Artificial intelligence for ultrasound scanning in regional anaesthesia: a scoping review of the evidence from multiple disciplines.

BACKGROUND: Artificial intelligence (AI) for ultrasound scanning in regional anaesthesia is a rapidly developing interdisciplinary field. There is a risk that work could be undertaken in parallel by different elements of the community but with a lack of knowledge transfer between disciplines, leading to repetition and diverging methodologies. This scoping review aimed to identify and map the available literature on the accuracy and utility of AI systems for ultrasound scanning in regional anaesthesia. METHODS: A literature search was conducted using Medline, Embase, CINAHL, IEEE Xplore, and ACM Digital Library. Clinical trial registries, a registry of doctoral theses, regulatory authority databases, and websites of learned societies in the field were searched. Online commercial sources were also reviewed. RESULTS: In total, 13,014 sources were identified; 116 were included for full-text review. A marked change in AI techniques was noted in 2016-17, from which point on the predominant technique used was deep learning. Methods of evaluating accuracy are variable, meaning it is impossible to compare the performance of one model with another. Evaluations of utility are more comparable, but predominantly gained from the simulation setting with limited clinical data on efficacy or safety. Study methodology and reporting lack standardisation. CONCLUSIONS: There is a lack of structure to the evaluation of accuracy and utility of AI for ultrasound scanning in regional anaesthesia, which hinders rigorous appraisal and clinical uptake. A framework for consistent evaluation is needed to inform model evaluation, allow comparison between approaches/models, and facilitate appropriate clinical adoption.

Adaptive strategies used by surgical teams under pressure: an interview study among senior healthcare professionals in four major hospitals in the United Kingdom.

BACKGROUND: Healthcare systems are operating under substantial pressures, and often simply cannot provide the standard of care they aspire to within the available resources. Organisations, managers, and individual clinicians make constant adaptations in response to these pressures, which are typically improvised, highly variable and not coordinated across clinical teams. The purpose of this study was to identify and describe the types of everyday pressures experienced by surgical teams and the adaptive strategies they use to respond to these pressures. METHODS: We conducted interviews with 20 senior multidisciplinary healthcare professionals from surgical teams in four major hospitals in the United Kingdom. The interviews explored the types of everyday pressures staff were experiencing, the strategies they use to adapt, and how these strategies might be taught to others. RESULTS: The primary pressures described by senior clinicians in surgery were increased numbers and complexity of patients alongside shortages in staff, theatre space and post-surgical beds. These pressures led to more difficult working conditions (e.g. high workloads) and problems with system functioning such as patient flow and cancellation of lists. Strategies for responding to these pressures were categorised into increasing or flexing resources, controlling and prioritising patient demand and strategies for managing the workload (scheduling for efficiency, communication and coordination, leadership, and teamwork strategies). CONCLUSIONS: Teams are deploying a range of strategies and making adaptations to the way care is delivered. These findings could be used as the basis for training programmes for surgical teams to develop coordinated strategies for adapting under pressure and to assess the impact of different combinations of strategies on patient safety and surgical outcomes.

Adopting human factors in early phase and experimental medicine research: A nested pilot study observing controlled human infection with SARS-CoV-2.

AIMS: The influence of human factors on safety in healthcare settings is well established, with targeted interventions reducing risk and enhancing team performance. In experimental and early phase clinical research participant safety is paramount and safeguarded by guidelines, protocolized care and staff training; however, the real-world interaction and implementation of these risk-mitigating measures has never been subjected to formal system-based assessment. METHODS: Independent structured observations, systematic review of study documents, and interviews and focus groups were used to collate data on three key tasks undertaken in a clinical research facility (CRF) during a SARS CoV-2 controlled human infection model (CHIM) study. The Systems Engineering Initiative for Patient Safety (SEIPS) was employed to analyse and categorize findings, and develop recommendations for safety interventions. RESULTS: High levels of team functioning and a clear focus on participant safety were evident throughout the study. Despite this, latent risks in both study-specific and CRF work systems were identified in all four SEIPS domains (people, environment, tasks and tools). Fourteen actionable recommendations were generated collaboratively. These included inter-organization and inter-study standardization, optimized checklists for safety critical tasks, and use of simulation for team training and exploration of work systems. CONCLUSIONS: This pioneering application of human factors techniques to analyse work systems during the conduct of research in a CRF revealed risks unidentified by routine review and appraisal, and despite international guideline adherence. SEIPS may aid categorization of system problems and the formulation of recommendations that reduce risk and mitigate potential harm applicable across a trials portfolio.

