The Effectiveness of Artificial Intelligence Conversational Agents in Health Care: Systematic Review.

BACKGROUND: The high demand for health care services and the growing capability of artificial intelligence have led to the development of conversational agents designed to support a variety of health-related activities, including behavior change, treatment support, health monitoring, training, triage, and screening support. Automation of these tasks could free clinicians to focus on more complex work and increase the accessibility to health care services for the public. An overarching assessment of the acceptability, usability, and effectiveness of these agents in health care is needed to collate the evidence so that future development can target areas for improvement and potential for sustainable adoption. OBJECTIVE: This systematic review aims to assess the effectiveness and usability of conversational agents in health care and identify the elements that users like and dislike to inform future research and development of these agents. METHODS: PubMed, Medline (Ovid), EMBASE (Excerpta Medica dataBASE), CINAHL (Cumulative Index to Nursing and Allied Health Literature), Web of Science, and the Association for Computing Machinery Digital Library were systematically searched for articles published since 2008 that evaluated unconstrained natural language processing conversational agents used in health care. EndNote (version X9, Clarivate Analytics) reference management software was used for initial screening, and full-text screening was conducted by 1 reviewer. Data were extracted, and the risk of bias was assessed by one reviewer and validated by another. RESULTS: A total of 31 studies were selected and included a variety of conversational agents, including 14 chatbots (2 of which were voice chatbots), 6 embodied conversational agents (3 of which were interactive voice response calls, virtual patients, and speech recognition screening systems), 1 contextual question-answering agent, and 1 voice recognition triage system. Overall, the evidence reported was mostly positive or mixed. Usability and satisfaction performed well (27/30 and 26/31), and positive or mixed effectiveness was found in three-quarters of the studies (23/30). However, there were several limitations of the agents highlighted in specific qualitative feedback. CONCLUSIONS: The studies generally reported positive or mixed evidence for the effectiveness, usability, and satisfactoriness of the conversational agents investigated, but qualitative user perceptions were more mixed. The quality of many of the studies was limited, and improved study design and reporting are necessary to more accurately evaluate the usefulness of the agents in health care and identify key areas for improvement. Further research should also analyze the cost-effectiveness, privacy, and security of the agents. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR2-10.2196/16934.

A comparison of 2-hydroxyglutarate detection at 3 and 7 T with long-TE semi-LASER.

Abnormally high levels of the 'oncometabolite' 2-hydroxyglutarate (2-HG) occur in many grade II and III gliomas, and correlate with mutations in the genes of isocitrate dehydrogenase (IDH) isoforms. In vivo measurement of 2-HG in patients, using magnetic resonance spectroscopy (MRS), has largely been carried out at 3 T, yet signal overlap continues to pose a challenge for 2-HG detection. To combat this, several groups have proposed MRS methods at ultra-high field (≥7 T) where theoretical increases in signal-to-noise ratio and spectral resolution could improve 2-HG detection. Long echo time (long-TE) semi-localization by adiabatic selective refocusing (semi-LASER) (TE = 110 ms) is a promising method for improved 2-HG detection in vivo at either 3 or 7 T owing to the use of broad-band adiabatic localization. Using previously published semi-LASER methods at 3 and 7 T, this study directly compares the detectability of 2-HG in phantoms and in vivo across nine patients. Cramér-Rao lower bounds (CRLBs) of 2-HG fitting were found to be significantly lower at 7 T (6 ± 2%) relative to 3 T (15 ± 7%) (p = 0.0019), yet were larger at 7 T in an IDH wild-type patient. Although no increase in SNR was detected at 7 T (77 ± 26) relative to 3 T (77 ± 30), the detection of 2-HG was greatly enhanced through an improved spectral profile and increased resolution at 7 T. 7 T had a large effect on pairwise fitting correlations between γ-aminobutyric acid (GABA) and 2-HG (p = 0.004), and resulted in smaller coefficients. The increased sensitivity for 2-HG detection using long-TE acquisition at 7 T may allow for more rapid estimation of 2-HG (within a few spectral averages) together with other associated metabolic markers in glioma.

Weighing benefits and risks in aspects of security, privacy and adoption of technology in a value-based healthcare system.

