Deep learning-based voice screening technique for cystic fibrosis related diabetes

Background: Cystic fibrosis-related diabetes (CFRD) is a unique type of diabetes that is associated with significantly increased morbidity and mortality in both children and adults with cystic fibrosis (CF). The prevalence of CFRD progressively increases with age such that more than half of adults with CF develop CFRD. Early diagnosis and treatment are associated with improvements in body weight and pulmonary function, reduction in the frequency of pulmonary exacerbations, and improved overall survival. Since patients with CFRD may present with no symptoms, screening is recommended starting from the age of 10 years with an annual oral glucose tolerance test (OGTT). Especially during the COVID-19 pandemic, requiring an in-person clinic visit can be challenging, which may lead to a delayed diagnosis of CFRD. Objectives: The purpose of this project was to develop a state-of-the-art technique to detect changes in glucose levels of patients with CF by developing a deep learning-based audio classification tool. Preliminary work by our group suggested that voice characteristics could distinguish between patients with CFRD patients and patients with CF but without CFRD. We hypothesize that high blood glucose levels may cause laryngeal soft tissue swelling leading to changes in voice characteristics. Methods: We performed a prospective cross-sectional study in adult patients with CF recruited from Emory CF Clinic from March to December 2021. We recorded 5-second voice samples of a sustained /a/ vowel via a portable digital microphone. The spectrogram was extracted via the Mel frequency cepstral coefficient. The training to test the dataset ratio was 80: 20. 20% of the training dataset were randomly selected to serve as a validation dataset. We designed a convolutional neural network (CNN) architecture for CFRD patients’ voice classification. Results: There were a total of 100 subjects consisting of 43 patients with CFRD and 57 patients with CF without diabetes. The male to female ratio was approximately 60: 40 in both groups. Patients with CFRD had similar mean age and mean BMI to patients without CFRD. There was a significantly higher point of care glucose level in CFRD patients. The mean duration of a CFRD diagnosis was 9 years and the mean HbA1c level was 7.26 in the CFRD group. The performance of the VGG model CNN classifier achieved 98.7% and 94.92% accuracy on training and validation datasets, respectively. On the test dataset, the model achieved 73.53% sensitivity 69.77% specificity and 71.43% accuracy. Conclusions: We found a deep learning-based audio screening tool for CFRD could be potentially used as an alternative tool for screening in the CF community. A convolutional neural network algorithm demonstrated high sensitivity and specificity to adequately differentiate between patients with and without CFRD. Larger prospective studies are required to test this technology in patients with every form of diabetes.Presentation: Sunday, June 12, 2022 12:15 p.m. - 12:30 p.m.

Implementation of Use of Monoclonal Antibody Therapy in a Large Academic Center for the Outpatient Treatment of Covid-19: Impact on 30 Day Hospitalization Rates, ED Visits and Death

Background. Monoclonal Antibody Therapy (MAbs) has been shown to reduce rates of ED visits and hospitalizations in patients at risk for severe Covid-19 infection in clinical trials. Since November, three Mabs received emergency use authorization: Bamlanivimab (Bam), Bamlanivimab/Etesevimab (Bam/Ete) and Casirivimab/ Imdevimab (Casi/imdevi). We describe here the real-world effectiveness of implementing early MAb therapy in the outpatient setting for individuals with Covid-19 at high risk of progression. Methods. We examined the records of 808 UCLA Health patients with a confirmed positive SARS-CoV2 PCR test who were either referred for outpatient Mab therapy or received Mab treatment in the emergency department (ED) between December 10, 2020, and May 3, 2021. The primary outcome of our analysis was the combined 30-day incidence of emergency department visits, hospitalizations, or death following the date of referral. SARS-CoV2 isolates of hospitalized patients who had received Mabs were sequenced to determine the presence of variants. Results. Of 808 patients, 383 were referred for treatment but did not receive treatment, 109 received Mabs in the ED and 316 patients were treated in an outpatient setting. Composite 30-day mortality, ED visits and hospital admissions were significantly reduced in the combination therapy group (Bam/Ete or Cas/Imd) compared with monotherapy (Bam alone) or no treatment groups (aHR 0.16, 95% CI .038, .67). Significant factors associated with the composite outcome included: history of lung disease (HR 4.46, 95% CI 2.89-6.90), cardiovascular disease (HR 1.87, 95% CI 1.12-3.12), kidney disease (HR 2.04, 95% CI 1.27-3.25), and immunocompromised state (HR 3.24, 95% CI 1.02-10.26) as well as high social vulnerability index (HR 1.87, 95% CI 1.13-3.10). Over one-third of hospitalized patients who had received Mabs were confirmed to have the California variant (B.1.427/29) (Figure 1). Figure 1. Covid-19 MAB Treatment Failure Lineages Conclusion. Our data show that in a real-world setting, combination monoclonal antibody therapy, not monotherapy, significantly reduced ED visits and hospital admissions, likely due to the presence of the California variants. High socioeconomic vulnerability and certain medical conditions increased risk of treatment failure.

