Smart Multimodal In-Bed Pose Estimation Framework Incorporating Generative Adversarial Neural Network.

Monitoring in-bed pose estimation based on the Internet of Medical Things (IoMT) and ambient technology has a significant impact on many applications such as sleep-related disorders including obstructive sleep apnea syndrome, assessment of sleep quality, and health risk of pressure ulcers. In this research, a new multimodal in-bed pose estimation has been proposed using a deep learning framework. The Simultaneously-collected multimodal Lying Pose (SLP) dataset has been used for performance evaluation of the proposed framework where two modalities including long wave infrared (LWIR) and depth images are used to train the proposed model. The main contribution of this research is the feature fusion network and the use of a generative model to generate RGB images having similar poses to other modalities (LWIR/depth). The inclusion of a generative model helps to improve the overall accuracy of the pose estimation algorithm. Moreover, the method can be generalized for situations to recover human pose both in home and hospital settings under various cover thickness levels. The proposed model is compared with other fusion-based models and shows an improved performance of 97.8% at PCKh @0.5. In addition, performance has been evaluated for different cover conditions, and under home and hospital environments which present improvements using our proposed model.

Enhancing Ophthalmic Triage: identification of new clinical features to support healthcare professionals in triage.

OBJECTIVE: To investigate which features from a patient's history are either high or low risk that could support healthcare professionals in ophthalmic emergency triage. METHODS: Prospective, 12,584 visits from 11,733 adult patients attending an Accident and Emergency department at a single tertiary centre were analysed. Data were collected by ophthalmic nurses working in triage, using an online form from August 2021 to April 2022. Multivariate analysis (MVA) was conducted to identify which features from the patients' history would be associated with emergency care. RESULTS: This study found that 45.5% (5731 patient visits (PV)) required a same day eye emergency examination (SDEE), 11.3% (1416 PV) needed urgent care, and 43.2% (5437 PV) were appropriate for elective consultations with a GP or optometrist. The MVA top ten features that were statistically significant (p < 0.05) that would warrant SDEE with odds ratio (95% CI) were: bilateral eye injury 36.5 [15.6-85.5], unilateral eye injury 25.8 [20.9-31.7], vision loss 4.8 [2.9-7.8], post-operative ophthalmic ( < 4 weeks) 4.6 [3.8-5.7], contact lens wearer 3.9 [3.3-4.7], history of uveitis 3.9 [3.3-4.7], photophobia 2.9 [2.4-3.6], unilateral dark shadow/curtain in vision 2.4 [1.8-3.0], unilateral injected red eye 2.0 [1.8-2.2] and rapid change in visual acuity 1.8 [1.5-2.2]. CONCLUSION: This study characterises presenting features covering almost 100 ophthalmic acute presentations that are commonly seen in emergency and elective care. This information could supplement current red flag indicators and support healthcare professionals in ophthalmic triage. Further research is required to evaluate the cost effectivity and safety of our findings for triaging acute presentations.

A machine learning system to optimise triage in an adult ophthalmic emergency department: a model development and validation study.