BACKGROUND: Technology can potentially enable the implementation of a value-based healthcare system, where the impact of quality of care is offered at optimised cost for maximised patient benefit. Technology can deliver value by aiding in data collection to evaluate outcomes and measure costs on a patient and population level. Healthcare organisations, however, face several challenges and risks that result almost exclusively from the use of these technologies. DISCUSSION: Some challenges associated with healthcare technology include their unsustainability, due to lack of scale-up plans and timely evaluations. Other risks include noncompliance with data protection policies, inadequate data governance, and overestimated expectations resulting from the rapid introduction of new technologies. CONCLUSION: Organisations need to consider the risks and challenges associated with the use of technology and develop comprehensive strategies that mitigate factors leading to non-adoption and to realise benefits for achieving a value-based healthcare system.

Feasibility of an artificial intelligence phone call for postoperative care following cataract surgery in a diverse population: two phase prospective study protocol.

INTRODUCTION: Artificial intelligence (AI) development has led to improvements in many areas of medicine. Canada has workforce pressures in delivering cataract care. A potential solution is using AI technology that can automate care delivery, increase effectiveness and decrease burdens placed on patients and the healthcare system. This study assesses the use of 'Dora', an example of an AI assistant that is able to deliver a regulated autonomous, voice-based, natural-language consultation with patients over the telephone. Dora is used in routine practice in the UK, but this study seeks to assess the safety, usability, acceptability and cost-effectiveness of using the technology in Canada. METHODS AND ANALYSIS: This is a two-phase prospective single-centred trial. An expected 250 patients will be recruited for each phase of the study. For Phase I of the study, Dora will phone patients at postoperative week 1 and for Phase II of the study, Dora will phone patients within 24hours of their cataract surgery and again at postoperative week 1. We will evaluate the agreement between Dora and a supervising clinician regarding the need for further review based on the patients' symptoms. A random sample of patients will undergo the System Usability Scale followed by an extended semi-structured interview. The primary outcome of agreement between Dora and the supervisor will be assessed using the kappa statistic. Qualitative data from the interviews will further gauge patient opinions about Dora's usability, appropriateness and level of satisfaction. ETHICS AND DISSEMINATION: Research Ethics Board William Osler Health System (ID: 22-0044) has approved this study and will be conducted by guidelines of Declaration of Helsinki. Master-linking sheet will contain the patient chart identification (ID), full name, date of birth and study ID. Results will be shared through peer-reviewed journals and presentations at conferences.

Successful treatment of testicular pain with peripheral nerve stimulation of the cutaneous branch of the ilioinguinal and genital branch of the genitofemoral nerves.

OBJECTIVES: To assess the effect of peripheral nerve stimulation on neuropathic testicular pain. MATERIAL AND METHODS: A 30-year-old man with a four-year history of chronic testicular pain following scrotal hydrocele surgery had two percutaneous leads implanted in his groin and low-frequency stimulation of the cutaneous branch of the ilioinguinal and genital branch of the genitofemoral nerves. RESULT: At seven-month follow-up, the pain intensity had declined from 9/10 to 2/10 on the numeric rating scale. CONCLUSION: We report the successful implantation of an ilioinguinal and genitofemoral nerve stimulator for sustained suppression of intractable neuropathic testicular pain.

Investigating the Impact of Automation on the Health Care Workforce Through Autonomous Telemedicine in the Cataract Pathway: Protocol for a Multicenter Study.