Detection of cystic fibrosis related diabetes through analysis of voice characteristics in telemedicine clinics

Purpose of Study Cystic fibrosis-related diabetes (CFRD) is one of many extrapulmonary co-morbidities associated with cystic fibrosis (CF), affecting an estimated 50% of all adults with the condition [1,2]. The standard test recommended by the CF Foundation for screening of CFRD is the oral glucose tolerance test (OGTT) [5]. Requiring a complex in-person clinic visit, the OGTT can be cumbersome to schedule, and for these reasons, may lead to a delayed diagnosis of CFRD. We are interested in developing a novel technique to detect changes in glucose levels by analyzing characteristics of the voice, specifically a method to capture voice recordings that requires no face-to-face interaction as restricted by the ongoing COVID-19 Pandemic. We hypothesized that high blood glucose levels may cause laryngeal soft tissue swelling and lead to changes in voice characteristics. The purpose of this study is to examine if changes in voice can distinguish patients with CFRD from patients without CFRD. Methods Used A prospective cross-sectional study was performed in adult CF patients recruited from the CF Telemedicine Clinic at Emory Healthcare from March to September 2021. We recorded 5-second voice samples of a sustained/a/ vowel via an electronic submission form to allow patients to submit recordings directly from a link sent to their smartphone. Voice parameters listed in Table 1 were analyzed using a Computerized Speech Lab with the Multi-Dimensional Voice Program. Summary of Results 5 patients with CFRD and 9 patients with CF alone were included in this study. Patients with CFRD had a similar mean age to patients without CFRD (39 ± 15 vs 33 ± 12 years old, p=0.948). Male CFRD patients were excluded due to low sample size. An acoustic parameter analysis categorized by sex showed vF0 in female patients who have CFRD was significantly higher compared with female patients with CF alone. FTRI was also significantly higher in CFRD individuals. Conclusions With a significant voice character distinction between CFRD and CF alone patients, we present a novel screening tool for diabetes that may have potential use in the CF community. This study is ongoing and will collect additional data on males with CFRD. The results indicate great potential for this technology to be used as a noninvasive test for earlier detection of undiagnosed CFRD. Studies in patients with diabetes without CF have also demonstrated a potential use of this technology [7].

Nurses Are More Exhausted Than Ever: What Should We Do About It? A review of the evidence leads to timely recommendations

For nurses, the challenges posed by demanding work environments and schedules often lead to fatigue, and this can be exacerbated during crises like the COVID-19 pandemic. In this article, the authors discuss causes and challenges of nurse fatigue and consider several evidence-based strategies and solutions for individual nurses and organizations. Barriers to implementation, including a negative workplace culture and inadequate staffing, are also described, and several resources are presented.

Telehealth in pain medicine: Lessons learned during the COVID-19 pandemic

Purpose Three selected cases of Pain Medicine clinical encounters are presented to illustrate three lessons learned during the conversion of face to face care to telehealth care (via video and telephone) mandated during the COVID-19 pandemic emergency situation. These lessons are described in three general categories. Those are: 1) care processes, 2) clinical impact, and 3) administrative to include costs. Lessons learned are important for improving clinical outcomes, conveying new information to clinicians planning to implement telehealth, conveying new information to clinicians utilizing telehealth, conveying new information to administrators promoting the use of telehealth and payers considering reimbursement for telehealth services. Care processes involve patient, staff, clinician and administrative perspectives. Methods Experience implementing telehealth services in three different health care systems in the state of Utah are the source of material to be described. Since the onset of the COVID-19 pandemic in the United States, different health systems have had different approaches to the implementation and acceleration of telehealth services. This senior Pain Medicine clinician has been practicing in these health systems for over twenty years and has made observations of the different approaches to implementing telehealth particularly since the COVID-19 pandemic hit the state of Utah. Three cases are chosen to illustrate the lessons learned from this experience. She has engaged with patients, other clinicians, and other administrators to share their observations of the impact of telehealth services on pain medicine care. Results There are differences in the outcomes of telehealth pain medicine services that depend on multiple factors. These factors can be divided into patient related, clinician related, technical and systems related factors. Some of the outcomes are positive and some may be negative. Some of the positive outcomes of telehealth pain medicine were surprising. The illustrative case reports are chosen to highlight both the positive and negative aspects of delivering telehealth pain medicine services. Conclusions Given the uncertainty of the COVID-19 pandemic in terms of changes to health care in the United States and the duration of the pandemic, we can expect to be using telehealth pain medicine in our care of patients well into the future. The lessons illustrated in these cases categorized by care processes, clinical impact and administrative issues will be very important to carry going forward. Additional lessons will likely be learned as we continue to provide telehealth pain medicine services.