Background: A substantial proportion of attendances to ophthalmic emergency departments are for non-urgent presentations. We developed and evaluated a machine learning system (DemDx Ophthalmology Triage System: DOTS) to optimise triage, with the aim of reducing inappropriate emergency attendances and streamlining case referral when necessary. Methods: DOTS was built using retrospective tabular data from 11,315 attendances between July 1st, 2021, to June 15th, 2022 at Moorfields Eye Hospital Emergency Department (MEH) in London, UK. Demographic and clinical features were used as inputs and a triage recommendation was given ("see immediately", "see within a week", or "see electively"). DOTS was validated temporally and compared with triage nurses' performance (1269 attendances at MEH) and validated externally (761 attendances at the Federal University of Minas Gerais - UFMG, Brazil). It was also tested for biases and robustness to variations in disease incidences. All attendances from patients aged at least 18 years with at least one confirmed diagnosis were included in the study. Findings: For identifying ophthalmic emergency attendances, on temporal validation, DOTS had a sensitivity of 94.5% [95% CI 92.3-96.1] and a specificity of 42.4% [38.8-46.1]. For comparison within the same dataset, triage nurses had a sensitivity of 96.4% [94.5-97.7] and a specificity of 25.1% [22.0-28.5]. On external validation at UFMG, DOTS had a sensitivity of 95.2% [92.5-97.0] and a specificity of 32.2% [27.4-37.0]. In simulated scenarios with varying disease incidences, the sensitivity was ≥92.2% and the specificity was ≥36.8%. No differences in sensitivity were found in subgroups of index of multiple deprivation, but the specificity was higher for Q2 when compared to Q4 (Q4 is less deprived than Q2). Interpretation: At MEH, DOTS had similar sensitivity to triage nurses in determining attendance priority; however, with a specificity of 17.3% higher, DOTS resulted in lower rates of patients triaged to be seen immediately at emergency. DOTS showed consistent performance in temporal and external validation, in social-demographic subgroups and was robust to varying relative disease incidences. Further trials are necessary to validate these findings. This system will be prospectively evaluated, considering human-computer interaction, in a clinical trial. Funding: The Artificial Intelligence in Health and Care Award (AI_AWARD01671) of the NHS AI Lab under National Institute for Health and Care Research (NIHR) and the Accelerated Access Collaborative (AAC).

An evaluation of optometric advanced skills within a UK tertiary based setting.

INTRODUCTION: Hospital-based optometrists are undertaking extended roles across ophthalmology that may require them to perform advanced skills (AS). Moorfields Eye Hospital (MEH) is the largest UK employer of hospital-based optometrists, it was sought to investigate which AS are being performed at this centre and how they align to the four pillars of advanced clinical practice (ACP). METHODS: An online survey was sent to MEH optometrists in May 2022 that asked about professional status, sub-specialties worked, qualifications, acquisition and validation of AS, research and leadership. RESULTS: Ninety-six optometrists with mean post-qualification experience was 16.2 years (SD 10.4) responded to the survey. There were 84 AS that covered clinical, leadership and research, with respondents achieving a mean of 11.8 (SD 10.3). Those with independent prescribing (IP) qualifications (n = 52) had a higher number of AS compared to non-IP optometrists (p = 0.03). There were 68 clinical AS across the sub-specialties (23 clinical AS were common in ≥2 sub-specialties), 49 out of 120 clinical AS could be performed by at least 60% of staff. Twenty-six optometrists identified with leadership, 56 had undertaken research/audit, 27 had published within a peer-reviewed journal and half of the time spent in active research was funded. CONCLUSION: AS are being performed by optometrists within a tertiary eye hospital that supports ACP. IP optometrists had higher self-reported AS but current educational frameworks don't accommodate for some AS. Targeted AS courses with competency-based sign-off may further support high-quality patient care. Further research is required on how advanced care practitioners can support workforce transformation.

Usability of an artificially intelligence-powered triage platform for adult ophthalmic emergencies: a mixed methods study.

There is growing demand for emergency-based eyecare services where the majority of those attending do not require urgent ophthalmic management. The Royal College of Ophthalmologists have recommended upskilling and supporting of allied health professionals to support eyecare delivery, where machine learning algorithms could help. A mixed methods study was conducted to evaluate the usability of an artificial intelligence (AI) powered online triage platform for ophthalmology. The interface, usability, safety and acceptability were investigated using a Think Aloud interview and usability questionnaires. Twenty participants who actively examine patients in ophthalmic triage within a tertiary eye centre or primary care setting completed the interview and questionnaires. 90% or more of participants found the platform easy to use, reflected their triage process and were able to interpret the triage outcome, 85% found it safe to use and 95% felt the processing time was fast. A quarter of clinicians reported that they have experienced some uncertainty when triaging in their career and were unsure of using AI, after this study 95% of clinicians were willing to use the platform in their clinical workflow. This study showed the platform interface was acceptable and usable for clinicians actively working in ophthalmic emergency triage.

Technologies for the diagnosis of angle closure glaucoma (ACE): protocol of a prospective, multicentre, cross-sectional diagnostic study.