A study of validity and usability evidence for non-technical skills assessment tools in simulated adult resuscitation scenarios.

BACKGROUND: Non-technical skills (NTS) assessment tools are widely used to provide formative and summative assessment for healthcare professionals and there are now many of them. This study has examined three different tools designed for similar settings and gathered evidence to test their validity and usability. METHODS: Three NTS assessment tools designed for use in the UK were used by three experienced faculty to review standardized videos of simulated cardiac arrest scenarios: ANTS (Anesthetists' Non-Technical Skills), Oxford NOTECHS (Oxford NOn-TECHnical Skills) and OSCAR (Observational Skill based Clinical Assessment tool for Resuscitation). Internal consistency, interrater reliability and quantitative and qualitative analysis of usability were analyzed for each tool. RESULTS: Internal consistency and interrater reliability (IRR) varied considerably for the three tools across NTS categories and elements. Intraclass correlation scores of three expert raters ranged from poor (task management in ANTS [0.26] and situation awareness (SA) in Oxford NOTECHS [0.34]) to very good (problem solving in Oxford NOTECHS [0.81] and cooperation [0.84] and SA [0.87] in OSCAR). Furthermore, different statistical tests of IRR produced different results for each tool. Quantitative and qualitative examination of usability also revealed challenges in using each tool. CONCLUSIONS: The lack of standardization of NTS assessment tools and training in their use is unhelpful for healthcare educators and students. Educators require ongoing support in the use of NTS assessment tools for the evaluation of individual healthcare professionals or healthcare teams. Summative or high-stakes examinations using NTS assessment tools should be undertaken with at least two assessors to provide consensus scoring. In light of the renewed focus on simulation as an educational tool to support and enhance training recovery in the aftermath of COVID-19, it is even more important that assessment of these vital skills is standardized, simplified and supported with adequate training.

Evaluation of the impact of assistive artificial intelligence on ultrasound scanning for regional anaesthesia.

BACKGROUND: Ultrasound-guided regional anaesthesia relies on the visualisation of key landmark, target, and safety structures on ultrasound. However, this can be challenging, particularly for inexperienced practitioners. Artificial intelligence (AI) is increasingly being applied to medical image interpretation, including ultrasound. In this exploratory study, we evaluated ultrasound scanning performance by non-experts in ultrasound-guided regional anaesthesia, with and without the use of an assistive AI device. METHODS: Twenty-one anaesthetists, all non-experts in ultrasound-guided regional anaesthesia, underwent a standardised teaching session in ultrasound scanning for six peripheral nerve blocks. All then performed a scan for each block; half of the scans were performed with AI assistance and half without. Experts assessed acquisition of the correct block view and correct identification of sono-anatomical structures on each view. Participants reported scan confidence, experts provided a global rating score of scan performance, and scans were timed. RESULTS: Experts assessed 126 ultrasound scans. Participants acquired the correct block view in 56/62 (90.3%) scans with the device compared with 47/62 (75.1%) without (P=0.031, two data points lost). Correct identification of sono-anatomical structures on the view was 188/212 (88.8%) with the device compared with 161/208 (77.4%) without (P=0.002). There was no significant overall difference in participant confidence, expert global performance score, or scan time. CONCLUSIONS: Use of an assistive AI device was associated with improved ultrasound image acquisition and interpretation. Such technology holds potential to augment performance of ultrasound scanning for regional anaesthesia by non-experts, potentially expanding patient access to these techniques. CLINICAL TRIAL REGISTRATION: NCT05156099.

