Remote Screening Research Articles

Coswara: A respiratory sounds and symptoms dataset for remote screening of SARS-CoV-2 infection

This paper presents the Coswara dataset, a dataset containing diverse set of respiratory sounds and rich meta-data, recorded between April-2020 and February-2022 from 2635 individuals (1819 SARS-CoV-2 negative, 674 positive, and 142 recovered subjects). The respiratory sounds contained nine sound categories associated with variants of breathing, cough and speech. The rich metadata contained demographic information associated with age, gender and geographic location, as well as the health information relating to the symptoms, pre-existing respiratory ailments, comorbidity and SARS-CoV-2 test status. Our study is the first of its kind to manually annotate the audio quality of the entire dataset (amounting to 65 hours) through manual listening. The paper summarizes the data collection procedure, demographic, symptoms and audio data information. A COVID-19 classifier based on bi-directional long short-term (BLSTM) architecture, is trained and evaluated on the different population sub-groups contained in the dataset to understand the bias/fairness of the model. This enabled the analysis of the impact of gender, geographic location, date of recording, and language proficiency on the COVID-19 detection performance.

Non-Melanoma Skin Cancer Detection in the Age of Advanced Technology: A Review.

Skin cancer is the most common cancer diagnosis in the United States, with approximately one in five Americans expected to be diagnosed within their lifetime. Non-melanoma skin cancer is the most prevalent type of skin cancer, and as cases rise globally, physicians need reliable tools for early detection. Artificial intelligence has gained substantial interest as a decision support tool in medicine, particularly in image analysis, where deep learning has proven to be an effective tool. Because specialties such as dermatology rely primarily on visual diagnoses, deep learning could have many diagnostic applications, including the diagnosis of skin cancer. Furthermore, with the advancement of mobile smartphones and their increasingly powerful cameras, deep learning technology could also be utilized in remote skin cancer screening applications. Ultimately, the available data for the detection and diagnosis of skin cancer using deep learning technology are promising, revealing sensitivity and specificity that are not inferior to those of trained dermatologists. Work is still needed to increase the clinical use of AI-based tools, but based on the current data and the attitudes of patients and physicians, deep learning technology could be used effectively as a clinical decision-making tool in collaboration with physicians to improve diagnostic efficiency and accuracy.

Excitation gratings in cross-beam filament wake channels in a dense argon gas: Formation, control, and Rabi sideband manifestation.

When two intense laser beams cross at a small angle, the interference in the crossing area results in a finite intensity grating. We consider femtosecond laser filamentation in such a grating, in a situation when the process is largely confined to the grating maxima and leads to formation of a structured filament wake channel. In a dense gas, electron impact processes during the laser pulse cause a copious excitation of neutral atoms, resulting in formation of a finite grating of the density of excited atoms. Numerically solving the equations of laser-driven kinetics, we obtain the properties of this grating, as depending on the characteristics of the interfering beams and especially on the interbeam phase delay. The excitation gratings thus formed give rise to a hallmark effect of Rabi sideband emission when probed by a picosecond 800 nm laser pulse, which couples with transitions in the excited states manifold. Spectral and spatial interference of the emitted radiation forms four-dimensional spatial-spectral fringe patterns accessible for observation on a remote screen. The patterns are indicative of the excitation grating structure; their sensitivity to the phase delay between the crossing pump pulses warrants experimental verification.

Using Machine Learning Algorithms to Determine the Post-COVID State of a Person by Their Rhythmogram.

This study introduces a novel method for detecting the post-COVID state using ECG data. By leveraging a convolutional neural network, we identify "cardiospikes" present in the ECG data of individuals who have experienced a COVID-19 infection. With a test sample, we achieve an 87 percent accuracy in detecting these cardiospikes. Importantly, our research demonstrates that these observed cardiospikes are not artifacts of hardware-software signal distortions, but rather possess an inherent nature, indicating their potential as markers for COVID-specific modes of heart rhythm regulation. Additionally, we conduct blood parameter measurements on recovered COVID-19 patients and construct corresponding profiles. These findings contribute to the field of remote screening using mobile devices and heart rate telemetry for diagnosing and monitoring COVID-19.