BACKGROUND: While digital health innovations are increasingly being adopted by health care organizations, implementation is often carried out without considering the impacts on frontline staff who will be using the technology and who will be affected by its introduction. The enthusiasm surrounding the use of artificial intelligence (AI)-enabled digital solutions in health care is tempered by uncertainty around how it will change the working lives and practices of health care professionals. Digital enablement can be viewed as facilitating enhanced effectiveness and efficiency by improving services and automating cognitive labor, yet the implementation of such AI technology comes with challenges related to changes in work practices brought by automation. This research explores staff experiences before and after care pathway automation with an autonomous clinical conversational assistant, Dora (Ufonia Ltd), that is able to automate routine clinical conversations. OBJECTIVE: The primary objective is to examine the impact of AI-enabled automation on clinicians, allied health professionals, and administrators who provide or facilitate health care to patients in high-volume, low-complexity care pathways. In the process of transforming care pathways through automation of routine tasks, staff will increasingly "work at the top of their license." The impact of this fundamental change on the professional identity, well-being, and work practices of the individual is poorly understood at present. METHODS: We will adopt a multiple case study approach, combining qualitative and quantitative data collection methods, over 2 distinct phases, namely phase A (preimplementation) and phase B (postimplementation). RESULTS: The analysis is expected to reveal the interrelationship between Dora and those affected by its introduction. This will reveal how tasks and responsibilities have changed or shifted, current tensions and contradictions, ways of working, and challenges, benefits, and opportunities as perceived by those on the frontlines of the health care system. The findings will enable a better understanding of the resistance or susceptibility of different stakeholders within the health care workforce and encourage managerial awareness of differing needs, demands, and uncertainties. CONCLUSIONS: The implementation of AI in the health care sector, as well as the body of research on this topic, remain in their infancy. The project's key contribution will be to understand the impact of AI-enabled automation on the health care workforce and their work practices. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): PRR1-10.2196/49374.

User-acceptability of an automated telephone call for post-operative follow-up after uncomplicated cataract surgery.

BACKGROUND: Innovative technology is recommended to address the current capacity challenges facing the NHS. This study evaluates the patient acceptability of automated telephone follow-up after routine cataract surgery using Dora (Ufonia Limited, Oxford, United Kingdom), which to our knowledge is the first AI-powered clinical assistant to be used in the NHS. Dora has a natural-language, phone conversation with patients about their symptoms after cataract surgery. METHODS: This is a prospective mixed-methods cohort study that was conducted at Buckinghamshire Healthcare NHS Foundation Trust. All patients who were followed up using Dora were asked to give a Net Promoter Score (NPS), and 24 patients were randomly selected to complete the validated Telephone Usability Questionnaire (TUQ) as well as extended semi-structured interviews that underwent thematic analysis. RESULTS: A total of 170 autonomous calls were completed. The median NPS score was 9 out of 10. The TUQ (scored out of 5) showed high rates of acceptability, with an overall mean score of 4.0. Simplicity, time saving, and ease of use scored the highest with a median of 5, whilst 'speaking to Dora feels the same as speaking to a clinician' scored a median of 3. The main themes extracted from the qualitative data were 'I can see why you're doing it', 'It went quite well actually', 'I just trust human beings I suppose'. CONCLUSION: We found high levels of patient acceptability when using Dora across three acceptability measures. Dora provides a potential solution to reduce pressure on hospital capacity whilst also providing a convenient service for patients.

Safety and Acceptability of a Natural Language Artificial Intelligence Assistant to Deliver Clinical Follow-up to Cataract Surgery Patients: Proposal.

BACKGROUND: Due to an aging population, the demand for many services is exceeding the capacity of the clinical workforce. As a result, staff are facing a crisis of burnout from being pressured to deliver high-volume workloads, driving increasing costs for providers. Artificial intelligence (AI), in the form of conversational agents, presents a possible opportunity to enable efficiency in the delivery of care. OBJECTIVE: This study aims to evaluate the effectiveness, usability, and acceptability of Dora agent: Ufonia's autonomous voice conversational agent, an AI-enabled autonomous telemedicine call for the detection of postoperative cataract surgery patients who require further assessment. The objectives of this study are to establish Dora's efficacy in comparison with an expert clinician, determine baseline sensitivity and specificity for the detection of true complications, evaluate patient acceptability, collect evidence for cost-effectiveness, and capture data to support further development and evaluation. METHODS: Using an implementation science construct, the interdisciplinary study will be a mixed methods phase 1 pilot establishing interobserver reliability of the system, usability, and acceptability. This will be done using the following scales and frameworks: the system usability scale; assessment of Health Information Technology Interventions in Evidence-Based Medicine Evaluation Framework; the telehealth usability questionnaire; and the Non-Adoption, Abandonment, and Challenges to the Scale-up, Spread and Suitability framework. RESULTS: The evaluation is expected to show that conversational technology can be used to conduct an accurate assessment and that it is acceptable to different populations with different backgrounds. In addition, the results will demonstrate how successfully the system can be delivered in organizations with different clinical pathways and how it can be integrated with their existing platforms. CONCLUSIONS: The project's key contributions will be evidence of the effectiveness of AI voice conversational agents and their associated usability and acceptability. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): PRR1-10.2196/27227.