Assistive artificial intelligence for ultrasound image interpretation in regional anaesthesia: an external validation study.

BACKGROUND: Ultrasonound is used to identify anatomical structures during regional anaesthesia and to guide needle insertion and injection of local anaesthetic. ScanNav Anatomy Peripheral Nerve Block (Intelligent Ultrasound, Cardiff, UK) is an artificial intelligence-based device that produces a colour overlay on real-time B-mode ultrasound to highlight anatomical structures of interest. We evaluated the accuracy of the artificial-intelligence colour overlay and its perceived influence on risk of adverse events or block failure. METHODS: Ultrasound-guided regional anaesthesia experts acquired 720 videos from 40 volunteers (across nine anatomical regions) without using the device. The artificial-intelligence colour overlay was subsequently applied. Three more experts independently reviewed each video (with the original unmodified video) to assess accuracy of the colour overlay in relation to key anatomical structures (true positive/negative and false positive/negative) and the potential for highlighting to modify perceived risk of adverse events (needle trauma to nerves, arteries, pleura, and peritoneum) or block failure. RESULTS: The artificial-intelligence models identified the structure of interest in 93.5% of cases (1519/1624), with a false-negative rate of 3.0% (48/1624) and a false-positive rate of 3.5% (57/1624). Highlighting was judged to reduce the risk of unwanted needle trauma to nerves, arteries, pleura, and peritoneum in 62.9-86.4% of cases (302/480 to 345/400), and to increase the risk in 0.0-1.7% (0/160 to 8/480). Risk of block failure was reported to be reduced in 81.3% of scans (585/720) and to be increased in 1.8% (13/720). CONCLUSIONS: Artificial intelligence-based devices can potentially aid image acquisition and interpretation in ultrasound-guided regional anaesthesia. Further studies are necessary to demonstrate their effectiveness in supporting training and clinical practice. CLINICAL TRIAL REGISTRATION: NCT04906018.

Recommendations for anatomical structures to identify on ultrasound for the performance of intermediate and advanced blocks in ultrasound-guided regional anesthesia.

Recent recommendations describe a set of core anatomical structures to identify on ultrasound for the performance of basic blocks in ultrasound-guided regional anesthesia (UGRA). This project aimed to generate consensus recommendations for core structures to identify during the performance of intermediate and advanced blocks. An initial longlist of structures was refined by an international panel of key opinion leaders in UGRA over a three-round Delphi process. All rounds were conducted virtually and anonymously. Blocks were considered twice in each round: for "orientation scanning" (the dynamic process of acquiring the final view) and for "block view" (which visualizes the block site and is maintained for needle insertion/injection). A "strong recommendation" was made if ≥75% of participants rated any structure as "definitely include" in any round. A "weak recommendation" was made if >50% of participants rated it as "definitely include" or "probably include" for all rounds, but the criterion for strong recommendation was never met. Structures which did not meet either criterion were excluded. Forty-one participants were invited and 40 accepted; 38 completed all three rounds. Participants considered the ultrasound scanning for 19 peripheral nerve blocks across all three rounds. Two hundred and seventy-four structures were reviewed for both orientation scanning and block view; a "strong recommendation" was made for 60 structures on orientation scanning and 44 on the block view. A "weak recommendation" was made for 107 and 62 structures, respectively. These recommendations are intended to help standardize teaching and research in UGRA and support widespread and consistent practice.

Non-technical Skills for Medical Students: Validating the Tools of the Trade.