Tele-Global Examination of Mental State (Tele-GEMS): an open tool for the remote neuropsychological screening.

Tele-neuropsychology, i.e., the application of remote audio-visual technologies to neuropsychological evaluation or rehabilitation, has become increasingly popular and widespread during and after the COVID-19 pandemic. New tools with updated normative data and appropriate methodological developments are necessary. We present Tele-GEMS, a telephone-based cognitive screening developed on N = 601 Italian participants. It yields a global score tapping on orientation, memory, spatial representation, language, and pragmatic abilities. Its administration lasts about 10 min. Clinical cut-offs are provided, accounting for demographic variables (age, education, and sex) and also for a comprehensive index taking into account cognitively stimulating life experiences that canbuild up a cognitive reserve. Tele-GEMS shows good internal consistency and a good inter-rater agreement. The test includes the thresholds for estimating a significant change after repeated measurements. Tele-GEMS has a good construct validity as assessed with MoCA and a suitable criterion validity assessed with its in-person version (GEMS). All the materials andthe instructions, including scripts and an online Application for the automatic calculation of cut-offs, are accessible on OSF at https://osf.io/t3bma/ under a Creative Commons license.

Active monitoring for atrial fibrillation (AMALFI): a streamlined randomised controlled trial of remote screening for subclinical atrial fibrillation

Abstract Funding Acknowledgements Type of funding sources: Public Institution(s). Main funding source(s): National Institute for Health and Care Research Oxford Biomedical Research Centre ECG patches (and respective monitoring data analysis) were granted free of charge by the manufacturer (iRhythm) Background Incidence of stroke could be reduced by detection of subclinical atrial fibrillation (AF) and appropriate anticoagulation. Opportunistic screening by pulse-taking or ECG is recommended by some international guidelines. In contrast, there is uncertainty regarding the benefits and harms of systematic screening across populations and/or using longer monitoring durations. Purpose To assess if screening using a one-off single-lead continuous non-invasive ECG patch worn for 14 days increases AF detection at 2.5 years compared to standard care, and whether this approach is cost-effective. Methods AMALFI is an ongoing mail-based open-label randomised controlled trial of remote AF screening in England. Streamlined design includes integration with national electronic health data and care pathways for participant identification, invitation, and management. The target population are individuals aged ≥65 years with CHA2DS2VASc score of ≥3 in men and ≥4 in women, and with no prior history of AF or allergy to latex. Participants in the intervention arm self-apply the ECG patch and return it for analysis after 14 days. AF findings (duration ≥30 seconds at any rate) are communicated to primary care physicians, with further clinical management at their discretion. Participants in the control arm receive usual care (which is expected to include opportunistic AF screening). The primary outcome is the proportion of participants with newly-detected AF recorded in primary care electronic health records in each group at 2.5 years after randomisation. Results From 2019 to 2022, AMALFI invited 22,188 people, of whom 5,040 (23%) were randomised. Baseline characteristics included age 77±6 years, 47% female, and median CHA2SD2VASc score 3 (Table 1). Of 2520 individuals randomised to the intervention, 2127 (84%) wore and returned the patch. Median wear time was 13.9 days and median proportion of time analysable was 98.8%. AF was present in 102 patch reports, representing 4.8% of reports received and 4% of the intervention arm (in line with a priori estimates for power calculations for the primary outcome); median AF burden was 11.3%. AF was present at commencement of recording in 40% of these reports, by the end of the first day in 58%, and by the end of the first week in 83% (Figure 1). Incidental findings included supraventricular tachycardia (in 90% of reports), non-sustained ventricular tachycardia (VT; in 32%), sustained VT (in 0.1%), Mobitz II/high-grade/3rd degree AV block (in 1.1%), and pauses >6s (in 0.3%). Conclusion AMALFI’s innovative design facilitated recruitment and demonstrates that a remote ECG patch-based AF screening strategy is feasible, resulting in a new diagnosis of AF in ~4% of individuals selected due to moderate-to-high clinical stroke risk. Longer term follow-up will provide data on the incremental value and cost-effectiveness of AF diagnosis by systematic screening compared to routine opportunistic detection.

