Tele-ACP amidst COVID-19 pandemic and beyond

BackgroundWhen COVID-19 hit Singapore in 2020, the public was advised to avoid visiting the hospitals unless for essential services. Advance Care Planning (ACP) services in hospital and community had to be stopped to reduce exposure for the public. However, it was not feasible for ACP services to stop with no foresight of when it could resume. Ironically, ACP should all the more be advocated amidst the pandemic.Henceforth, the team planned and implemented a tele-ACP workflow in February 2020 to ensure accessibility in continuity of care and reduce waiting time for ACP appointments.MethodsTele-ACP is conducted only via Zoom, given its security and encryption features. Criteria of patients include ability to read basic English, having electronic device with internet access, and having no severe hearing or speech impairment.Pre-ACP appointment: Zoom details including a guide were sent to patients and/or NHS.On appointment day: Before session starts, ACP Facilitator will ensure that patient and/or NHS are at a space where there is privacy. Internet stability will also be checked.Post-ACP appointment: Signatures will be obtained electronically or via post, while ensuring personal data is well-protected.ResultsFrom February 2020 to November 2022, 105 tele-ACPs (14 General ACPs and 91 Preferred Plan of Care) were completed. 45 were completed in 2020 and 2021 each, while 15 were done in 2022 (as of November). The average duration for tele vs in-person ACPs is both about 90 minutes, indicating that the effort and time spent are not any less despite ACP discussions being done virtually.ConclusionsLooking at the number of tele-ACPs completed and how it is still actively carried out despite COVID-19 situation being stable and restrictions lifted, tele-ACP is clearly in healthy demand. This shows that tele-ACP is here to stay, being both sustainable and transferable to multiple settings.

Analysis of Antibodies Induced after SARS-CoV-2 Vaccination Using Antigen Coded Bead Array Luminex Technology.

Objectives. Since the outbreak of SARS-CoV-2 in late 2019, nearly 12.2 billion doses of the COVID-19 vaccine have been administered worldwide; however, the humoral immune responses induced by different types of vaccines are yet to be fully validated. Methods. We analyzed antibody levels in 100 serum samples after vaccination with different types of COVID-19 vaccines and their reactivity against the RBD antigen of Delta and Omicron variants using a bead-based microarray. Results. Elevated levels of anti-wild-type (WT)-RBD IgG and anti-WT-NP IgG were detected in participants who received two doses of the inactivated vaccines (CoronaVac or BBIBP-CorV) and three doses of the recombinant spike protein vaccine (ZF2001), indicating that antibody responses to SARS-CoV-2 were generated regardless of the vaccine administered. We found highly correlated levels of serum anti-RBD IgG and anti-NP IgG (r = 0.432, p < 0.001). We observed that the antibodies produced in vivo after COVID-19 vaccination still reacted with variants of SARS-CoV-2 (p < 0.0001). Conclusions. Our results show that high levels of specific antibodies can be produced after completion of COVID-19 vaccination (two doses of the inactivated vaccines or three doses of ZF2001), with some degree of cross-reactivity to the RBD antigen of Delta and Omicron variants, and provide an accessible and practical experimental method for post-vaccination antibody detection.

Selection of Pediatric Mental Health Quality Measures for Health System Improvement in British Columbia Based on a Modified Delphi Approach.

Background: The COVID-19 pandemic has highlighted the importance of mental wellbeing. The identification and implementation of quality measures can improve health outcomes and patient experience. The objective was to identify and define a core set of valid and relevant pediatric mental health quality measures that will support health system evaluation and quality improvement in British Columbia, Canada. Methods: The study consisted of four phases. First, a comprehensive database search identified valid pediatric quality measures focused on mental health and substance use (MH/SU). Second, the identified quality measures were mapped to focus areas, which were then prioritized by two stakeholder groups consisting of 26 members. Third, up to two representative measures for each prioritized focus area were pre-selected by an expert panel (n = 9). And fourth, a three-step modified Delphi approach was employed to (1) assess each quality measure on a 7-point Likert scale against three relevance criteria (representative of a quality problem, value to intended audience and actionable), (2) discuss the results, and (3) select and rank the most relevant measures. Forty-eight stakeholders were invited to participate; of those 24 completed the round 1 survey, 21 participated in the round 2 discussion and 18 voted in the round 3 selection and ranking survey. For round 1, consensus was determined when at least 70% of the response rates were within the range of five to seven. For round 3, Kendall's coefficient of concordance W was used as an estimator of inter-rater reliability. Results: One-hundred pediatric mental health quality measures were identified in the database search. Of those, 37 were mapped to ten focus areas. Pre-selection resulted in 19 representative measures moving forward to the Delphi study. Eleven measures met the consensus thresholds and were brought forward to the round 2 discussion. Round 3 ranking showed moderate to strong raters' agreement (Kendall's W = 0.595; p < 0.01) and resulted in the following five highest-ranked measures: level of satisfaction after discharge from inpatient admission due to MH/SU, number of patients experiencing seclusion or restraint, length of time from eating disorder referral to assessment, number of ED visits due to MH/SU, and number of readmissions to ED. Conclusion: The selected core set of valid and relevant pediatric quality measures will support sustainable system change in British Columbia. The five top-ranked measures will be refined and tested for data collection feasibility before being implemented in the province.