The Medical Students' Non-Technical Skills (Medi-StuNTS) is a behavioural marker scheme (BMS) designed to assess non-technical skills (NTS) in medical students in emergency simulations. This study aimed to assess the evidence for validity and usability of Medi-StuNTS by naive, near-peer educators. Nine doctors assessed four students in simulations of common medical emergencies. The scores were used to assess inter-rater reliability, inter-class correlation, and observability. Students and assessors completed questionnaires that assessed the tool's usability and consequence. Inter-rater agreement across all skill elements was "high" with rWG scores >0.8. An inter-class correlation was "good" with ICC3K kappa scores of 0.86 and 0.89 overall, when measured per simulation and per skills element respectively. Overall skill observability was high (>80%) except for coping with stress. Assessors found the tool "difficult to use" but "useful for feeding back in a constructive way". Students appreciated the comprehensiveness of the feedback as well as knowing what to expect during debriefs. This study has shown that the Medi-StuNTS BMS has good usability and evidence of validity in naive assessors and near-peer educators. It shows the particularly good internal structure and overall beneficial consequences. Further study will be necessary to understand how best to deploy it in formative and summative contexts.

Artificial Intelligence: Innovation to Assist in the Identification of Sono-anatomy for Ultrasound-Guided Regional Anaesthesia.

Ultrasound-guided regional anaesthesia (UGRA) involves the targeted deposition of local anaesthesia to inhibit the function of peripheral nerves. Ultrasound allows the visualisation of nerves and the surrounding structures, to guide needle insertion to a perineural or fascial plane end point for injection. However, it is challenging to develop the necessary skills to acquire and interpret optimal ultrasound images. Sound anatomical knowledge is required and human image analysis is fallible, limited by heuristic behaviours and fatigue, while its subjectivity leads to varied interpretation even amongst experts. Therefore, to maximise the potential benefit of ultrasound guidance, innovation in sono-anatomical identification is required.Artificial intelligence (AI) is rapidly infiltrating many aspects of everyday life. Advances related to medicine have been slower, in part because of the regulatory approval process needing to thoroughly evaluate the risk-benefit ratio of new devices. One area of AI to show significant promise is computer vision (a branch of AI dealing with how computers interpret the visual world), which is particularly relevant to medical image interpretation. AI includes the subfields of machine learning and deep learning, techniques used to interpret or label images. Deep learning systems may hold potential to support ultrasound image interpretation in UGRA but must be trained and validated on data prior to clinical use.Review of the current UGRA literature compares the success and generalisability of deep learning and non-deep learning approaches to image segmentation and explains how computers are able to track structures such as nerves through image frames. We conclude this review with a case study from industry (ScanNav Anatomy Peripheral Nerve Block; Intelligent Ultrasound Limited). This includes a more detailed discussion of the AI approach involved in this system and reviews current evidence of the system performance.The authors discuss how this technology may be best used to assist anaesthetists and what effects this may have on the future of learning and practice of UGRA. Finally, we discuss possible avenues for AI within UGRA and the associated implications.

Exploring the utility of assistive artificial intelligence for ultrasound scanning in regional anesthesia.

INTRODUCTION: Ultrasound-guided regional anesthesia (UGRA) involves the acquisition and interpretation of ultrasound images to delineate sonoanatomy. This study explores the utility of a novel artificial intelligence (AI) device designed to assist in this task (ScanNav Anatomy Peripheral Nerve Block; ScanNav), which applies a color overlay on real-time ultrasound to highlight key anatomical structures. METHODS: Thirty anesthesiologists, 15 non-experts and 15 experts in UGRA, performed 240 ultrasound scans across nine peripheral nerve block regions. Half were performed with ScanNav. After scanning each block region, participants completed a questionnaire on the utility of the device in relation to training, teaching, and clinical practice in ultrasound scanning for UGRA. Ultrasound and color overlay output were recorded from scans performed with ScanNav. Experts present during the scans (real-time experts) were asked to assess potential for increased risk associated with use of the device (eg, needle trauma to safety structures). This was compared with experts who viewed the AI scans remotely. RESULTS: Non-experts were more likely to provide positive and less likely to provide negative feedback than experts (p=0.001). Positive feedback was provided most frequently by non-experts on the potential role for training (37/60, 61.7%); for experts, it was for its utility in teaching (30/60, 50%). Real-time and remote experts reported a potentially increased risk in 12/254 (4.7%) vs 8/254 (3.1%, p=0.362) scans, respectively. DISCUSSION: ScanNav shows potential to support non-experts in training and clinical practice, and experts in teaching UGRA. Such technology may aid the uptake and generalizability of UGRA. TRIAL REGISTRATION NUMBER: NCT04918693.