Telemedicine remote multidisciplinary questionnaire screening as a tool for early detection of chronic non-communicable diseases

Chronic non-communicable diseases (CNID) are an important problem for health care worldwide. In this connection, there was a need to develop preventive measures aimed at their early detection and timely correction. Taking into account the available evidence base on the effectiveness of preventive measures to reduce risk factors of CNID, special attention should be paid to people with high risks of their development and patients with an already established diagnosis. To improve the quality of preventive medicine, it is necessary to consistently develop and implement modern technologies aimed at early detection of risk factors at all stages of human life using available tests. To date, remote screening is considered as one of the promising areas of telemedicine technologies, which can be successfully used during medical examinations, medical examinations, etc. We have developed a methodology for telemedicine multidisciplinary remote questionnaire screening of the adult population using a system of analysis of unstructured data for population monitoring of CNID. Our telemedicine program was tested based on the city polyclinic for the first stage of medical examination. Depending on the identified risk factors, it was found that patients with the greatest frequency are concerned about complaints from the endocrine, digestive, respiratory, cardiovascular systems and oncological alertness. The interim results of the study once again show that the transition to new social conditions causes an active destabilization of the physical and psycho-emotional capabilities of the human body, which, under excessive loads, are able to increase the degree of activity of the CNID risk factors.

Mobile Assessments to Improve Screening and Novel Patient Engagement to Diagnose and Manage Maternal Mental Health (MIND) Study Midpoint Analysis: Compliance Rate of Antenatal Screening [ID: 1363058

INTRODUCTION: One in five to seven birthing people are affected by perinatal mood disorders. While screening for gestational diabetes is well defined, the efficacy of standardized assessment of mood disorders is less understood. Between 2018 and 2020, screening at an academic center was highly variable, demonstrating need for improvement. Remote surveillance using novel integrated technology was used to improve patient participation in routine antenatal screening. METHODS: A prospective observational cohort of 81 patients from two clinical sites in the Babyscripts (BRx) Maternal Mental Health Module received digital Edinburgh Postnatal Depression Scale (EPDS) screening at 12 and 28 weeks of gestation. The control group consisted of 81 randomly selected patients who received in-person standardized screening from 2018 to 2020. In a midpoint analysis, EPDS completion was assessed at weeks 12 and 28 in the intervention group and compared to the control group using χ2 analysis. The study was reviewed and exempted by the George Washington University IRB. RESULTS: The BRx treatment group totaled 45 patients enrolled prior to week 14 and 74 patients enrolled prior to week 30. At week 14, 38% of BRx patients (17/45) had completed an EPDS survey, compared with 10% of patients in the control group (8/81). At week 30, 36% of BRx patients (27/74) had completed an EPDS survey, compared with 21% of control (17/81). Improved EPDS response rates of the BRx cohort were statistically significant at both assessments (P<.01 and P=.03). CONCLUSION: Remote mood screening resulted in significantly higher rates of adherence in the first and second trimesters as compared to standard methods. In the next phase of analysis, we plan to assess adherence of screening through 12 months postpartum. Future analysis is planned to determine noninferiority as compared to in-office methods among a larger cohort. We hypothesize that novel telehealth mood screenings will help to sustain engagement with providers and reduce time to intervention.

Symptoms Predicting SARS-CoV-2 Test Results in Resident Physicians and Fellows in New York City

