1
|
Wang Y, Wang H, Li Z, Zhang H, Yang L, Li J, Tang Z, Hou S, Wang Q. Sound as a bell: a deep learning approach for health status classification through speech acoustic biomarkers. Chin Med 2024; 19:101. [PMID: 39049005 PMCID: PMC11267751 DOI: 10.1186/s13020-024-00973-3] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/07/2024] [Accepted: 07/16/2024] [Indexed: 07/27/2024] Open
Abstract
BACKGROUND Human health is a complex, dynamic concept encompassing a spectrum of states influenced by genetic, environmental, physiological, and psychological factors. Traditional Chinese Medicine categorizes health into nine body constitutional types, each reflecting unique balances or imbalances in vital energies, influencing physical, mental, and emotional states. Advances in machine learning models offer promising avenues for diagnosing conditions like Alzheimer's, dementia, and respiratory diseases by analyzing speech patterns, enabling complementary non-invasive disease diagnosis. The study aims to use speech audio to identify subhealth populations characterized by unbalanced constitution types. METHODS Participants, aged 18-45, were selected from the Acoustic Study of Health. Audio recordings were collected using ATR2500X-USB microphones and Praat software. Exclusion criteria included recent illness, dental issues, and specific medical histories. The audio data were preprocessed to Mel-frequency cepstral coefficients (MFCCs) for model training. Three deep learning models-1-Dimensional Convolution Network (Conv1D), 2-Dimensional Convolution Network (Conv2D), and Long Short-Term Memory (LSTM)-were implemented using Python to classify health status. Saliency maps were generated to provide model explainability. RESULTS The study used 1,378 recordings from balanced (healthy) and 1,413 from unbalanced (subhealth) types. The Conv1D model achieved a training accuracy of 91.91% and validation accuracy of 84.19%. The Conv2D model had 96.19% training accuracy and 84.93% validation accuracy. The LSTM model showed 92.79% training accuracy and 87.13% validation accuracy, with early signs of overfitting. AUC scores were 0.92 and 0.94 (Conv1D), 0.99 (Conv2D), and 0.97 (LSTM). All models demonstrated robust performance, with Conv2D excelling in discrimination accuracy. CONCLUSIONS The deep learning classification of human speech audio for health status using body constitution types showed promising results with Conv1D, Conv2D, and LSTM models. Analysis of ROC curves, training accuracy, and validation accuracy showed all models robustly distinguished between balanced and unbalanced constitution types. Conv2D excelled with good accuracy, while Conv1D and LSTM also performed well, affirming their reliability. The study integrates constitution theory and deep learning technologies to classify subhealth populations using noninvasive approach, thereby promoting personalized medicine and early intervention strategies.
Collapse
Affiliation(s)
- Yanbing Wang
- School of Traditional Chinese Medicine, Beijing University of Chinese Medicine, Beijing, 100029, China
| | - Haiyan Wang
- School of Traditional Chinese Medicine, Beijing University of Chinese Medicine, Beijing, 100029, China
| | - Zhuoxuan Li
- School of Traditional Chinese Medicine, Beijing University of Chinese Medicine, Beijing, 100029, China
| | - Haoran Zhang
- School of Management, Beijing University of Chinese Medicine, Beijing, 100029, China
| | - Liwen Yang
- School of Traditional Chinese Medicine, Beijing University of Chinese Medicine, Beijing, 100029, China
| | - Jiarui Li
- School of Traditional Chinese Medicine, Beijing University of Chinese Medicine, Beijing, 100029, China
| | - Zixiang Tang
- School of Traditional Chinese Medicine, Beijing University of Chinese Medicine, Beijing, 100029, China
| | - Shujuan Hou
- National Institute of TCM Constitution and Preventive Medicine, Beijing University of Chinese Medicine, Beijing, 100029, China.
| | - Qi Wang
- National Institute of TCM Constitution and Preventive Medicine, Beijing University of Chinese Medicine, Beijing, 100029, China.
