CUSCO: An Unobtrusive Custom Secure Audio-Visual Recording System for Ambient Assisted Living.

The ubiquity of digital technology has facilitated detailed recording of human behaviour. Ambient technology has been used to capture behaviours in a broad range of applications ranging from healthcare and monitoring to assessment of cooperative work. However, existing systems often face challenges in terms of autonomy, usability, and privacy. This paper presents a portable, easy-to-use and privacy-preserving system for capturing behavioural signals unobtrusively in home or in office settings. The system focuses on the capture of audio, video, and depth imaging. It is based on a device built on a small-factor platform that incorporates ambient sensors which can be integrated with the audio and depth video hardware for multimodal behaviour tracking. The system can be accessed remotely and integrated into a network of sensors. Data are encrypted in real time to ensure safety and privacy. We illustrate uses of the device in two different settings, namely, a healthy-ageing IoT application, where the device is used in conjunction with a range of IoT sensors to monitor an older person's mental well-being at home, and a healthcare communication quality assessment application, where the device is used to capture a patient-clinician interaction for consultation quality appraisal. CUSCO can automatically detect active speakers, extract acoustic features, record video and depth streams, and recognise emotions and cognitive impairment with promising accuracy.

Predicting sex, age, general cognition and mental health with machine learning on brain structural connectomes.

There is an increasing expectation that advanced, computationally expensive machine learning (ML) techniques, when applied to large population-wide neuroimaging datasets, will help to uncover key differences in the human brain in health and disease. We take a comprehensive approach to explore how multiple aspects of brain structural connectivity can predict sex, age, general cognitive function and general psychopathology, testing different ML algorithms from deep learning (DL) model (BrainNetCNN) to classical ML methods. We modelled N = 8183 structural connectomes from UK Biobank using six different structural network weightings obtained from diffusion MRI. Streamline count generally provided the highest prediction accuracies in all prediction tasks. DL did not improve on prediction accuracies from simpler linear models. Further, high correlations between gradient attribution coefficients from DL and model coefficients from linear models suggested the models ranked the importance of features in similar ways, which indirectly suggested the similarity in models' strategies for making predictive decision to some extent. This highlights that model complexity is unlikely to improve detection of associations between structural connectomes and complex phenotypes with the current sample size.

Epistemologies of evidence-based medicine: a plea for corpus-based conceptual research in the medical humanities.

Evidence-based medicine has been the subject of much controversy within and outside the field of medicine, with its detractors characterizing it as reductionist and authoritarian, and its proponents rejecting such characterization as a caricature of the actual practice. At the heart of this controversy is a complex linguistic and social process that cannot be illuminated by appealing to the semantics of the modifier evidence-based. The complexity lies in the nature of evidence as a basic concept that circulates in both expert and non-expert spheres of communication, supports different interpretations in different contexts, and is inherently open to contestation. We outline a new methodology that combines a social epistemological perspective with advanced methods of corpus linguistics and elements of conceptual history to investigate this and other basic concepts that underpin the practice and ethos of modern medicine. The potential of this methodology to offer new insights into controversies such as those surrounding EBM is demonstrated through a case study of the various meanings supported by evidence and based, as attested in a large electronic corpus of online material written by non-experts as well as a variety of experts in different fields, including medicine.

The successful uptake and sustainability of rapid infectious disease and antimicrobial resistance point-of-care testing requires a complex 'mix-and-match' implementation package.

The emergence and spread of antimicrobial resistance is one of the major global issues currently threatening the health and wealth of nations, with effective guidelines and intervention strategies urgently required. Such guidelines and interventions should ideally be targeted at individuals, communities, and nations, requiring international coordination for maximum effect. In this respect, the European Joint Programming Initiative on Antimicrobial Resistance Transnational Working Group 'Antimicrobial Resistance - Rapid Diagnostic Tests' (JPIAMR AMR-RDT) is proposing to consider a 'mix-and-match' package for the implementation of point-of-care testing (PoCT), which is described in this publication. The working group was established with the remit of identifying barriers and solutions to the development and implementation of rapid infectious disease PoCT for combatting the global spread of antimicrobial resistance. It constitutes a multi-sectoral collaboration between medical, technological, and industrial opinion leaders involved in in vitro diagnostics development, medical microbiology, and clinical infectious diseases. The mix-and-match implementation package is designed to encourage the implementation of rapid infectious disease and antimicrobial resistance PoCT in transnational medical environments for use in the fight against increasing antimicrobial resistance.