International consensus on anatomical structures to identify on ultrasound for the performance of basic blocks in ultrasound-guided regional anesthesia.

There is no universally agreed set of anatomical structures that must be identified on ultrasound for the performance of ultrasound-guided regional anesthesia (UGRA) techniques. This study aimed to produce standardized recommendations for core (minimum) structures to identify during seven basic blocks. An international consensus was sought through a modified Delphi process. A long-list of anatomical structures was refined through serial review by key opinion leaders in UGRA. All rounds were conducted remotely and anonymously to facilitate equal contribution of each participant. Blocks were considered twice in each round: for "orientation scanning" (the dynamic process of acquiring the final view) and for the "block view" (which visualizes the block site and is maintained for needle insertion/injection). Strong recommendations for inclusion were made if ≥75% of participants rated a structure as "definitely include" in any round. Weak recommendations were made if >50% of participants rated a structure as "definitely include" or "probably include" for all rounds (but the criterion for "strong recommendation" was never met). Thirty-six participants (94.7%) completed all rounds. 128 structures were reviewed; a "strong recommendation" is made for 35 structures on orientation scanning and 28 for the block view. A "weak recommendation" is made for 36 and 20 structures, respectively. This study provides recommendations on the core (minimum) set of anatomical structures to identify during ultrasound scanning for seven basic blocks in UGRA. They are intended to support consistent practice, empower non-experts using basic UGRA techniques, and standardize teaching and research.

Simulation past, present and future-a decade of progress in simulation-based education in the UK.

This review will present developments in simulation-based education (SBE) over the past decade with a focus on activity in the UK's National Health Service and the role of the national society (the Association for Simulation Practice in Healthcare). The article covers the evolution of strategic changes for the use of SBE in the UK and the operational challenges faced by clinicians and other faculty in SBE. The expansion of the evidence base to support SBE in healthcare both in technical skills and, more broadly, in interprofessional team training is explored. Finally, the wider role of simulation in patient safety and healthcare systems, including testing pathways and the development of cognitive aids and involvement of patients in SBE is considered.

Validation of the Partners at Care Transitions Measure (PACT-M): assessing the quality and safety of care transitions for older people in the UK.

BACKGROUND: The Partners at Care Transitions Measure (PACT-M) is a patient-reported questionnaire for evaluation of the quality and safety of care transitions from hospital to home, as experienced by older adults. PACT-M has two components; PACT-M 1 to capture the immediate post discharge period and PACT-M 2 to assess the experience of managing care at home. In this study, we aim to examine the psychometric properties, factor structure, validity and reliability of the PACT-M. METHODS: We administered the PACT-M over the phone and by mail, within one week post discharge with 138 participants and one month after discharge with 110 participants. We performed principal components analysis and factors were assessed for internal consistency, reliability and construct validity. RESULTS: Reliability was assessed by calculating Cronbach's alpha for the 9-item PACT-M 1 and 8-item PACT-M 2 and exploratory factor analysis was performed to evaluate dimensionality of the scales. Principal components analysis was chosen using pair-wise deletion. Both PACT-M 1 and PACT-M 2 showed high internal consistency and good internal reliability values and conveyed unidimensional scale characteristics with high reliability scores; above 0.8. CONCLUSIONS: The PACT-M has shown evidence to suggest that it is a reliable measure to capture patients' perception of the quality of discharge arrangements and also on patients' ability to manage their care at home one month post discharge. PACT-M 1 is a marker of patient experience of transition and PACT-M 2 of coping at home.