| |
Collapse
|
2
|
Millman LSM, Williams IA, Jungilligens J, Pick S. Neurocognitive performance in functional neurological disorder: A systematic review and meta-analysis. Eur J Neurol 2024:e16386. [PMID: 38953473 DOI: 10.1111/ene.16386] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/08/2023] [Revised: 04/10/2024] [Accepted: 05/30/2024] [Indexed: 07/04/2024]
Abstract
BACKGROUND AND PURPOSE Cognitive complaints are common in functional neurological disorder (FND), but it is unclear whether objective neurocognitive deficits are present. This systematic review summarized validated/standardized cognitive test performance in FND samples across cognitive domains. METHODS Embase, PsycInfo and MEDLINE were searched from inception to 15 May 2023, combining terms for FND and cognitive domains (e.g., attention, memory, executive functioning). Studies included a range of FND phenotypes (seizures, motor, cognitive disorder, mixed), compared to healthy or clinical controls. Risk of bias was assessed with the modified Newcastle-Ottawa Scale and a qualitative synthesis/narrative review of cognitive performance in FND was conducted. Test performance scores were extracted, and random effects meta-analyses were run where appropriate. This review was registered on PROSPERO, CRD42023423139. RESULTS Fifty-six studies including 2260 individuals with FND were eligible. Although evidence for some impairments emerged across domains of executive functioning, attention, memory and psychomotor/processing speed, this was inconsistent across studies and FND phenotypes. Common confounds included group differences in demographics, medication and intellectual functioning. Only 24% of studies objectively assessed performance validity. Meta-analyses revealed higher scores on tests of naming (g = 0.67, 95% confidence interval [CI] 0.50, 0.84) and long-term memory (g = 0.43, 95% CI 0.13, 0.74) in functional seizures versus epilepsy, but no significant differences in working (g = -0.08, 95% CI -0.44, 0.29) or immediate (g = 0.25, 95% CI -0.02, 0.53) memory and cognitive flexibility (g = -0.01, 95% CI -0.29, 0.28). CONCLUSIONS There is mixed evidence for objective cognitive deficits in FND. Future research should control for confounds, include tests of performance validity, and assess relationships between objective and subjective neurocognitive functioning.
Collapse
Affiliation(s)
- L S Merritt Millman
- Institute of Psychiatry, Psychology and Neuroscience, King's College London, London, UK
| | - Isobel A Williams
- Psychology in Healthcare, Newcastle Upon Tyne Hospitals NHS Foundation Trust and the Translational and Clinical Research Institute, Newcastle University, Callaghan, UK
| | - Johannes Jungilligens
- Department of Neurology, University Hospital Knappschaftskrankenhaus, Ruhr University Bochum, Bochum, Germany
| | - Susannah Pick
- Institute of Psychiatry, Psychology and Neuroscience, King's College London, London, UK
| |
Collapse
|
3
|
Cabreira V, Alty J, Antic S, Araújo R, Aybek S, Ball HA, Baslet G, Bhome R, Coebergh J, Dubois B, Edwards M, Filipović SR, Frederiksen KS, Harbo T, Hayhow B, Howard R, Huntley J, Isaacs J, LaFrance WC, Larner AJ, Di Lorenzo F, Main J, Mallam E, Marra C, Massano J, McGrath ER, McWhirter L, Moreira IP, Nobili F, Pennington C, Tábuas-Pereira M, Perez DL, Popkirov S, Rayment D, Rossor M, Russo M, Santana I, Schott J, Scott EP, Taipa R, Tinazzi M, Tomic S, Toniolo S, Tørring CW, Wilkinson T, Frostholm L, Stone J, Carson A. Perspectives on the diagnosis and management of functional cognitive disorder: An international Delphi study. Eur J Neurol 2024:e16318. [PMID: 38700361 DOI: 10.1111/ene.16318] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/11/2023] [Revised: 03/18/2024] [Accepted: 04/11/2024] [Indexed: 05/05/2024]
Abstract
BACKGROUND Current proposed criteria for functional cognitive disorder (FCD) have not been externally validated. We sought to analyse the current perspectives of cognitive specialists in the diagnosis and management of FCD in comparison with neurodegenerative conditions. METHODS International experts in cognitive disorders were invited to assess seven illustrative clinical vignettes containing history and bedside characteristics alone. Participants assigned a probable diagnosis and selected the appropriate investigation and treatment. Qualitative, quantitative and inter-rater agreement analyses were undertaken. RESULTS Eighteen diagnostic terminologies were assigned by 45 cognitive experts from 12 countries with a median of 13 years of experience, across the seven scenarios. Accurate discrimination between FCD and neurodegeneration was observed, independently of background and years of experience: 100% of the neurodegenerative vignettes were correctly classified and 75%-88% of the FCD diagnoses were attributed to non-neurodegenerative causes. There was <50% agreement in the terminology used for FCD, in comparison with 87%-92% agreement for neurodegenerative syndromes. Blood tests and neuropsychological evaluation were the leading diagnostic modalities for FCD. Diagnostic communication, psychotherapy and psychiatry referral were the main suggested management strategies in FCD. CONCLUSIONS Our study demonstrates the feasibility of distinguishing between FCD and neurodegeneration based on relevant patient characteristics and history details. These characteristics need further validation and operationalisation. Heterogeneous labelling and framing pose clinical and research challenges reflecting a lack of agreement in the field. Careful consideration of FCD diagnosis is advised, particularly in the presence of comorbidities. This study informs future research on diagnostic tools and evidence-based interventions.