What matters to people with memory problems, healthy volunteers and health and social care professionals in the context of developing treatment to prevent Alzheimer's dementia? A qualitative study.

BACKGROUND: Alzheimer's disease (AD) is recognized as one of the greatest global public health challenges. There is increasing consensus that optimal disease modification using pharmaceuticals may best be achieved earlier in the disease continuum before symptoms occur. However, more needs to be understood about what outcomes are meaningful to potential participants in clinical trials within this preventative paradigm and how people make trade-offs between risks and benefits. The Electronic Person-Specific Outcome Measure (ePSOM) programme is developing an app to capture person-specific outcomes and preferences in clinical trials. OBJECTIVE: As one phase in the ePSOM programme, this study explored what matters when developing new treatments to prevent AD and how trade-offs are made between risks and benefits, from three perspectives. DESIGN: Focus groups were conducted with people living with memory problems (n = 21) and healthy volunteers (n = 10), and telephone interviews with health and social care professionals (n = 10). Differences and overlap between the three groups were explored. RESULTS: Outcomes that matter lie in five key domains in relation to what matters in everyday life: Everyday Functioning; Relationships and Social Connections; Enjoying Life; Sense of Identity; and Alleviating  Symptoms. Insights were gained into the significance of reducing the risk of developing dementia with drugs and the processes of weighing up risks versus benefits. DISCUSSION AND CONCLUSIONS: The key domains identified are being used to inform the next stage of the ePSOM programme which is to develop a survey to be distributed nationally in the UK to explore these issues further.

An Overview of the ADReSS-M Signal Processing Grand Challenge on Multilingual Alzheimer's Dementia Recognition Through Spontaneous Speech.

The ADReSS-M Signal Processing Grand Challenge was held at the 2023 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023. The challenge targeted difficult automatic prediction problems of great societal and medical relevance, namely, the detection of Alzheimer's Dementia (AD) and the estimation of cognitive test scoress. Participants were invited to create models for the assessment of cognitive function based on spontaneous speech data. Most of these models employed signal processing and machine learning methods. The ADReSS-M challenge was designed to assess the extent to which predictive models built based on speech in one language generalise to another language. The language data compiled and made available for ADReSS-M comprised English, for model training, and Greek, for model testing and validation. To the best of our knowledge no previous shared research task investigated acoustic features of the speech signal or linguistic characteristics in the context of multilingual AD detection. This paper describes the context of the ADReSS-M challenge, its data sets, its predictive tasks, the evaluation methodology we employed, our baseline models and results, and the top five submissions. The paper concludes with a summary discussion of the ADReSS-M results, and our critical assessment of the future outlook in this field.

Longitudinal observational cohort study: Speech for Intelligent cognition change tracking and DEtection of Alzheimer's Disease (SIDE-AD).

INTRODUCTION: There is emerging evidence that speech may be a potential indicator and manifestation of early Alzheimer's disease (AD) pathology. Therefore, the University of Edinburgh and Sony Research have partnered to create the Speech for Intelligent cognition change tracking and DEtection of Alzheimer's Disease (SIDE-AD) study, which aims to develop digital speech-based biomarkers for use in neurodegenerative disease. METHODS AND ANALYSIS: SIDE-AD is an observational longitudinal study, collecting samples of spontaneous speech. Participants are recruited from existing cohort studies as well as from the National Health Service (NHS)memory clinics in Scotland. Using an online platform, participants record a voice sample talking about their brain health and rate their mood, anxiety and apathy. The speech biomarkers will be analysed longitudinally, and we will use machine learning and natural language processing technology to automate the assessment of the respondents' speech patterns. ETHICS AND DISSEMINATION: The SIDE-AD study has been approved by the NHS Research Ethics Committee (REC reference: 23/WM/0153, protocol number AC23046, IRAS Project ID 323311) and received NHS management approvals from Lothian, Fife and Forth Valley NHS boards. Our main ethical considerations pertain to the remote administration of the study, such as taking remote consent. To address this, we implemented a consent process, whereby the first step of the consent is done entirely remotely but a member of the research team contacts the participant over the phone to consent participants to the optional, most sensitive, elements of the study. Results will be presented at conferences, published in peer-reviewed journals and communicated to study participants.

Machine learning on cardiotocography data to classify fetal outcomes: A scoping review.