Secretion of functionally active complement factor H related protein 5 (FHR5) by primary tumour cells derived from Glioblastoma Multiforme patients.

The complement system is an important humoral immune surveillance mechanism against tumours. However, many malignant tumours are resistant to complement mediated lysis. Here, we report secretion of complement factor H related protein 5 (FHR5) by primary tumour cells derived from Glioblastoma multiforme (GBM) patients. We investigated whether the secreted FHR5 exhibited functional activity similar to factor H, including inhibition of complement mediated lysis, acting as a co-factor for factor I mediated cleavage of C3b, and decay acceleration of C3 convertase. Immunoblotting analysis of primary GBM cells (B30, B31 and B33) supernatant showed the active secretion of FHR5, but not of Factor H. ELISA revealed that the secretion of soluble GBM-FHR5 by cultured GBM cells increased in a time-dependent manner. Primary GBM-FHR5 inhibited complement mediated lysis, possessed co-factor activity for factor I mediated cleavage and displayed decay acceleration of C3 convertase. In summary, we detected the secretion of FHR5 by primary GBM cells B30, B31 and B33. The results demonstrated that GBM-FHR5 shares biological function with FH as a mechanism primary GBM cells potentially use to resist complement mediated lysis.

A Noninvasive Comparison Study between Human Gliomas with IDH1 and IDH2 Mutations by MR Spectroscopy.

The oncogenes that are expressed in gliomas reprogram particular pathways of glucose, amino acids, and fatty acid metabolism. Mutations in isocitrate dehydrogenase genes (IDH1/2) in diffuse gliomas are associated with abnormally high levels of 2-hydroxyglutarate (2-HG) levels. The aim of this study was to determine whether metabolic reprogramming associated with IDH mutant gliomas leads to additional ¹H MRS-detectable differences between IDH1 and IDH2 mutations, and to identify metabolites correlated with 2-HG. A total of 21 glioma patients (age= 37 ± 11, 13 males) were recruited for magnetic resonance spectroscopy (MRS) using semi-localization by adiabatic selective refocusing pulse sequence at an ultra-high-field (7T). For 20 patients, the tumor mutation subtype was confirmed by immunohistochemistry and DNA sequencing. LCModel analysis was applied for metabolite quantification. A two-sample t-test was used for metabolite comparisons between IDH1 (n = 15) and IDH2 (n = 5) mutant gliomas. The Pearson correlation coefficients between 2-HG and associated metabolites were calculated. A Bonferroni correction was applied for multiple comparison. IDH2 mutant gliomas have a higher level of 2-HG/tCho (total choline=phosphocholine+glycerylphosphorylcholine) (2.48 ± 1.01vs.0.72 ± 0.38, Pc < 0.001) and myo-Inositol/tCho (2.70 ± 0.90 vs. 1.46 ± 0.51, Pc = 0.011) compared to IDH1 mutation gliomas. Associated metabolites, myo-Inositol and glucose+taurine were correlated with 2-HG levels. These results show the improved characterization of the metabolic pathways in IDH1 and IDH2 gliomas for precision medicine.

The Internet of Things in Health Care in Oxford: Protocol for Proof-of-Concept Projects.