Collapse
Affiliation(s)
- Verónica Cabreira
- Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, UK
| | - Jane Alty
- Wicking Dementia Research and Education Centre, University of Tasmania, Hobart, Tasmania, Australia
| | - Sonja Antic
- Department of Neurology, Aarhus University Hospital, Aarhus, Denmark
| | - Rui Araújo
- Department of Neurology, Centro Hospitalar Universitário São João, Porto, Portugal
- Department of Clinical Neurosciences and Mental Health, Faculty of Medicine University of Porto, Porto, Portugal
| | - Selma Aybek
- Neurology, Faculty of Sciences and Medicine, Fribourg University, Fribourg, Switzerland
| | - Harriet A Ball
- Population Health Sciences, Bristol Medical School, Bristol, UK
| | - Gaston Baslet
- Department of Psychiatry, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, USA
| | - Rohan Bhome
- Dementia Research Centre, UCL Queen Square Institute of Neurology, London, UK
- Centre for Medical Image Computing, University College London, London, UK
| | - Jan Coebergh
- Department of Neurology, St George's University of London, London, UK
| | - Bruno Dubois
- Department of Neurology, Institut de la mémoire et de la maladie d'Alzheimer (IM2A), AP-HP, Brain Institute, Sorbonne University, Paris, France
| | - Mark Edwards
- Department of Basic and Clinical Neurosciences, Institute of Psychiatry Psychology and Neurosciences, Kings College London, London, UK
| | - Saša R Filipović
- University of Belgrade Institute for Medical Research, Belgrade, Serbia
| | - Kristian Steen Frederiksen
- Clinical Trial Unit, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark
- Department of Clinical Medicine, University of Copenhagen, Copenhagen, Denmark
| | - Thomas Harbo
- Department of Neurology, Aarhus University Hospital, Aarhus, Denmark
| | - Bradleigh Hayhow
- Department of Neurology, Fiona Stanley Hospital, Murdoch, Western Australia, Australia
- School of Medicine, University of Notre Dame Australia, Fremantle, Western Australia, Australia
| | - Robert Howard
- Division of Psychiatry, University College London, London, UK
| | - Jonathan Huntley
- Division of Psychiatry, University College London, London, UK
- Camden and Islington NHS Foundation Trust, London, UK
| | - Jeremy Isaacs
- Department of Neurology, St George's University of London, London, UK
| | - William Curt LaFrance
- Alpert Medical School, Brown University, Providence, Rhode Island, USA
- Neuropsychiatry and Behavioral Neurology, Rhode Island Hospital, Providence, Rhode Island, USA
| | - Andrew J Larner
- Cognitive Function Clinic, Walton Centre for Neurology and Neurosurgery, Liverpool, UK
| | - Francesco Di Lorenzo
- Department of Clinical and Behavioural Neurology, Santa Lucia Foundation IRCCS, Rome, Italy
| | - James Main
- Bristol Dementia Wellbeing Service, Devon Partnership NHS Trust, Bristol, UK
| | | | - Camillo Marra
- Department of Neuroscience, Catholic University of the Sacred Heart, Memory Clinic - Fondazione Policlinico Agostino Gemelli IRCCS, Rome, Italy
| | - João Massano
- Department of Neurology, Centro Hospitalar Universitário São João, Porto, Portugal
- Department of Clinical Neurosciences and Mental Health, Faculty of Medicine University of Porto, Porto, Portugal
| | - Emer R McGrath
- School of Medicine, University of Galway, Galway, Ireland
| | - Laura McWhirter
- Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, UK
| | - Isabel Portela Moreira
- Neurology Department, Private Hospital of Gaia of the Trofa Saúde Group, Vila Nova de Gaia, Portugal
| | - Flavio Nobili
- IRCCS Ospedale Policlinico San Martino, Genoa, Italy
| | - Catherine Pennington
- Clinical Lecturer, University of Edinburgh, Edinburgh, UK
- Neurology Department, NHS Forth Valley, Larbert, UK
- Department of Clinical Neurosciences, NHS Lothian, Edinburgh, UK
| | - Miguel Tábuas-Pereira
- Neurology Department, Centro Hospitalar e Universitário de Coimbra, Praceta Prof. Mota Pinto, Coimbra, Portugal
- Faculty of Medicine, University of Coimbra, Coimbra, Portugal
- Center for Innovative Biomedicine and Biotechnology, University of Coimbra, Coimbra, Portugal
| | - David L Perez
- Department of Neurology and Psychiatry, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA
| | - Stoyan Popkirov
- Department of Neurology, University Hospital Essen, Essen, Germany
| | - Dane Rayment
- Rosa Burden Centre for Neuropsychiatry, Southmead Hospital, Bristol, UK
| | - Martin Rossor
- Dementia Research Centre, UCL Queen Square Institute of Neurology, London, UK
| | - Mirella Russo
- Department of Neuroscience, Imaging and Clinical Sciences G. d'Annunzio University of Chieti-Pescara, Chieti, Italy
| | - Isabel Santana
- Faculty of Medicine, University of Coimbra, Coimbra, Portugal
- Center for Innovative Biomedicine and Biotechnology, University of Coimbra, Coimbra, Portugal