INTRODUCTION: Uterine contractions during labour constrict maternal blood flow and oxygen delivery to the developing baby, causing transient hypoxia. While most babies are physiologically adapted to withstand such intrapartum hypoxia, those exposed to severe hypoxia or with poor physiological reserves may experience neurological injury or death during labour. Cardiotocography (CTG) monitoring was developed to identify babies at risk of hypoxia by detecting changes in fetal heart rate (FHR) patterns. CTG monitoring is in widespread use in intrapartum care for the detection of fetal hypoxia, but the clinical utility is limited by a relatively poor positive predictive value (PPV) of an abnormal CTG and significant inter and intra observer variability in CTG interpretation. Clinical risk and human factors may impact the quality of CTG interpretation. Misclassification of CTG traces may lead to both under-treatment (with the risk of fetal injury or death) or over-treatment (which may include unnecessary operative interventions that put both mother and baby at risk of complications). Machine learning (ML) has been applied to this problem since early 2000 and has shown potential to predict fetal hypoxia more accurately than visual interpretation of CTG alone. To consider how these tools might be translated for clinical practice, we conducted a review of ML techniques already applied to CTG classification and identified research gaps requiring investigation in order to progress towards clinical implementation. MATERIALS AND METHOD: We used identified keywords to search databases for relevant publications on PubMed, EMBASE and IEEE Xplore. We used Preferred Reporting Items for Systematic Review and Meta-Analysis for Scoping Reviews (PRISMA-ScR). Title, abstract and full text were screened according to the inclusion criteria. RESULTS: We included 36 studies that used signal processing and ML techniques to classify CTG. Most studies used an open-access CTG database and predominantly used fetal metabolic acidosis as the benchmark for hypoxia with varying pH levels. Various methods were used to process and extract CTG signals and several ML algorithms were used to classify CTG. We identified significant concerns over the practicality of using varying pH levels as the CTG classification benchmark. Furthermore, studies needed to be more generalised as most used the same database with a low number of subjects for an ML study. CONCLUSION: ML studies demonstrate potential in predicting fetal hypoxia from CTG. However, more diverse datasets, standardisation of hypoxia benchmarks and enhancement of algorithms and features are needed for future clinical implementation.

Towards a lifelong personalized brain health program: empowering individuals to define, pursue, and monitor meaningful outcomes.

Incorporating person-centered outcomes into clinical trials for neurodegenerative diseases has been challenging due to a deficiency in quantitative measures. Meanwhile, the integration of personally meaningful treatment targets in clinical practice remains qualitative, failing to truly inform evaluations, therapeutic interventions and longitudinal monitoring and support. We discuss the current advances and future directions in capturing individualized brain health outcomes and present an approach to integrate person-centered outcome in a scalable manner. Our approach stems from the evidence-based electronic Person-Specific Outcome Measure (ePSOM) program which prompts an individual to define personally meaningful treatment priorities and report level of confidence in managing items that matter to the individual the most (e.g., "Do I feel confident in my ability to contribute to a conversation?"). Deployed either as a single version (person only) or a dyad version (person and care partner), our proposed tool could be used as an endpoint in clinical trials, offering proof of meaningful intervention benefits and in clinical practice, by establishing an anchor for the therapeutic objectives sought by the individual.

Enhancing volleyball training: empowering athletes and coaches through advanced sensing and analysis.

Modern sensing technologies and data analysis methods usher in a new era for sports training and practice. Hidden insights can be uncovered and interactive training environments can be created by means of data analysis. We present a system to support volleyball training which makes use of Inertial Measurement Units, a pressure sensitive display floor, and machine learning techniques to automatically detect relevant behaviours and provides the user with the appropriate information. While working with trainers and amateur athletes, we also explore potential applications that are driven by automatic action recognition, that contribute various requirements to the platform. The first application is an automatic video-tagging protocol that marks key events (captured on video) based on the automatic recognition of volleyball-specific actions with an unweighted average recall of 78.71% in the 10-fold cross-validation setting with convolution neural network and 73.84% in leave-one-subject-out cross-validation setting with active data representation method using wearable sensors, as an exemplification of how dashboard and retrieval systems would work with the platform. In the context of action recognition, we have evaluated statistical functions and their transformation using active data representation besides raw signal of IMUs sensor. The second application is the "bump-set-spike" trainer, which uses automatic action recognition to provide real-time feedback about performance to steer player behaviour in volleyball, as an example of rich learning environments enabled by live action detection. In addition to describing these applications, we detail the system components and architecture and discuss the implications that our system might have for sports in general and for volleyball in particular.

Machine Learning models for detection and assessment of progression in Alzheimer's disease based on blood and cerebrospinal fluid biomarkers.