Accurate prediction of SARS-CoV-2 infection based on symptoms can be a cost-efficient tool for remote screening in healthcare settings with limited SARS-CoV-2 testing capacity. We used a machine learning approach to determine self-reported symptoms that best predict a positive SARS-CoV-2 test result in physician trainees from a large healthcare system in New York. We used survey data on symptoms history and SARS-CoV-2 testing results collected retrospectively from 328 physician trainees in the Mount Sinai Health System, over the period 1 February 2020 to 31 July 2020. Prospective data on symptoms reported prior to SARS-CoV-2 test results were available from the employee health service COVID-19 registry for 186 trainees and analyzed to confirm absence of recall bias. We estimated the associations between symptoms and IgG antibody and/or reverse transcriptase polymerase chain reaction test results using Bayesian generalized linear mixed effect regression models adjusted for confounders. We identified symptoms predicting a positive SARS-CoV-2 test result using extreme gradient boosting (XGBoost). Cough, chills, fever, fatigue, myalgia, headache, shortness of breath, diarrhea, nausea/vomiting, loss of smell, loss of taste, malaise and runny nose were associated with a positive SARS-CoV-2 test result. Loss of taste, myalgia, loss of smell, cough and fever were identified as key predictors for a positive SARS-CoV-2 test result in the XGBoost model. Inclusion of sociodemographic and occupational risk factors in the model improved prediction only slightly (from AUC = 0.822 to AUC = 0.838). Loss of taste, myalgia, loss of smell, cough and fever are key predictors for symptom-based screening of SARS-CoV-2 infection in healthcare settings with remote screening and/or limited testing capacity.

Decentralized approaches in TRAILBLAZER-ALZ 3 (P7-6.006)

<h3>Objective:</h3> To provide an overview of key decentralized design characteristics utilized in the TRAILBLAZER-ALZ 3 (TB3) (NCT05026866) study, an ongoing Phase 3 research trial with a decentralized design, testing donanemab in preclinical Alzheimer’s disease (AD). <h3>Background:</h3> Decentralized trials (DCTs) offer flexibility typically limited in traditional trial design, which may increase geographical and ethnic/racial diversity in trial populations, improve participant engagement and retention, and reduce trial cost. Clinical trials focused on AD pose specialized challenges to remotely assessing cognitive outcomes. <h3>Design/Methods:</h3> TRAILBLAZER-ALZ 3 is a multicenter, randomized, double-blind, placebo-controlled Phase 3 trial designed to assess the impact of donanemab versus placebo in cognitively unimpaired participants with evidence of AD pathology. <h3>Results:</h3> The remote screening process, which includes mobile research units and health fairs, uses the modified Telephone Interview for Cognitive status (TICS-m) to help select for cognitively unimpaired individuals. Plasma AD assays assess an AD biomarker as part of the inclusion screening criteria and minimize participant burden. Optional remote genetic counseling for APOE disclosure is offered as well as two optional sub-studies testing amyloid PET and tau PET. Clinical outcomes are assessed throughout the study using central raters who administer the Clinical Dementia Rating scale (CDR) to study partners and participants, and psychometric examinations to study participants. Self-administered tests are proctored remotely by a centralized study coordinator who also helps coordinate and facilitate all remote appointments for the participant and study partner. Infusion and imaging centers outside of traditional study “sites” are available to improve participant convenience and access throughout the study. <h3>Conclusions:</h3> TRAILBLAZER-ALZ 3 represents an innovative design that could potentially pave the way for future AD clinical studies by utilizing decentralized trial design with central raters, time-to-clinical-event model, blood-based AD biomarker selection criterion, and limited dosing of anti-amyloid antibody. <b>Disclosure:</b> Dr. Yaari has received personal compensation for serving as an employee of Eli Lilly and Company. Dr. Yaari has stock in Eli Lilly and Company. Karen C. Holdridge, 19157 has received personal compensation for serving as an employee of Eli Lilly and Company. Karen C. Holdridge, 19157 has stock in Eli Lilly and Company. Margaret Ferguson has received personal compensation for serving as an employee of Eli Lilly. Margaret Ferguson has stock in Eli Lilly. Dr. Wessels has received personal compensation for serving as an employee of Eli Lilly and Company. Dr. Wessels has received stock or an ownership interest from Eli Lilly and Company. Dr. Sims has received personal compensation for serving as an employee of Eli Lilly and Company. Dr. Sims has stock in Eli Lilly and Company.