| | - Jonathan Schott
- Dementia Research Centre, UCL Queen Square Institute of Neurology, London, UK
| | - Emmi P Scott
- Medical University of South Carolina, Charleston, South Carolina, USA
| | - Ricardo Taipa
- Neuropathology Department, Centro Hospitalar Universitário de Santo António, Porto, Portugal
| | - Michele Tinazzi
- Department of Neurosciences, Biomedicine and Movement, University of Verona, Verona, Italy
| | - Svetlana Tomic
- Department of Neurology, University Hospital Center Osijek, Medical School on University of Osijek, Osijek, Croatia
| | - Sofia Toniolo
- Cognitive Disorder Clinic, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, UK
| | | | - Tim Wilkinson
- Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, UK
| | - Lisbeth Frostholm
- Department of Clinical Medicine, Aarhus University Hospital, Aarhus, Denmark
- Department of Functional Disorders and Psychosomatics, Aarhus University Hospital, Aarhus, Denmark
| | - Jon Stone
- Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, UK
| | - Alan Carson
- Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, UK
| |
Collapse
|
4
|
Gagliardi G. Natural language processing techniques for studying language in pathological ageing: A scoping review. INTERNATIONAL JOURNAL OF LANGUAGE & COMMUNICATION DISORDERS 2024; 59:110-122. [PMID: 36960885 DOI: 10.1111/1460-6984.12870] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 01/07/2022] [Accepted: 02/27/2023] [Indexed: 06/18/2023]
Abstract
BACKGROUND In the past few years there has been a growing interest in the employment of verbal productions as digital biomarkers, namely objective, quantifiable behavioural data that can be collected and measured by means of digital devices, allowing for a low-cost pathology detection, classification and monitoring. Numerous research papers have been published on the automatic detection of subtle verbal alteration, starting from written texts, raw speech recordings and transcripts, and such linguistic analysis has been singled out as a cost-effective method for diagnosing dementia and other medical conditions common among elderly patients (e.g., cognitive dysfunctions associated with metabolic disorders, dysarthria). AIMS To provide a critical appraisal and synthesis of evidence concerning the application of natural language processing (NLP) techniques for clinical purposes in the geriatric population. In particular, we discuss the state of the art on studying language in healthy and pathological ageing, focusing on the latest research efforts to build non-intrusive language-based tools for the early identification of cognitive frailty due to dementia. We also discuss some challenges and open problems raised by this approach. METHODS & PROCEDURES We performed a scoping review to examine emerging evidence about this novel domain. Potentially relevant studies published up to November 2021 were identified from the databases of MEDLINE, Cochrane and Web of Science. We also browsed the proceedings of leading international conferences (e.g., ACL, COLING, Interspeech, LREC) from 2017 to 2021, and checked the reference lists of relevant studies and reviews. MAIN CONTRIBUTION The paper provides an introductory, but complete, overview of the application of NLP techniques for studying language disruption due to dementia. We also suggest that this technique can be fruitfully applied to other medical conditions (e.g., cognitive dysfunctions associated with dysarthria, cerebrovascular disease and mood disorders). CONCLUSIONS & IMPLICATIONS Despite several critical points need to be addressed by the scientific community, a growing body of empirical evidence shows that NLP techniques can represent a promising tool for studying language changes in pathological aging, with a high potential to lead a significant shift in clinical practice. WHAT THIS PAPER ADDS What is already known on this subject Speech and languages abilities change due to non-pathological neurocognitive ageing and neurodegenerative processes. These subtle verbal modifications can be measured through NLP techniques and used as biomarkers for screening/diagnostic purposes in the geriatric population (i.e., digital linguistic biomarkers-DLBs). What this paper adds to existing knowledge The review shows that DLBs can represent a promising clinical tool, with a high potential to spark a major shift to dementia assessment in the elderly. Some challenges and open problems are also discussed. What are the potential or actual clinical implications of this work? This methodological review represents a starting point for clinicians approaching the DLB research field for studying language in healthy and pathological ageing. It summarizes the state of the art and future research directions of this novel approach.