INTRODUCTION: Angle-closure is responsible for half of all glaucoma blindness globally. Patients with suspected glaucoma require assessment of the drainage angle by an experienced clinician. The goal of this study is to evaluate the diagnostic performance and cost-effectiveness of two non-contact tests, anterior segment OCT (Optical Coherence Tomography) (AS-OCT) and limbal anterior chamber depth for patients referred to hospital with suspected angle closure compared with gonioscopy by ophthalmologist. METHODS AND ANALYSIS: Study design: prospective, multicentre, cross-sectional diagnostic accuracy study. INCLUSION CRITERIA: adults referred from community optometry to hospital with suspected angle closure. PRIMARY OUTCOME: Sensitivity and specificity. SECONDARY OUTCOMES: Positive/negative likelihood ratios, concordance, cost-effectiveness, proportion of patients requiring subsequent clinical assessment by ophthalmologist. SAMPLE SIZE: 600 individuals who have been referred with suspected angle closure from primary care (community optometry). We will have a 95% probability of detecting the true sensitivity of either test to within ±3.5% based on a sensitivity of 90%. The study would also have a 95% probability of detecting the true specificity of either test to within ±5%, assuming a specificity of 75%. ETHICS AND DISSEMINATION: Ethical Review Board approval was obtained. REC reference: 22/LO/0885. Our findings will be disseminated to those involved in eye care services. We will have a knowledge exchange event at the end of the study, published via the Health Technology Assessment web page and in specialist journals. The results will be presented at professional conferences and directly to patients via patient group meetings and the Glaucoma UK charity. TRIAL REGISTRATION NUMBER: ISRCTN15115867.

A Privacy-Preserving Authentication Scheme for Real-Time Medical Monitoring Systems.

In real-time medical monitoring systems, given the significance of medical data and disease symptoms, a secure and always-on connection with the medical centre over the public channels is essential. To this end, an edge-enabled Internet of Medical Things (IoMT) scheme is designed to improve flexibility and scalability of the network and provide seamless connectivity with minimum latency. The entities involved in such network are vulnerable to various attacks and can potentially be compromised. To address this issue, an authentication scheme comprised of digital signature and Authenticated Key Exchange (AKE) protocol is proposed which guarantees only authorized entities get access to the services available in the medical system. Moreover, to fulfill the privacy-preserving, each entity is mapped to a different pseudo-identity. The non-mathematical and performance analysis show that the proposed scheme is robust against various attacks such as impersonation and replay attacks.

Non-contact tests for identifying people at risk of primary angle closure glaucoma.