BACKGROUND: Demands on health services across are increasing because of the combined challenges of an expanding and aging population, alongside complex comorbidities that transcend the classical boundaries of modern health care. Continuing to provide and coordinate care in the current manner is not a viable route to sustain the improvements in health outcomes observed in recent history. To ensure that there continues to be improvement in patient care, prevention of disease, and reduced burden on health systems, it is essential that we adapt our models of delivery. Providers of health and social care are evolving to face these pressures by changing the way they think about the care system and, importantly, how to involve patients in the planning and delivery of services. OBJECTIVE: The objective of this paper is to provide (1) an overview of the current state of Internet of Things (IoT) and key implementation considerations, (2) key use cases demonstrating technology capabilities, (3) an overview of the landscape for health care IoT use in Oxford, and (4) recommendations for promoting the IoT via collaborations between higher education institutions and industry proof-of-concept (PoC) projects. METHODS: This study describes the PoC projects that will be created to explore cost-effectiveness, clinical efficacy, and user adoption of Internet of Medical Things systems. The projects will focus on 3 areas: (1) bring your own device integration, (2) chronic disease management, and (3) personal health records. RESULTS: This study is funded by Research England's Connecting Capability Fund. The study started in March 2018, and results are expected by the end of 2019. CONCLUSIONS: Embracing digital solutions to support the evolution and transformation of health services is essential. Importantly, this should not simply be undertaken by providers in isolation. It must embrace and exploit the advances being seen in the consumer devices, national rollout of high-speed broadband services, and the rapidly expanding medical device industry centered on mobile and wearable technologies. Oxford University Hospitals and its partner providers, patients, and stakeholders are building on their leading position as an exemplar site for digital maturity in the National Health Service to implement and evaluate technologies and solutions that will capitalize on the IoT. Although early in the application to health, the IoT and the potential it provides to make the patient a partner at the center of decisions about care represent an exciting opportunity. If achieved, a fully connected and interoperable health care environment will enable continuous acquisition and real-time analysis of patient data, offering unprecedented ability to monitor patients, manage disease, and potentially deliver early diagnosis. The clinical benefit of this is clear, but additional patient benefit and value will be gained from being able to provide expert care at home or close to home. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/12077.

Noninvasive Quantification of 2-Hydroxyglutarate in Human Gliomas with IDH1 and IDH2 Mutations.

Mutations in the isocitrate dehydrogenase genes (IDH1/2) occur often in diffuse gliomas, where they are associated with abnormal accumulation of the oncometabolite 2-hydroxyglutarate (2-HG). Monitoring 2-HG levels could provide prognostic information in this disease, but detection strategies that are noninvasive and sufficiently quantitative have yet to be developed. In this study, we address this need by presenting a proton magnetic resonance spectroscopy ((1)H-MRS) acquisition scheme that uses an ultrahigh magnetic field (≥ 7T) capable of noninvasively detecting 2-HG with quantitative measurements sufficient to differentiate mutant cytosolic IDH1 and mitochondrial IDH2 in human brain tumors. Untargeted metabolomics analysis of in vivo (1)H-MRS spectra discriminated between IDH-mutant tumors and healthy tissue, and separated IDH1 from IDH2 mutations. High-quality spectra enabled the quantification of neurochemical profiles consisting of at least eight metabolites, including 2-HG, glutamate, lactate, and glutathione in both tumor and healthy tissue voxels. Notably, IDH2 mutation produced more 2-HG than IDH1 mutation, consistent with previous findings in cell culture. By offering enhanced sensitivity and specificity, this scheme can quantitatively detect 2-HG and associated metabolites that may accumulate during tumor progression, with implications to better monitor patient responses to therapy.

Short-term adaptation of the VOR: non-retinal-slip error signals and saccade substitution.

We studied short-term (30 min) adaptation of the vestibulo-ocular reflex (VOR) in five normal humans using a "position error" stimulus without retinal image motion. Both before and after adaptation a velocity gain (peak slow-phase eye velocity/peak head velocity) and a position gain (total eye movement during chair rotation/amplitude of chair motion) were measured in darkness using search coils. The vestibular stimulus was a brief ( approximately 700 ms), 15 degrees chair rotation in darkness (peak velocity 43 degrees /s). To elicit adaptation, a straight-ahead fixation target disappeared during chair movement and when the chair stopped the target reappeared at a new location in front of the subject for gain-decrease (x0) adaptation, or 10 degrees opposite to chair motion for gain-increase (x1.67) adaptation. This position-error stimulus was effective at inducing VOR adaptation, though for gain-increase adaptation the primary strategy was to substitute augmenting saccades during rotation while for gain-decrease adaptation both corrective saccades and a decrease in slow-phase velocity occurred. Finally, the presence of the position-error signal alone, at the end of head rotation, without any attempt to fix upon it, was not sufficient to induce adaptation. Adaptation did occur, however, if the subject did make a saccade to the target after head rotation, or even if the subject paid attention to the new location of the target without actually looking at it.