Collapse
Affiliation(s)
- Gloria Gagliardi
- Department of Classical Philology and Italian Studies, University of Bologna, Bologna, Italy
| |
Collapse
|
5
|
Ivanova O, Martínez-Nicolás I, Meilán JJG. Speech changes in old age: Methodological considerations for speech-based discrimination of healthy ageing and Alzheimer's disease. INTERNATIONAL JOURNAL OF LANGUAGE & COMMUNICATION DISORDERS 2024; 59:13-37. [PMID: 37140204 DOI: 10.1111/1460-6984.12888] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 09/28/2022] [Accepted: 04/03/2023] [Indexed: 05/05/2023]
Abstract
BACKGROUND Recent evidence suggests that speech substantially changes in ageing. As a complex neurophysiological process, it can accurately reflect changes in the motor and cognitive systems underpinning human speech. Since healthy ageing is not always easily discriminable from early stages of dementia based on cognitive and behavioural hallmarks, speech is explored as a preclinical biomarker of pathological itineraries in old age. A greater and more specific impairment of neuromuscular activation, as well as a specific cognitive and linguistic impairment in dementia, unchain discriminating changes in speech. Yet, there is no consensus on such discriminatory speech parameters, neither on how they should be elicited and assessed. AIMS To provide a state-of-the-art on speech parameters that allow for early discrimination between healthy and pathological ageing; the aetiology of these parameters; the effect of the type of experimental stimuli on speech elicitation and the predictive power of different speech parameters; and the most promising methods for speech analysis and their clinical implications. METHODS & PROCEDURES A scoping review methodology is used in accordance with the PRISMA model. Following a systematic search of PubMed, PsycINFO and CINAHL, 24 studies are included and analysed in the review. MAIN CONTRIBUTION The results of this review yield three key questions for the clinical assessment of speech in ageing. First, acoustic and temporal parameters are more sensitive to changes in pathological ageing and, of these two, temporal variables are more affected by cognitive impairment. Second, different types of stimuli can trigger speech parameters with different degree of accuracy for the discrimination of clinical groups. Tasks with higher cognitive load are more precise in eliciting higher levels of accuracy. Finally, automatic speech analysis for the discrimination of healthy and pathological ageing should be improved for both research and clinical practice. CONCLUSIONS & IMPLICATIONS Speech analysis is a promising non-invasive tool for the preclinical screening of healthy and pathological ageing. The main current challenges of speech analysis in ageing are the automatization of its clinical assessment and the consideration of the speaker's cognitive background during evaluation. WHAT THIS PAPER ADDS What is already known on the subject Societal aging goes hand in hand with the rising incidence of ageing-related neurodegenerations, mainly Alzheimer's disease (AD). This is particularly noteworthy in countries with longer life expectancies. Healthy ageing and early stages of AD share a set of cognitive and behavioural characteristics. Since there is no cure for dementias, developing methods for accurate discrimination of healthy ageing and early AD is currently a priority. Speech has been described as one of the most significantly impaired features in AD. Neuropathological alterations in motor and cognitive systems would underlie specific speech impairment in dementia. Since speech can be evaluated quickly, non-invasively and inexpensively, its value for the clinical assessment of ageing itineraries may be particularly high. What this paper adds to existing knowledge Theoretical and experimental advances in the assessment of speech as a marker of AD have developed rapidly over the last decade. Yet, they are not always known to clinicians. Furthermore, there is a need to provide an updated state-of-the-art on which speech features are discriminatory to AD, how they can be assessed, what kind of results they can yield, and how such results should be interpreted. This article provides an updated overview of speech profiling, methods of speech measurement and analysis, and the clinical power of speech assessment for early discrimination of AD as the most common cause of dementia. What are the potential or actual clinical implications of this work? This article provides an overview of the predictive potential of different speech parameters in relation to AD cognitive impairment. In addition, it discusses the effect that the cognitive state, the type of elicitation task and the type of assessment method may have on the results of the speech-based analysis in ageing.