Abstract 3334: Epigenetic immune cell markers for CLL

Abstract CLL is a rare hematological cancer and is classified as low-malignancy non-Hodgkin lymphoma. While rare, it is the most frequent form of leukemia in Europe and US accounting for approximately a quarter of all leukemias. As it is a slowly growing tumor, its 5-year survival rate is high compared to faster developing B cell tumors and survival has further improved since introduction of Anti-CD20 Antibody therapies. However, overall the 5-year survival of rare cancers (49%) is significantly lower than for their common counterparts (63%), even though hematological cancers improved outcome over the last years. Due to the currently very late and often incidental diagnosis, some cases turn into aggressively progressing cancers prior to treatment and with concomitant poor outcome. Therefore, diagnosis of CLL and other B cell leukemias remains a challenge. To further improve outcome of CLL and other lymphatic leukemias earlier diagnosis is important. To this end, simple home testing could broaden access to diagnostic tools, but no current method would provide adequate remote initial screening of lymphatic leukemia in populations at risk, such as persons older than 65 years. Epigenetic qPCR has been shown to detect and quantify immune cell populations efficiently from liquid and dried blood samples (DBS). This system is based on specifically unmethylated regulatory elements in gene loci associated with certain cell types. These unique methylation patterns allow specific quantification of cells of interest via qPCR. Different epigenetic qPCR systems were developed for total B cells, naïve and memory B cells as well as IgM producing B cells. While these gene loci are fully unmethylated in the respective target cells, they are fully methylated in all relevant control cell types. The cell type specificity of these markers was shown using bisulfite specific sequencing using purified cells from healthy donors. Here, we investigated blood samples from healthy donors (n = 112) as well as patients at the time of their initial CLL diagnosis. The mean total B cell count of healthy donors at 5.7% was clearly lower than the 60.4% obtained for CLL patients. Similarly drastic differences were observed for naïve B (2.3% normal vs. 20.8% CLL) and IgM (3.3% vs. 49.5%) as well as memory B cells (1.6% vs. 32.6%). All differences were statistically significant. To investigate if these mean differences may also lead to differential (and potentially early) diagnosis, we designed receiver operator characteristics and contingency tables, both indicating near perfect segregation between CLL patients and healthy donors in Ig M, memory and all B cells. Our data indicate applicability of epigenetic immune cell counting as easy, potentially DBS- and home-based approach to test increase of immune cells in people potentially at risk of hematological cancers. Citation Format: Janine Jung, Jeannette Werner, Konstantin Schildknecht, Janika Schulze, Steffi Walter, Christoph Sachsenmaier, Deborah Phippard, Barbara Seliger, Claudia Wickenhauser, Sven Olek. Epigenetic immune cell markers for CLL [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3334.

Feasibility of Remote Unsupervised Cognitive Screening With SATURN in Older Adults.

Widespread cognitive test screening as part of tele-public health initiatives necessitates a test that is self-administered online and automatically scored, with no clinician effort. The feasibility of unsupervised cognitive screening is unclear. We adapted the Self-Administered Tasks Uncovering Risk of Neurodegeneration (SATURN) to make it suitable for self-administration and automatic scoring. A sample of 364 healthy older adults completed SATURN via a web browser, in a fully independent manner. SATURN's overall score was not modulated by gender, education, reading speed, the time of day at which the test was taken, or an individual's familiarity with technology. SATURN proved extremely portable across operating systems. Importantly, comments from participants reported satisfaction with the experience and the clarity of the instructions. SATURN represents a fast and easy screening tool that can be used for a first assessment, during a routine test or clinical evaluation, or during periodic health monitoring, in person or remotely.

Comparison of the telephone-Montreal Cognitive Assessment (T-MoCA) and Telephone Interview for Cognitive Status (TICS) as screening tests for early Alzheimer's disease.