BACKGROUND: Primary angle closure glaucoma (PACG) accounts for 50% of glaucoma blindness worldwide. More than three-quarters of individuals with PACG reside in Asia. In these populations, PACG often develops insidiously leading to chronically raised intraocular pressure and optic nerve damage, which is often asymptomatic. Non-contact tests to identify people at risk of angle closure are relatively quick and can be carried out by appropriately trained healthcare professionals or technicians as a triage test. If the test is positive, the person will be referred for further specialist assessment. OBJECTIVES: To determine the diagnostic accuracy of non-contact tests (limbal anterior chamber depth (LACD) (van Herick test); oblique flashlight test; scanning peripheral anterior chamber depth analyser (SPAC), Scheimpflug photography; anterior segment optical coherence tomography (AS-OCT), for identifying people with an occludable angle. SEARCH METHODS: We searched the following bibliographic databases 3 October 2019: CENTRAL; MEDLINE; Embase; BIOSIS; OpenGrey; ARIF and clinical trials registries. The searches were limited to remove case reports. There were no date or language restrictions in the searches. SELECTION CRITERIA: We included prospective and retrospective cross-sectional, cohort and case-control studies conducted in any setting that evaluated the accuracy of one or more index tests for identifying people with an occludable angle compared to a gonioscopic reference standard. DATA COLLECTION AND ANALYSIS: Two review authors independently performed data extraction and quality assessment using QUADAS2 for each study. For each test, 2 x 2 tables were constructed and sensitivity and specificity were calculated. When four or more studies provided data at fixed thresholds for each test, we fitted a bivariate model using the METADAS macro in SAS to calculate pooled point estimates for sensitivity and specificity. For comparisons between index tests and subgroups, we performed a likelihood ratio test comparing the model with and without the covariate. MAIN RESULTS: We included 47 studies involving 26,151 participants and analysing data from 23,440. Most studies were conducted in Asia (36, 76.6%). Twenty-seven studies assessed AS-OCT (analysing 15,580 participants), 17 studies LACD (7385 participants), nine studies Scheimpflug photography (1616 participants), six studies SPAC (5239 participants) and five studies evaluated the oblique flashlight test (998 participants). Regarding study quality, 36 of the included studies (76.6%) were judged to have a high risk of bias in at least one domain.The use of a case-control design (13 studies) or inappropriate exclusions (6 studies) raised patient selection concerns in 40.4% of studies and concerns in the index test domain in 59.6% of studies were due to lack of masking or post-hoc determination of optimal thresholds. Among studies that did not use a case-control design, 16 studies (20,599 participants) were conducted in a primary care/community setting and 18 studies (2590 participants) in secondary care settings, of which 15 investigated LACD. Summary estimates were calculated for commonly reported parameters and thresholds for each test; LACD ≤ 25% (16 studies, 7540 eyes): sensitivity 0.83 (95% confidence interval (CI) 0.74, 0.90), specificity 0.88 (95% CI 0.84, 0.92) (moderate-certainty); flashlight (grade1) (5 studies, 1188 eyes): sensitivity 0.51 (95% CI 0.25, 0.76), specificity 0.92 (95% CI 0.70, 0.98) (low-certainty); SPAC (≤ 5 and/or S or P) (4 studies, 4677 eyes): sensitivity 0.83 (95% CI 0.70, 0.91), specificity 0.78 (95% CI 0.70, 0.83) (moderate-certainty); Scheimpflug photography (central ACD) (9 studies, 1698 eyes): sensitivity 0.92 (95% CI 0.84, 0.96), specificity 0.86 (95% CI 0.76, 0.93) (moderate-certainty); AS-OCT (subjective opinion of occludability) (13 studies, 9242 eyes): sensitivity 0.85 (95% CI 0.76, 0.91); specificity 0.71 (95% CI 0.62, 0.78) (moderate-certainty). For comparisons of sensitivity and specificity between index tests we used LACD (≤ 25%) as the reference category. The flashlight test (grade 1 threshold) showed a statistically significant lower sensitivity than LACD (≤ 25%), whereas AS-OCT (subjective judgement) had a statistically significant lower specificity. There were no statistically significant differences for the other index test comparisons. A subgroup analysis was conducted for LACD (≤ 25%), comparing community (7 studies, 14.4% prevalence) vs secondary care (7 studies, 42% prevalence) settings. We found no evidence of a statistically significant difference in test performance according to setting. Performing LACD on 1000 people at risk of angle closure with a prevalence of occludable angles of 10%, LACD would miss about 17 cases out of the 100 with occludable angles and incorrectly classify 108 out of 900 without angle closure. AUTHORS' CONCLUSIONS: The finding that LACD performed as well as index tests that use sophisticated imaging technologies, confirms the potential for this test for case-detection of occludable angles in high-risk populations. However, methodological issues across studies may have led to our estimates of test accuracy being higher than would be expected in standard clinical practice. There is still a need for high-quality studies to evaluate the performance of non-invasive tests for angle assessment in both community-based and secondary care settings.

Impact of optical coherence tomography on diagnostic decision-making by UK community optometrists: a clinical vignette study.