Collapse
Affiliation(s)
- Olga Ivanova
- Spanish Language Department, Faculty of Philology, University of Salamanca, Salamanca, Spain
- Institute of Neuroscience of Castilla y León, Salamanca, Spain
| | - Israel Martínez-Nicolás
- Department of Basic Psychology, Psychobiology and Behavioral Science Methodology, Faculty of Psychology, University of Salamanca, Salamanca, Spain
- Institute of Neuroscience of Castilla y León, Salamanca, Spain
| | - Juan José García Meilán
- Department of Basic Psychology, Psychobiology and Behavioral Science Methodology, Faculty of Psychology, University of Salamanca, Salamanca, Spain
- Institute of Neuroscience of Castilla y León, Salamanca, Spain
| |
Collapse
|
6
|
Walker G, Pevy N, O'Malley R, Mirheidari B, Reuber M, Christensen H, Blackburn DJ. Speech patterns in responses to questions asked by an intelligent virtual agent can help to distinguish between people with early stage neurodegenerative disorders and healthy controls. CLINICAL LINGUISTICS & PHONETICS 2023:1-22. [PMID: 37722818 DOI: 10.1080/02699206.2023.2254458] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 08/19/2022] [Accepted: 08/28/2023] [Indexed: 09/20/2023]
Abstract
Previous research has provided strong evidence that speech patterns can help to distinguish between people with early stage neurodegenerative disorders (ND) and healthy controls. This study examined speech patterns in responses to questions asked by an intelligent virtual agent (IVA): a talking head on a computer which asks pre-recorded questions. The study investigated whether measures of response length, speech rate and pausing in responses to questions asked by an IVA help to distinguish between healthy control participants and people diagnosed with Mild Cognitive Impairment (MCI) or Alzheimer's disease (AD). The study also considered whether those measures can further help to distinguish between people with MCI, people with AD, and healthy control participants (HC). There were 38 people with ND (31 people with MCI, 7 people with AD) and 26 HC. All interactions took place in English. People with MCI spoke fewer words compared to HC, and people with AD and people with MCI spoke for less time than HC. People with AD spoke at a slower rate than people with MCI and HC. There were significant differences across all three groups for the proportion of time spent pausing and the average pause duration: silent pauses make up the greatest proportion of responses from people with AD, who also have the longest average silent pause duration, followed by people with MCI then HC. Therefore, the study demonstrates the potential of an IVA as a method for collecting data showing patterns which can help to distinguish between diagnostic groups.
Collapse
Affiliation(s)
- Gareth Walker
- School of English, University of Sheffield, Sheffield, UK
| | - Nathan Pevy
- Department of Computer Science, University of Sheffield, Sheffield, UK
| | - Ronan O'Malley
- Department of Neuroscience, University of Sheffield, Sheffield, UK
| | - Bahman Mirheidari
- Department of Computer Science, University of Sheffield, Sheffield, UK
| | - Markus Reuber
- Academic Neurology Unit, Royal Hallamshire Hospital, University of Sheffield, Sheffield, UK
| | - Heidi Christensen
- Department of Computer Science, University of Sheffield, Sheffield, UK
| | | |
Collapse
|
7
|
Martínez-Nicolás I, Martínez-Sánchez F, Ivanova O, Meilán JJG. Reading and lexical-semantic retrieval tasks outperforms single task speech analysis in the screening of mild cognitive impairment and Alzheimer's disease. Sci Rep 2023; 13:9728. [PMID: 37322073 PMCID: PMC10272227 DOI: 10.1038/s41598-023-36804-y] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/27/2023] [Accepted: 06/12/2023] [Indexed: 06/17/2023] Open
Abstract
Age-related cognitive impairment have increased dramatically in recent years, which has risen the interes in developing screening tools for mild cognitive impairment and Alzheimer's disease. Speech analysis allows to exploit the behavioral consequences of cognitive deficits on the patient's vocal performance so that it is possible to identify pathologies affecting speech production such as dementia. Previous studies have further shown that the speech task used determines how the speech parameters are altered. We aim to combine the impairments in several speech production tasks in order to improve the accuracy of screening through speech analysis. The sample consists of 72 participants divided into three equal groups of healthy older adults, people with mild cognitive impairment, or Alzheimer's disease, matched by age and education. A complete neuropsychological assessment and two voice recordings were performed. The tasks required the participants to read a text, and complete a sentence with semantic information. A stepwise linear discriminant analysis was performed to select speech parameters with discriminative power. The discriminative functions obtained an accuracy of 83.3% in simultaneous classifications of several levels of cognitive impairment. It would therefore be a promising screening tool for dementia.