Machine-learning techniques were applied to human blood plasma and cerebrospinal fluid (CSF) biomarker data related to cognitive decline in Alzheimer's Disease (AD) patients available via Alzheimer Disease Neuroimaging Initiative (ADNI) study. We observed the accuracy of AD diagnosis is greatest when protein biomarkers from cerebrospinal fluid are combined with plasma proteins using Support Vector Machines (SVM); this is not improved by adding age and sex. The area under the receiver operator characteristic (ROC) curve for our model of AD diagnosis based on a full (unbiased) set of plasma proteins was 0.94 in cross-validation and 0.82 on an external validation (test) set. Taking plasma in combination with CSF, the model reaches 0.98 area under the ROC curve on the test set. Accuracy of prediction of risk of mild cognitive impairment progressing to AD is the same for blood plasma biomarkers as for CSF and is not improved by combining them or adding age and sex as covariates.Clinical relevance- The identification of accurate and cost-effective biomarkers to screen for risk of developing AD and monitoring its progression is crucial for improved understanding of its causes and stratification of patients for treatments under development. This paper demonstrates the feasibility of AD detection and prognosis based on blood plasma biomarkers.

Machine Learning to Classify Cardiotocography for Fetal Hypoxia Detection.

Fetal hypoxia can cause damaging consequences on babies' such as stillbirth and cerebral palsy. Cardiotocography (CTG) has been used to detect intrapartum fetal hypoxia during labor. It is a non-invasive machine that measures the fetal heart rate and uterine contractions. Visual CTG suffers inconsistencies in interpretations among clinicians that can delay interventions. Machine learning (ML) showed potential in classifying abnormal CTG, allowing automatic interpretation. In the absence of a gold standard, researchers used various surrogate biomarkers to classify CTG, where some were clinically irrelevant. We proposed using Apgar scores as the surrogate benchmark of babies' ability to recover from birth. Apgar scores measure newborns' ability to recover from active uterine contraction, which measures appearance, pulse, grimace, activity and respiration. The higher the Apgar score, the healthier the baby is.We employ signal processing methods to pre-process and extract validated features of 552 raw CTG. We also included CTG-specific characteristics as outlined in the NICE guidelines. We employed ML techniques using 22 features and measured performances between ML classifiers. While we found that ML can distinguish CTG with low Apgar scores, results for the lowest Apgar scores, which are rare in the dataset we used, would benefit from more CTG data for better performance. We need an external dataset to validate our model for generalizability to ensure that it does not overfit a specific population.Clinical Relevance- This study demonstrated the potential of using a clinically relevant benchmark for classifying CTG to allow automatic early detection of hypoxia to reduce decision-making time in maternity units.

Expert guidance on target product profile development for AMR diagnostic tests.

Diagnostics are widely considered crucial in the fight against antimicrobial resistance (AMR), which is expected to kill 10 million people annually by 2030. Nevertheless, there remains a substantial gap between the need for AMR diagnostics versus their development and implementation. To help address this problem, target product profiles (TPP) have been developed to focus developers' attention on the key aspects of AMR diagnostic tests. However, during discussion between a multisectoral working group of 51 international experts from industry, academia and healthcare, it was noted that specific AMR-related TPPs could be extended by incorporating the interdependencies between the key characteristics associated with the development of such TPPs. Subsequently, the working group identified 46 characteristics associated with six main categories (ie, Intended Use, Diagnostic Question, Test Description, Assay Protocol, Performance and Commercial). The interdependencies of these characteristics were then identified and mapped against each other to generate new insights for use by stakeholders. Specifically, it may not be possible for diagnostics developers to achieve all of the recommendations in every category of a TPP and this publication indicates how prioritising specific TPP characteristics during diagnostics development may influence (or not) a range of other TPP characteristics associated with the diagnostic. The use of such guidance, in conjunction with specific TPPs, could lead to more efficient AMR diagnostics development.

Remote data collection speech analysis and prediction of the identification of Alzheimer's disease biomarkers in people at risk for Alzheimer's disease dementia: the Speech on the Phone Assessment (SPeAk) prospective observational study protocol.