In situ deteriorating patient simulation in general practice.

BACKGROUND: GP practices have limited access to medical emergency training and basic life support is often taught out of context as a skills-based event. AIM: To develop and evaluate a whole team integrated simulation-based education, to enhance learning, change behaviours and provide safer care. METHOD: Phase 1: 10 practices piloted a 3-hour programme delivering 40 minutes BLS and AED skills and 2-hour deteriorating patient simulation. Three scenarios where developed: adult chest pain, child anaphylaxis and baby bronchiolitis. An adult simulation patient and relative were used and a child and baby manikin. Two facilitators trained in coaching and debriefing used the 3D debriefing model. Phase 2: 12 new practices undertook identical training derived from Phase 1, with pre- and post-course questionnaires. Teams were scored on: team working, communication, early recognition and systematic approach. The team developed action plans derived from their learning to inform future response. Ten of the 12 practices from Phase 2 received an emergency drill within 6 months of the original session. Three to four members of the whole team integrated training, attended the drill, but were unaware of the nature of the scenario before. Scoring was repeated and action plans were revisited to determine behaviour changes. RESULTS: Every emergency drill demonstrated improved scoring in skills and behaviour. CONCLUSION: A combination of: in situ GP simulation, appropriately qualified facilitators in simulation and debriefing, and action plans developed by the whole team suggests safer care for patients experiencing a medical emergency.

Developing a measure to assess the quality of care transitions for older people.

BACKGROUND: The transition of older patients (over 65 years of age) from hospital to their own home is a time when patients are at high risk. No measure currently exists to assess the experience, quality and safety of care transitions relevant to UK population. We aim to describe the development and initial testing of the Partners at Care Transitions Measure (PACT-M) as a patient-reported questionnaire for evaluation of the quality and safety of care transitions from hospital to home in older patients. METHODS: We used an established measure development procedure which includes conceptualising the components of care transitions, item development, conducting a modified Delphi process and pilot-testing of the PACT-M with patients over 65 years old using telephone administration. RESULTS: Pilot testing of the PACT-M suggests that the components identified cover the issues of most importance to patients. Face validity testing showed that the measure in its current form is acceptable to older patients. CONCLUSIONS: The measure developed in this study shows promise for use by those involved in planning, implementing and evaluating discharge care, and could be used to inform interventions to improve the transition from hospital to home for older patients.

Observer-based tools for non-technical skills assessment in simulated and real clinical environments in healthcare: a systematic review.

BACKGROUND: Over the past three decades multiple tools have been developed for the assessment of non-technical skills (NTS) in healthcare. This study was designed primarily to analyse how they have been designed and tested but also to consider guidance on how to select them. OBJECTIVES: To analyse the context of use, method of development, evidence of validity (including reliability) and usability of tools for the observer-based assessment of NTS in healthcare. DESIGN: Systematic review. DATA SOURCES: Search of electronic resources, including PubMed, Embase, CINAHL, ERIC, PsycNet, Scopus, Google Scholar and Web of Science. Additional records identified through searching grey literature (OpenGrey, ProQuest, AHRQ, King's Fund, Health Foundation). STUDY SELECTION: Studies of observer-based tools for NTS assessment in healthcare professionals (or undergraduates) were included if they: were available in English; published between January 1990 and March 2018; assessed two or more NTS; were designed for simulated or real clinical settings and had provided evidence of validity plus or minus usability. 11,101 articles were identified. After limits were applied, 576 were retrieved for evaluation and 118 articles included in this review. RESULTS: One hundred and eighteen studies describing 76 tools for assessment of NTS in healthcare met the eligibility criteria. There was substantial variation in the method of design of the tools and the extent of validity, and usability testing. There was considerable overlap in the skills assessed, and the contexts of use of the tools. CONCLUSION: This study suggests a need for rationalisation and standardisation of the way we assess NTS in healthcare and greater consistency in how tools are developed and deployed.