Remote screening for cognitive impairment associated with Alzheimer's disease (AD) has grown in importance with the expected rise in prevalence of AD in an aging population and with new potential treatment options. The Telephone Interview for Cognitive Status (TICS) and new telephone adaptation of the Montreal Cognitive Assessment (T-MoCA) were administered to participants independently classified through in-person clinical evaluation as cognitively normal (CN; n = 167), mild cognitive impairment (MCI; n = 25), or dementia (n = 23). Cerebrospinal fluid AD biomarkers were measured (n = 79). TICS and T-MoCA were highly correlated (r = 0.787; P <0.001): groups differed on both (CN<MCI<dementia; TICS: F [2212] = 156.66; P <0.001; T-MoCA: F [2210] = 143.72; P <0.001), both effectively detected cognitive impairment (receiver operating characteristic area under the curve: TICS = 0.889; T-MoCA = 0.902), and both negatively correlated with a composite AD biomarker (tau/amyloid beta 1-42; TICS: r = -0.372; P = 0.001; T-MoCA: r = -0.480; P <0.001). TICS and T-MoCA are effective for remotely detecting cognitive impairment associated with AD in older adults. Strong correlation between tests provides construct validity for the newer T-MoCA. Construct validity for the telephone adaptation of the Montreal Cognitive Assessment (T-MoCA) was newly established against the Telephone Interview for Cognitive Status (TICS). TICS and T-MoCA effectively detected cognitive impairment with remote administration. Both tests negatively correlated with a composite cerebrospinal fluid Alzheimer's disease (AD) biomarker (tau/amyloid beta 1-42).

Diagnostic Ability of a Smartphone App for Dry Eye Disease: Protocol for a Multicenter, Open-Label, Prospective, and Cross-sectional Study.

Dry eye disease (DED) is one of the most common ocular surface diseases. Numerous patients with DED remain undiagnosed and inadequately treated, experiencing various subjective symptoms and a decrease in quality of life and work productivity. A mobile health smartphone app, namely, the DEA01, has been developed as a noninvasive, noncontact, and remote screening device, in the context of an ongoing paradigm shift in the health care system, to facilitate a diagnosis of DED. This study aimed to evaluate the capabilities of the DEA01 smartphone app to facilitate a DED diagnosis. In this multicenter, open-label, prospective, and cross-sectional study, the test method will involve using the DEA01 smartphone app to collect and evaluate DED symptoms, based on the Japanese version of the Ocular Surface Disease Index (J-OSDI), and to measure the maximum blink interval (MBI). The standard method will then involve a paper-based J-OSDI evaluation of subjective symptoms of DED and tear film breakup time (TFBUT) measurement in an in-person encounter. We will allocate 220 patients to DED and non-DED groups, based on the standard method. The primary outcome will be the sensitivity and specificity of the DED diagnosis according to the test method. Secondary outcomes will be the validity and reliability of the test method. The concordance rate, positive and negative predictive values, and the likelihood ratio between the test and standard methods will be assessed. The area under the curve of the test method will be evaluated using a receiver operating characteristic curve. The internal consistency of the app-based J-OSDI and the correlation between the app-based J-OSDI and paper-based J-OSDI will be assessed. A DED diagnosis cutoff value for the app-based MBI will be determined using a receiver operating characteristic curve. The app-based MBI will be assessed to determine a correlation between a slit lamp-based MBI and TFBUT. Adverse events and DEA01 failure data will be collected. Operability and usability will be assessed using a 5-point Likert scale questionnaire. Patient enrollment will start in February 2023 and end in July 2023. The findings will be analyzed in August 2023, and the results will be reported from March 2024 onward. This study may have implications in identifying a noninvasive, noncontact route to facilitate a diagnosis of DED. The DEA01 may enable a comprehensive diagnostic evaluation within a telemedicine setting and facilitate early intervention for undiagnosed patients with DED confronting health care access barriers. Japan Registry of Clinical Trials jRCTs032220524; https://jrct.niph.go.jp/latest-detail/jRCTs032220524. PRR1-10.2196/45218.