PURPOSE: In recent years, there has been widespread investment in imaging technologies by community optometrists in the UK, most notably optical coherence tomography (OCT). The aim of the current study was to determine the value of OCT in the diagnosis of posterior segment diseases in a representative sample of community optometrists using a clinical vignette methodology. METHODS: A group of community optometrists (n = 50) initially completed a standardised training package on OCT interpretation followed by a computer-based assessment featuring 52 clinical vignettes, containing images of healthy (n = 8) or glaucomatous (n = 18) discs or healthy (n = 8) or diseased (n = 18) fundi. Each vignette featured either a single fundus/disc photographic image, or a combination of a fundus/disc image with the corresponding OCT scan. An expert panel confirmed that the fundus images presented alone and those in combination with OCT data were of a similar level of difficulty and that the cases were typical of those seen in primary care. For each case, the optometrist selected their diagnosis from a pull-down list and reported their confidence in their decision using a 10-point Likert scale. Pairwise comparisons of the fundus image alone and fundus image/OCT combination were made for both diagnostic performance and confidence. RESULTS: The mean percentage of correct diagnoses using fundus imaging alone was 62% (95% CI 59-64%) and for the combination of fundus image/OCT was 80% (95% CI 77-82%). The mean false negative rate with fundus alone was 27% reducing to 13% with the OCT combination. Median confidence scores for fundus imaging alone was 8.0 (IQR 7.0-8.0) and 8.3 (IQR 8.0-9.0) for the combination. Improvements in performance and confidence were statistically significant (p < 0.001). CONCLUSION: The results from this vignette study suggests that OCT improves optometrists' diagnostic performance compared to fundus observation alone. These initial results suggest that OCT provides valuable additional data that could augment case-finding for glaucoma and retinal disease; however, further research is needed to assess its diagnostic performance in a routine clinical practice setting.

Role of advanced technology in the detection of sight-threatening eye disease in a UK community setting.

BACKGROUND/AIMS: To determine the performance of combinations of structural and functional screening tests in detecting sight-threatening eye disease in a cohort of elderly subjects recruited from primary care. METHODS: 505 subjects aged ≥60 years underwent frequency doubling technology (FDT) perimetry, iVue optical coherence tomography (iWellness and peripapillary retinal nerve fibre layer (RNFL) scans) and intraocular pressure with the Ocular Response Analyzer, all performed by an ophthalmic technician. The reference standard was a full ophthalmic examination by an experienced clinician who was masked to the index test results. Subjects were classified as presence or absence of sight-threatening eye disease (clinically significant cataract, primary open-angle glaucoma, intermediate or advanced age-related macular degeneration and significant diabetic retinopathy). Univariate and multivariate logistic regression analyses were used to determine the association between abnormal screening test results and the presence of sight-threatening eye disease. RESULTS: 171 subjects (33.8%) had one or more sight-threatening eye diseases. The multivariate analysis found significant associations with any of the target conditions for visual acuity of <6/12, an abnormal FDT and peripapillary RNFL thickness outside the 99% normal limit. The sensitivity of this optimised screening panel was 61.3% (95% CI 53.5 to 68.7), with a specificity of 78.8% (95% CI 74.0 to 83.1), a positive predictive value of 59.5% (95% CI 53.7 to 65.2) and an overall diagnostic accuracy of 72.9% (95% CI 68.8 to 76.8). CONCLUSIONS: A subset of screening tests may provide an accurate and efficient means of population screening for significant eye disease in the elderly. This study provides useful preliminary data to inform the development of further larger, multicentre screening studies to validate this screening panel.

Development and validation of a new glaucoma screening test using temporally modulated flicker.

PURPOSE: Describing the psychometric characteristics and diagnostic accuracy of the Accelerator 4-Alternative Forced-Choice Flicker Test prototype (A4FTp) for detecting chronic open angle glaucoma (COAG). METHODS: A4FTp measures temporally-modulated flicker thresholds in regions of the visual field with high susceptibility to glaucomatous loss. We initially evaluated its psychometric properties on 20 normals (aged 33.8 ± 8.5 years) who were tested multiple times over a period of 3 months. All subjects underwent four repetitions for shorter (T8) and longer (T12) staircase termination criteria, to determine the most suitable threshold criterion. Four randomly selected subjects underwent a total of 10 repetitions to study test-retest repeatability and learning effects. To determine its diagnostic accuracy, one eye of 40 participants with COAG and 38 normal controls were tested with the A4FTp in comparison with the Frequency Doubling Technology (FDT; C20-5 programme) and iVue Spectral Domain Optical Coherence Tomography (SD-OCT). Tests were conducted in a random order with results masked to the clinician conducting the reference ophthalmic examination. The accuracy of each test was determined by analysis of the area under the receiver operator characteristic curve (AUROC). RESULTS: A4FTp flicker thresholds were stable, with standard deviations of only 0.52 decilog (dL) for T8, increasing to 1.32 dL for T12, and no significant flicker sensitivity threshold improvement over the 10 repeat runs. T8 was superior to T12 on several other measures, so it was used for the remaining comparisons. In terms of diagnostic accuracy, the mean AUROC for the three tests were A4FTp [T8 criterion; 0.82, 95% confidence interval (0.73-0.92)]; SD-OCT [any RNFL parameter p < 1% level; 0.90 (0.83-0.97)]; and FDT [one or more locations missed at p < 5% level; 0.91 (0.82-0.96)]. There was no statistical difference in AUROC between A4FTp and SD-OCT (p = 0.18) or FDT (p = 0.12). The A4FTp test duration averaged just over 2 min per eye, taking approximately one-third of the time for completion of the HFA SITA 24-2 algorithm (conducted as part of the reference examination) and twice the time for the suprathreshold FDT. CONCLUSION: Test accuracy for the A4FTp was comparable to those of the FDT and SD-OCT for the detection of COAG. Time taken to complete the A4FTp was relatively short and initial results are promising. With further refinement, the A4FTp could have a future role in glaucoma detection.