Collapse
Affiliation(s)
| | | | - Olga Ivanova
- Faculty of Philology, University of Salamanca, 37008, Salamanca, Spain
| | - Juan J G Meilán
- Faculty of Psychology, University of Salamanca, 37008, Salamanca, Spain
- Institute of Neuroscience of Castilla y León, 37007, Salamanca, Spain
| |
Collapse
|
8
|
Knox L, McDermott C, Hobson E. Telehealth in long-term neurological conditions: the potential, the challenges and the key recommendations. J Med Eng Technol 2022; 46:506-517. [PMID: 35212580 DOI: 10.1080/03091902.2022.2040625] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/11/2023]
Abstract
Long-term neurological conditions (LTNCs) cause physical and psychological symptoms that have a significant impact on activities of daily living and quality of life. Multidisciplinary teams are effective at providing treatment for people with LTNCs; however, access to such services by people with disabilities can be difficult and as a result, good quality care is not universal. One potential solution is telehealth. This review describes the potential of telehealth to support people with LTNCs, the challenges of designing and implementing these systems, and the key recommendations for those involved in telehealth to facilitate connected services that can benefit patients, carers and healthcare professionals. These recommendations include understanding the problems posed by LTNCs and the needs of the end-user through a person-centred approach. We discuss how to work collaboratively and use shared learning, and consider how to effectively evaluate the intervention at every stage of the development process.
Collapse
Affiliation(s)
- Liam Knox
- Department of Neuroscience, University of Sheffield, Sheffield, UK
| | - Christopher McDermott
- Department of Neuroscience, University of Sheffield, Sheffield, UK.,Department of Neuroscience, Sheffield Teaching Hospitals, Sheffield, UK
| | - Esther Hobson
- Department of Neuroscience, University of Sheffield, Sheffield, UK.,Department of Neuroscience, Sheffield Teaching Hospitals, Sheffield, UK
| |
Collapse
|
9
|
Li R, Wang X, Lawler K, Garg S, Bai Q, Alty J. Applications of Artificial Intelligence to aid detection of dementia: a scoping review on current capabilities and future directions. J Biomed Inform 2022; 127:104030. [DOI: 10.1016/j.jbi.2022.104030] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/21/2021] [Revised: 01/21/2022] [Accepted: 02/12/2022] [Indexed: 12/17/2022]
|
10
|
Pevy N, Christensen H, Walker T, Reuber M. Feasibility of using an automated analysis of formulation effort in patients' spoken seizure descriptions in the differential diagnosis of epileptic and nonepileptic seizures. Seizure 2021; 91:141-145. [PMID: 34157636 DOI: 10.1016/j.seizure.2021.06.009] [Citation(s) in RCA: 9] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/17/2021] [Revised: 05/17/2021] [Accepted: 06/08/2021] [Indexed: 11/28/2022] Open
Abstract
OBJECTIVE There are three common causes of Transient Loss of Consciousness (TLOC), syncope, epileptic and psychogenic nonepileptic seizures (PNES). Many individuals who have experienced TLOC initially receive an incorrect diagnosis and inappropriate treatment. Whereas syncope can be distinguished relatively easily with a small number of "yes"/"no" questions, the differentiation of the other two causes of TLOC is more challenging. Previous qualitative research based on the methodology of Conversation Analysis has demonstrated that the descriptions of epileptic seizures contain more formulation effort than accounts of PNES. This research investigates whether features likely to reflect the level of formulation effort can be automatically elicited from audio recordings and transcripts of speech and used to differentiate between epileptic and nonepileptic seizures. METHOD Verbatim transcripts of conversations between patients and neurologists were manually produced from video and audio recordings of 45 interactions (21 epilepsy and 24 PNES). The subsection of each transcript containing the person's account of their first seizure was manually extracted for the analysis. Seven automatically detectable features were designed as markers of formulation effort. These features were used to train a Random Forest machine learning classifier. RESULT There were significantly more hesitations and repetitions in descriptions of epileptic than nonepileptic seizures. Using a nested leave-one-out cross validation approach, 71% of seizures were correctly classified by the Random Forest classifier. DISCUSSION This pilot study provides proof of principle that linguistic features that have been automatically extracted from audio recordings and transcripts could be used to distinguish between epileptic seizures and PNES and thereby contribute to the differential diagnosis of TLOC. Future research should explore whether additional observations can be incorporated into a diagnostic stratification tool and compare the performance of these features when they are combined with additional information provided by patients and witnesses about seizure manifestations and medical history.