A proactive learning approach toward building adaptive capacity during COVID-19: A radiology case study

The COVID-19 pandemic has challenged organizations to adapt under uncertainty and time pressure, with no pre-existing protocols or guidelines available. For organizations to learn to adapt effectively, there is a need to understand the perspectives of the frontline workforce involved in everyday operations. This study implemented a survey-tool to elicit narratives of successful adaptation based on the lived experiences frontline radiology staff at a large multispecialty pediatric hospital. Fifty-eight members of the radiology frontline staff responded to the tool between July and October of 2020. Qualitative analysis of the free-text data revealed five categories of themes that underpinned adaptive capacity of the radiology department during the pandemic: information flow, attitudes and initiative, new and adjusted workflows, availability and utilization of resources, and collaboration and teamwork. Enablers of adaptive capacity included timely and clear communication about procedures and policies from the leadership to frontline staff, and revised workflows with flexible work arrangements, such as remote patient screening. Responses to multiple choice questions in the tool helped identify the main categories of challenges faced by staff, factors that enabled successful adaptation, and resources used. The study demonstrates the use of a survey-tool to proactively identify frontline adaptations. The paper also reports a system-wide intervention resulting directly from a discovery enabled by the findings based on the use of RETIPS in the radiology department. In general, the tool could be used in concert with existing learning mechanisms, such as safety event reporting systems, to inform leadership-level decisions to support adaptive capacity.

Patients' Willingness to Accept Social Needs Navigation After In-Person versus Remote Screening.

Social needs screening and referral interventions are increasingly common in health care settings. Although remote screening offers a potentially more practical alternative to traditional in-person screening, there is concern that screening patients remotely could adversely affect patient engagement, including interest in accepting social needs navigation. We conducted a cross-sectional study using a multivariable logistic regression analysis and data from the Accountable Health Communities (AHC) model in Oregon. Participants were Medicare and Medicaid beneficiaries in the AHC model from October 2018 through December 2020. The outcome variable was patients' willingness to accept social needs navigation assistance. We included an interaction term (total number of social needs + screening mode) to test whether in-person versus remote screening was an effect modifier. The study included participants who screened positive for ≥1 social need(s); 43% were screened in person and 57% remotely. Overall, 71% of participants were willing to accept help with social needs. Neither screening mode nor interaction term were significantly associated with willingness to accept navigation assistance. Among patients presenting with similar numbers of social needs, results indicate that type of screening mode may not adversely affect patients' willingness to accept health care-based navigation for social needs.

BioscoreNet: Traumatic Brain Injury (TBI) detection using a multimodal self-attention fusion neural network and a passive bioscore monitoring framework from smartphone sensor data

Traumatic Brain Injury (TBI) affects millions of individuals globally and can cause motor, cognitive and emotional deficits. Emerging research in the field of mobile health has demonstrated that smartphone sensor data can be used to monitor various chronic diseases, which can facilitate early ailment detection and continuous monitoring of recovering patients. The goal of the paper’s contributions is two-fold: Proposing (1) A Deep Learning framework that uses smartphone sensor data mobility data (accelerometer, magnetometer, gyroscope, pedometer, pressure, altitude, accessibility) to detect whether a subject has TBI after a head injury. (2) The concept of a TBI Bioscore, a number that quantifies the certainty (0–1) that a subject has TBI. The TBI deep learning detection model uses a self-attention mechanism for multimodal feature fusion and a stacked LSTM for prediction. This model achieves a Balanced Accuracy (BA) of 90.2% and a True Positive Rate (TPR) of 83.3% in correctly identifying TBI instances. The Bioscore is generated using Monte-Carlo Dropout uncertainty estimation. When a users’ TBI Bioscores are visualized throughout, it steadily decreases for users who have normal recoveries. The proposed TBI detection and Bioscore framework facilitates population-level passive, continuous and remote TBI screening and monitoring. A glanceable display of color-coded Bioscores can enable a small team of medical staff to monitor a large pool of patients and intervene preemptively. • This paper proposed a deep learning framework that uses smartphone sensor data (accelerometer, magnetometer, gyroscope, pedometer, pressure, altitude, accessibility) to detect whether a subject has Traumatic Brain Injury (TBI) after a head injury. • This paper also proposed the concept of a TBI Bioscore, a number that quantifies the certainty (0–1) that a subject has TBI, generated using Monte-Carlo Dropout uncertainty estimation that facilitates populationlevel monitoring. • The TBI deep learning detection model uses a self-attention mechanism for multimodal feature fusion and a stacked LSTM for prediction. • The TBI detection model achieves a Balanced Accuracy (BA) of 90.2% and a True Positive Rate (TPR) of 83.3% in correctly identifying TBI instances.