INTRODUCTION: Identifying cost-effective, non-invasive biomarkers of Alzheimer's disease (AD) is a clinical and research priority. Speech data are easy to collect, and studies suggest it can identify those with AD. We do not know if speech features can predict AD biomarkers in a preclinical population. METHODS AND ANALYSIS: The Speech on the Phone Assessment (SPeAk) study is a prospective observational study. SPeAk recruits participants aged 50 years and over who have previously completed studies with AD biomarker collection. Participants complete a baseline telephone assessment, including spontaneous speech and cognitive tests. A 3-month visit will repeat the cognitive tests with a conversational artificial intelligence bot. Participants complete acceptability questionnaires after each visit. Participants are randomised to receive their cognitive test results either after each visit or only after they have completed the study. We will combine SPeAK data with AD biomarker data collected in a previous study and analyse for correlations between extracted speech features and AD biomarkers. The outcome of this analysis will inform the development of an algorithm for prediction of AD risk based on speech features. ETHICS AND DISSEMINATION: This study has been approved by the Edinburgh Medical School Research Ethics Committee (REC reference 20-EMREC-007). All participants will provide informed consent before completing any study-related procedures, participants must have capacity to consent to participate in this study. Participants may find the tests, or receiving their scores, causes anxiety or stress. Previous exposure to similar tests may make this more familiar and reduce this anxiety. The study information will include signposting in case of distress. Study results will be disseminated to study participants, presented at conferences and published in a peer reviewed journal. No study participants will be identifiable in the study results.

Temporal Integration of Text Transcripts and Acoustic Features for Alzheimer's Diagnosis Based on Spontaneous Speech.

Background: Advances in machine learning (ML) technology have opened new avenues for detection and monitoring of cognitive decline. In this study, a multimodal approach to Alzheimer's dementia detection based on the patient's spontaneous speech is presented. This approach was tested on a standard, publicly available Alzheimer's speech dataset for comparability. The data comprise voice samples from 156 participants (1:1 ratio of Alzheimer's to control), matched by age and gender. Materials and Methods: A recently developed Active Data Representation (ADR) technique for voice processing was employed as a framework for fusion of acoustic and textual features at sentence and word level. Temporal aspects of textual features were investigated in conjunction with acoustic features in order to shed light on the temporal interplay between paralinguistic (acoustic) and linguistic (textual) aspects of Alzheimer's speech. Combinations between several configurations of ADR features and more traditional bag-of-n-grams approaches were used in an ensemble of classifiers built and evaluated on a standardised dataset containing recorded speech of scene descriptions and textual transcripts. Results: Employing only semantic bag-of-n-grams features, an accuracy of 89.58% was achieved in distinguishing between Alzheimer's patients and healthy controls. Adding temporal and structural information by combining bag-of-n-grams features with ADR audio/textual features, the accuracy could be improved to 91.67% on the test set. An accuracy of 93.75% was achieved through late fusion of the three best feature configurations, which corresponds to a 4.7% improvement over the best result reported in the literature for this dataset. Conclusion: The proposed combination of ADR audio and textual features is capable of successfully modelling temporal aspects of the data. The machine learning approach toward dementia detection achieves best performance when ADR features are combined with strong semantic bag-of-n-grams features. This combination leads to state-of-the-art performance on the AD classification task.

COVID-19: Affect recognition through voice analysis during the winter lockdown in Scotland.

The COVID-19 pandemic has led to unprecedented restrictions in people's lifestyle which have affected their psychological wellbeing. In this context, this paper investigates the use of social signal processing techniques for remote assessment of emotions. It presents a machine learning method for affect recognition applied to recordings taken during the COVID-19 winter lockdown in Scotland (UK). This method is exclusively based on acoustic features extracted from voice recordings collected through home and mobile devices (i.e. phones, tablets), thus providing insight into the feasibility of monitoring people's psychological wellbeing remotely, automatically and at scale. The proposed model is able to predict affect with a concordance correlation coefficient of 0.4230 (using Random Forest) and 0.3354 (using Decision Trees) for arousal and valence respectively.Clinical relevance- In 2018/2019, 12% and 14% of Scottish adults reported depression and anxiety symptoms. Remote emotion recognition through home devices would support the detection of these difficulties, which are often underdiagnosed and, if untreated, may lead to temporal or chronic disability.

A Network-Based Embedding Method for Drug-Target Interaction Prediction.

Integration of multi-omics and pharmacological data can help researchers understand the impact of drugs on dynamic biological systems. Network-based approaches to such integration explore the interaction of different cellular components and drugs. However, with ever-increasing amounts of data, processing these high-dimensional biological networks requires powerful tools. We investigate whether network embeddings can address this problem by providing an effective method for dimensionality reduction in drug-related networks. A neural network-based embedding method is employed to encode protein-protein, protein-disease, drug-drug and drug-disease networks for the prediction of novel drug-target interactions. We found that drug-target interaction prediction using embeddings of heterogeneous networks as input features performs comparably to state-of-the-art methods, exhibiting an area under the ROC curve of 84%, outperforming methods such as BLM-NII and NetLapRLS, and coming very close to the best performing network methods such as HNM, CMF and DTINet. These encouraging results suggest that further investigation of this approach is warranted.