Collapse
Affiliation(s)
- Nathan Pevy
- Sheffield Institute for Translational Neuroscience (SITraN), University of Sheffield, Sheffield, United Kingdom.
| | - Heidi Christensen
- Department of Computer Science, University of Sheffield, Sheffield, United Kingdom
| | - Traci Walker
- Division of Human Communication Sciences, University of Sheffield, Sheffield, United Kingdom
| | - Markus Reuber
- Academic Neurology Unit, University of Sheffield, Royal Hallamshire Hospital, Sheffield, United Kingdom
| |
Collapse
|
11
|
De Looze C, Dehsarvi A, Crosby L, Vourdanou A, Coen RF, Lawlor BA, Reilly RB. Cognitive and Structural Correlates of Conversational Speech Timing in Mild Cognitive Impairment and Mild-to-Moderate Alzheimer's Disease: Relevance for Early Detection Approaches. Front Aging Neurosci 2021; 13:637404. [PMID: 33986656 PMCID: PMC8110716 DOI: 10.3389/fnagi.2021.637404] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/03/2020] [Accepted: 03/31/2021] [Indexed: 11/19/2022] Open
Abstract
Background: Increasing efforts have focused on the establishment of novel biomarkers for the early detection of Alzheimer’s disease (AD) and prediction of Mild Cognitive Impairment (MCI)-to-AD conversion. Behavioral changes over the course of healthy ageing, at disease onset and during disease progression, have been recently put forward as promising markers for the detection of MCI and AD. The present study examines whether the temporal characteristics of speech in a collaborative referencing task are associated with cognitive function and the volumes of brain regions involved in speech production and known to be reduced in MCI and AD pathology. We then explore the discriminative ability of the temporal speech measures for the classification of MCI and AD. Method: Individuals with MCI, mild-to-moderate AD and healthy controls (HCs) underwent a structural MRI scan and a battery of neuropsychological tests. They also engaged in a collaborative referencing task with a caregiver. The associations between the conversational speech timing features, cognitive function (domain-specific) and regional brain volumes were examined by means of linear mixed-effect modeling. Genetic programming was used to explore the discriminative ability of the conversational speech features. Results: MCI and mild-to-moderate AD are characterized by a general slowness of speech, attributed to slower speech rate and slower turn-taking in conversational settings. The speech characteristics appear to be reflective of episodic, lexico-semantic, executive functioning and visuospatial deficits and underlying volume reductions in frontal, temporal and cerebellar areas. Conclusion: The implementation of conversational speech timing-based technologies in clinical and community settings may provide additional markers for the early detection of cognitive deficits and structural changes associated with MCI and AD.
Collapse
Affiliation(s)
- Céline De Looze
- Trinity Centre for Biomedical Engineering, School of Engineering, Trinity College Dublin, Dublin, Ireland
| | - Amir Dehsarvi
- Trinity Centre for Biomedical Engineering, School of Engineering, Trinity College Dublin, Dublin, Ireland
| | - Lisa Crosby
- Mercer's Institute for Successful Ageing, St James's Hospital, Dublin, Ireland
| | - Aisling Vourdanou
- Trinity Centre for Biomedical Engineering, School of Engineering, Trinity College Dublin, Dublin, Ireland
| | - Robert F Coen
- Mercer's Institute for Successful Ageing, St James's Hospital, Dublin, Ireland
| | - Brian A Lawlor
- Mercer's Institute for Successful Ageing, St James's Hospital, Dublin, Ireland.,Institute of Neuroscience, School of Medicine, Trinity College Dublin, Dublin, Ireland
| | - Richard B Reilly
- Trinity Centre for Biomedical Engineering, School of Engineering, Trinity College Dublin, Dublin, Ireland.,Institute of Neuroscience, School of Medicine, Trinity College Dublin, Dublin, Ireland
| |
Collapse
|