The feasibility of a digital health approach to facilitate remote dental screening among preschool children during COVID-19 and social restrictions.

Tele-dentistry can be useful to facilitate screening of children, especially those living in rural and remote communities, and during the COVID-19 pandemic. This study evaluated the feasibility of tele-dental screening for the identification of early childhood caries (ECC) in preschoolers using an app operated by their parents with remote review by oral-health therapists. This cross-sectional study was a sub-project nested in ORIGINS Project, a longitudinal birth cohort study in Western Australia. Initially, children were visually examined by a paediatric dentist (gold standard). Subsequently, dental photographs were taken by parents using a smartphone camera. Two trained oral health professionals asynchronously evaluated dental photographs. The presence of dental caries was recorded as per the International Caries Detection and Assessment System-II classification. The diagnostic accuracy and reliability of the tele-dental screening and the gold standard dental examinations were then compared. Forty-two children aged <4 years were enrolled in the study. Twenty-five per cent of examined children had dental caries (mean dmfs=0.7). A total of 370 dental photographs were obtained. Parents were able to take good-quality photographs, with 90% of photographs rated as good to fair quality. Tele-dental screening demonstrated high specificity (>=95.5%) for both reviewers compared to the gold standard dental examination. However, the sensitivity scores for the two reviewers varied, ranging from 44% to 88.4%. Tele-dental screening for ECC was shown to be a feasible approach following a brief training for primary caregivers. This approach can offer a potential low-cost and sustainable alternative for visual dental examinations for young children, particularly in times of COVID-19-related restrictions.

Augmented reality for minimally invasive spinal surgery.

Augmented reality (AR) is an emerging technology that can overlay computer graphics onto the real world and enhance visual feedback from information systems. Within the past several decades, innovations related to AR have been integrated into our daily lives; however, its application in medicine, specifically in minimally invasive spine surgery (MISS), may be most important to understand. AR navigation provides auditory and haptic feedback, which can further enhance surgeons' capabilities and improve safety. The purpose of this article is to address previous and current applications of AR, AR in MISS, limitations of today's technology, and future areas of innovation. A literature review related to applications of AR technology in previous and current generations was conducted. AR systems have been implemented for treatments related to spinal surgeries in recent years, and AR may be an alternative to current approaches such as traditional navigation, robotically assisted navigation, fluoroscopic guidance, and free hand. As AR is capable of projecting patient anatomy directly on the surgical field, it can eliminate concern for surgeon attention shift from the surgical field to navigated remote screens, line-of-sight interruption, and cumulative radiation exposure as the demand for MISS increases. AR is a novel technology that can improve spinal surgery, and limitations will likely have a great impact on future technology.

Remote Screening for Developmental Language Disorder in Bilingual Children: Preliminary Validation in Spanish–Italian Speaking Preschool Children

Due to the difficulties in differentiating bilingual children with Developmental Language Disorder (DLD) from bilingual children with temporary language difficulties that may be caused by heterogeneous language input, language assessments of bilingual children are challenging for clinicians. Research demonstrates that assessments of bilingual children should be in all the languages a bilingual child speaks. This can be arduous for clinicians, but computerised screening approaches provide potential solutions. MuLiMi is a new web-based platform designed to automatise screening procedures for bilingual children at risk of DLD. To validate this procedure and investigate its reliability, 36 Spanish-speaking children, aged 4–6 years old, living in Italy, were tested remotely using the Italian–Spanish MuLiMi DLD screening. Sixteen of the participants were previously diagnosed with DLD. L2 (second or societal language) as well as L1 (first or family language) language abilities in static (nonword repetition, grammaticality judgement, and verb comprehension) as well as dynamic tasks (dynamic novel word learning) were assessed. Speed and accuracy of the children’s responses were automatically recorded (except nonword repetition). Significant associations emerged between the results obtained in the screening tasks when comparing them to parental questionnaires and standardised tests. An exploratory analysis of the diagnostic accuracy indicates that the single screening scores as well as the overall total score significantly contribute to DLD (risk) identification.