was confirmed during the clinical visits. A total of 57% of the participants either supported or substituted their natural speech with high- or low- technology strategies. The majority of them reported that they used high-technology rather than low-technology AAC as the main communication means after speech had deteriorated. The classification of functional capacity of communication changed in 63% of participants, more in bulbar than spinal-onset ALS as the disease progressed. Natural speech is a strong means of communication and not effortlessly replaceable. However, AAC means, such as handwriting, text-to-speech devices and computer-based devices, are essential for persons with ALS as their speech deteriorates. Several AAC strategies to augment deteriorated speech and replace natural speech are often needed. Functional capacity of communication and its individual changes along the disease progression must be taken into account during the continuing communication aid processes.

Augmentative and alternative communication (AAC) is the study of the compensatory strategies for augmenting, supporting, or replacing communication via speech, e.g. speech generating devices (SGDs) and communication books. Because provision of AAC services (such as evaluation, treatment, and AAC system or device determination) is part of the scope of practice for speech and language pathologists (SLPs), it is important that students entering this field feel competent in their AAC knowledge. The purpose of this study was to examine the impact of additional components of AAC education (i.e., workshops and clinical practica) on SLP students’ preservice confidence levels in providing these services. An online survey was distributed via email to 49 SLP graduate students attending a large metropolitan university. The survey included questions in which students were asked to identify their education experiences, and then rate their levels of confidence in providing AAC services for individuals with complex communication needs (CCN). The results show trends indicating that additional clinical practica in AAC impacted student’s confidence levels in treating individuals with CCN and in determining appropriate AAC systems for the student. In conclusion, it is likely that required pre-service education in AAC needs to be more robust, especially in the area of device/system determination.

Prior research has highlighted the positive impact of augmentative and alternative communication (AAC) on speech and language development for children who cannot use speech alone. However, little is known about AAC use specifically for children with childhood apraxia of speech (CAS). Therefore, the purpose of this scoping review was to identify and synthesize twenty years of research investigating AAC use in children with CAS and determine gaps. A systematic search strategy was employed to identify relevant literature in five databases. and full-text screening was conducted by all authors. Forward and backward citation chaining was completed resulting in a total of 11 eligible studies. Study results were synthesized within three categories: therapy focused, profile and perspective studies. All studies highlighted that early AAC intervention may optimize communication outcomes for children with CAS. Therapy focused studies demonstrated that AAC fulfills a complementary role in the management of CAS. Perspective studies demonstrated that there are a variety of factors which influenced a SLP’s decision making surrounding AAC implementation for children with CAS. While prior systematic reviews have demonstrated the positive impact of AAC on speech production broadly, this review uniquely focuses on AAC use in children with CAS.

배경 및 목적: 말실행증을 동반한 표현성 실어증 환자는 말 산출에 어려움을 보여 일상생활에서 의사소통에 심한 어려움을 보이지만 보완대체의사소통(augmentative and alternative communication; 이하 AAC)을 적용하면 기능적인 의사소통 능력을 향상시킬 수 있다. 본 연구는 3명의 실어증 환자를 대상으로 AAC를 적용한 과정과 경과를 보고하고, 실어증에 서 AAC의 적용가능성에 대해서 알아보고자 하였다. <BR> 방법: 고려대학교 안암병원 언어치료실에서 심한 표현성 실어증으로 치료를 받던 3명의 환자를 대상으로 2013년 3월부터 2013년 9월까지 AAC 프로그램인 마이토키를 사용하여 훈련하였다. 사례 1은 72세 남자로 파라다이스`한국판-웨스턴실어증검사에서 실어증지수 10.8점, 사례 2는 52세 여자로 실어증지수 28.8점이었으며, 우뇌 병변 후 발생한 교차성 실어증이었다. 사례 3은 40세 남자로 실어증지수 16.2점으로 대상자 모두 브로카 실어증이었으며, 말실행증을 동반하고 있었다. 각 사례에 대해 개별적인 프로그램에 따라 AAC를 적용하였고, 훈련과정에서 환자의 반응과 치료 전후 의사소통능력의 변화를 기술하였다. <BR> 결과: 사례 1와 2은 각각 15회기와 17회기의 AAC훈련을 적용하였고, 치료 상황에서 자발적으로 간단한 의사표현을 하는 데에는 훈련이 가능하였다. 그러나 일상생활에서 의사소통기능의 변화는 관찰되지 않았고 사지실행증과 보속증으로 인해 프로그램을 학습하고 조작하는데 지속적인 어려움을 보였다. 사례 3은 총 8회기의 치료를 시행하였으나, AAC에 대한 요구도가 낮아 훈련을 지속하지 못하고 언어 이해와 말 산출 능력의 향상을 위한 언어치료로 전환하여 시행하였다. <BR> 논의 및 결론: 많은 실어증 환자는 언어기능 외에도 집행기능을 포함한 인지기능의 손상을 흔히 동반하게 된다. 실어증 환자에서 성공적으로 AAC를 적용하기 위해서는 실어증의 심한 정도 외에도 환자의 동기와 요구도, 인지기능의 손상여부, 실행증의 유무 등에 대한 평가를 통해 환자를 선별하는 절차가 필요하며, 실어증 환자의 특성에 따라 체계적으로 적용할 수 있는 AAC 훈련방법의 개발과 적절한 치료 전후의 변화 측정, 치료사에 대한 전문적인 교육 등이 중요할 것으로 생각된다.

Non-speaking individuals with motor disabilities heavily rely on augmentative and alternative communication (AAC) text entry systems to communicate. However, designing, developing, and evaluating AAC text entry systems for users and researchers gives rise to several challenges, such as the difficulty of eliciting requirements for a highly heterogeneous user population using a wide variety of assistive devices. This research aims to address such challenges and bridge the gaps between AAC stakeholders while exploring the potential of artificial intelligence (AI)-powered AAC systems to cater to diverse AAC needs and wants. To achieve these goals, we propose an imperfect surrogate user model for AAC system design and evaluation, develop a dedicated machine learning language model for AAC purposes, and design an AI-driven text entry system Tinkerable AAC (TAAC) which is ready for user testing. The objective of these efforts is to ultimately reduce the communication gap between AAC users and their speaking partners.

Purpose Augmentative and alternative communication (AAC) has been shown to support communication and quality of life for persons with aphasia (PWA), yet its use in Jordan remains underexplored. This mixed-methods study aimed to investigate the perspectives of Jordanian speech-language pathologists (SLPs) towards the use of AAC for PWA. Materials and Methods Phase 1 involved a nationwide survey of 142 SLPs, with 101 responses analysed descriptively. Phase 2 of the study comprised two focus group discussions with 10 SLPs whose caseloads predominantly included PWA. Results From Phase 1 of the study, it was noted that only 18.8% of respondents had introduced AAC to PWA, most commonly via mobile applications and gestures. While the majority believed AAC to be beneficial, some SLPs continued to have misconceptions about AAC. Qualitative content analysis of the transcripts from the focus group discussions revealed three themes, which were: (i) views about AAC, (ii) challenges faced when using AAC, and (iii) hopes for the future. Conclusion Jordanian SLPs expressed a strong interest in learning more about AAC and called for its inclusion in pre-service education and ongoing professional development. Findings highlight the need for culturally and linguistically appropriate AAC resources, caregiver training, and professional capacity building to promote AAC adoption in Jordan.