Providing Healthcare-as-a-Service Using Fuzzy Rule Based Big Data Analytics in Cloud Computing.

With advancements in information and communication technology, there is a steep increase in the remote healthcare applications in which patients can get treatment from the remote places also. The data collected about the patients by remote healthcare applications constitute big data because it varies with volume, velocity, variety, veracity, and value. To process such a large collection of heterogeneous data is one of the biggest challenges which requires a specialized approach. To address this challenge, a new fuzzy rule based classifier is presented in this paper with an aim to provide Healthcare-as-a-Service. The proposed scheme is based upon the initial cluster formation, retrieval, and processing of the big data in cloud environment. Then, a fuzzy rule based classifier is designed for efficient decision making for data classification in the proposed scheme. To perform inferencing from the collected data, membership functions are designed for fuzzification and defuzzification processes. The proposed scheme is evaluated on various evaluation metrics, such as average response time, accuracy, computation cost, classification time, and false positive ratio. The results obtained confirm the effectiveness of the proposed scheme with respect to various performance evaluation metrics in cloud computing environment.

Agreement among optometrists and ophthalmologists in estimating limbal anterior chamber depth using the van Herick method.

PURPOSE: To evaluate the inter-observer agreement for measuring limbal anterior chamber depth (LACD) using the van Herick test in community optometrists, glaucoma specialist optometrists and ophthalmologists. METHODS: The study was divided into two phases. In the first phase, a random sample of 100 UK community optometrists were given an opportunity to select and grade eight digital slit-lamp images of anterior chamber angles using the original van Herick 4 point grading scale. The images were included in a clinical decision making study using computerised virtual case vignettes. In the second phase, hospital-based glaucoma specialist optometrists and glaucoma sub-specialist ophthalmologists graded the LACD of the right eye using a 7-point % grading scale in 57 consecutively presenting patients with suspect glaucoma. Inter-observer agreement was assessed using linearly weighted kappa (κw ). RESULTS: Inter-observer agreement for community optometrists was moderate, with a mean κw for grading photographic images of 0.50 (95% confidence interval (CI) 0.43-0.57). Overall, ninety-two percent of observations were within one grade of the actual grade, although grading of narrow angles was associated with a 13% false negative error rate (based on a ≤ grade 2 threshold). For Phase 2 of the study, pairwise comparisons between optometrists and ophthalmologists showed that agreement was moderate to substantial (mean κw  = 0.54-0.65) with a false negative rate of 1.9% (based on a ≤ 25% threshold). Grading accuracy of specialist optometrists and ophthalmologists were equivalent. CONCLUSIONS: In summary, the present study found that community optometrists showed moderate inter-observer agreement for grading LACD. Glaucoma specialist optometrists showed moderate to substantial agreement with weighted kappa values that were equivalent to sub-specialist ophthalmologists. The augmented 7-point % grading scale is intuitive and potentially offers greater accuracy for grading narrow angles than the traditional 4-point scale for grading LACD.