Predicting Mini-Mental Status Examination Scores through Paralinguistic Acoustic Features of Spontaneous Speech.

Speech analysis could provide an indicator of cognitive health and help develop clinical tools for automatically detecting and monitoring cognitive health progression. The Mini Mental Status Examination (MMSE) is the most widely used screening tool for cognitive health. But the manual operation of MMSE restricts its screening within primary care facilities. An automatic screening tool has the potential to remedy this situation. This study aims to assess the association between acoustic features of spontaneous speech and assess whether acoustic features can be used to automatically predict MMSE score. We assessed the effectiveness of paralinguistic feature set for MMSE score prediction on a balanced sample of DementiaBank's Pitt spontaneous speech dataset, with patients matched by gender and age. Linear regression analysis shows that fusion of acoustic features, age, sex and years of education provides better results (mean absolute error, MAE = 4.97, and R2 = 0.261) than acoustic features alone (MAE = 5.66 and R2 =0.125) and age, gender and education level alone (MAE of 5.36 and R2 =0.17). This suggests that the acoustic features of spontaneous speech are an important part of an automatic screening tool for cognitive impairment detection.Clinical relevance- We hereby present a method for automatic screening of cognitive health. It is based on acoustic information of speech, a ubiquitous source of data, therefore being cost-efficient, non-invasive and with little infrastructure required.

Cross-corpus Feature Learning between Spontaneous Monologue and Dialogue for Automatic Classification of Alzheimer's Dementia Speech.

Speech analysis could help develop clinical tools for automatic detection of Alzheimer's disease and monitoring of its progression. However, datasets containing both clinical information and spontaneous speech suitable for statistical learning are relatively scarce. In addition, speech data are often collected under different conditions, such as monologue and dialogue recording protocols. Therefore, there is a need for methods to allow the combination of these scarce resources. In this paper, we propose two feature extraction and representation models, based on neural networks and trained on monologue and dialogue data recorded in clinical settings. These models are evaluated not only for AD recognition, but also with respect to their potential to generalise across both datasets. They provide good results when trained and tested on the same data set (72.56% UAR for monologue data and 85.21% for dialogue). A decrease in UAR is observed in transfer training, where feature extraction models trained on dialogues provide better average UAR on monologues (63.72%) than the other way around (58.94%). When the choice of classifiers is independent of feature extraction, transfer from monologue models to dialogues result in a maximum UAR of 81.04% and transfer from dialogue features to monologue achieve a maximum UAR of 70.73%, evidencing the generalisability of the feature model.

Artificial Intelligence, Speech, and Language Processing Approaches to Monitoring Alzheimer's Disease: A Systematic Review.

BACKGROUND: Language is a valuable source of clinical information in Alzheimer's disease, as it declines concurrently with neurodegeneration. Consequently, speech and language data have been extensively studied in connection with its diagnosis. OBJECTIVE: Firstly, to summarize the existing findings on the use of artificial intelligence, speech, and language processing to predict cognitive decline in the context of Alzheimer's disease. Secondly, to detail current research procedures, highlight their limitations, and suggest strategies to address them. METHODS: Systematic review of original research between 2000 and 2019, registered in PROSPERO (reference CRD42018116606). An interdisciplinary search covered six databases on engineering (ACM and IEEE), psychology (PsycINFO), medicine (PubMed and Embase), and Web of Science. Bibliographies of relevant papers were screened until December 2019. RESULTS: From 3,654 search results, 51 articles were selected against the eligibility criteria. Four tables summarize their findings: study details (aim, population, interventions, comparisons, methods, and outcomes), data details (size, type, modalities, annotation, balance, availability, and language of study), methodology (pre-processing, feature generation, machine learning, evaluation, and results), and clinical applicability (research implications, clinical potential, risk of bias, and strengths/limitations). CONCLUSION: Promising results are reported across nearly all 51 studies, but very few have been implemented in clinical research or practice. The main limitations of the field are poor standardization, limited comparability of results, and a degree of disconnect between study aims and clinical applications. Active attempts to close these gaps will support translation of future research into clinical practice.