Investigation of a Mask Fitness Test Based on Self-Efficacy and Diversified Training in the Assessment System for Nosocomial Infection Training.

Objective: To explore a mask fitness test based on self-efficacy and diversified training in the assessment system for nosocomial infection training. Methods: From March 15 to April 5, 2022, 442 staff members (272 male and 170 female) of the Third People's Hospital of Shenzhen who planned to enter the quarantine ward for secondary protection skill training assessment were selected. They comprised 56 doctors, 31 medical technicians, 72 nurses, and 283 property logistics staff. During the mask fitness test, a diversified training model based on self-efficacy was adopted to observe the passing status, the identification and selection of mask models, the method of mask-wearing, the fit between the mask and the face, and the changes in self-efficacy. Results: In the assessment system for nosocomial infection training, the passing rate of the mask fitness test was correlated with the identification and selection of mask models, the method of wearing masks, the fit between the mask and the face, and the diversified training, and the differences were statistically significant (P < 0.05). The difference in the self-efficacy in the test takers between those before and after the mask fitness test was statistically significant (P < 0.05). Conclusion: In the assessment system for nosocomial infection training, the mask fitness test based on self-efficacy and diversified training might improve the passing rate, the rate of correct mask model identification and selection, the rate of correct mask-wearing, and the degree of facial fit, thus to enhance the awareness of protection and improve self-efficacy.

Acceptability of Telegram-based Training among Aged Care Providers

During the COVID-19 pandemic, increasing infection prevention and control (IPC) in nursing homes was challenging due to the impracticality of typical face-to-face teaching. Concerns about device availability, Internet connectivity, and acceptance of online training for care workers led to the selection of Telegram for its minimal bandwidth use and appeal in the socio-educational setting. An online survey assessing the acceptability of a Telegram-based IPC training program was conducted on 60 care providers from 14 aged care facilities. Despite the strong acceptability of Telegram-based training, there are challenges to be overcome when delivering online learning for aged care providers. [ FROM AUTHOR]

Using Text Mining and Data Visualization Approaches for Investigating Mental Illness from the Perspective of Traditional Chinese Medicine.

Background and Objectives. Anxiety and depressive disorders are the most prevalent mental disorders, and due to the COVID-19 pandemic, more people are suffering from anxiety and depressive disorders, and a considerable fraction of COVID-19 survivors have a variety of persistent neuropsychiatric problems after the initial infection. Traditional Chinese Medicine (TCM) offers a different perspective on mental disorders from Western biomedicine. Effective management of mental disorders has become an increasing concern in recent decades due to the high social and economic costs involved. This study attempts to express and ontologize the relationships between different mental disorders and physical organs from the perspective of TCM, so as to bridge the gap between the unique terminology used in TCM and a medical professional. Materials and Methods. Natural language processing (NLP) is introduced to quantify the importance of different mental disorder descriptions relative to the five depots and two palaces, stomach and gallbladder, through the classical medical text Huangdi Neijing and construct a mental disorder ontology based on the TCM classic text. Results. The results demonstrate that our proposed framework integrates NLP and data visualization, enabling clinicians to gain insights into mental health, in addition to biomedicine. According to the results of the relationship analysis of mental disorders, depots, palaces, and symptoms, the organ/depot most related to mental disorders is the heart, and the two most important emotion factors associated with mental disorders are anger and worry & think. The mental disorders described in TCM are related to more than one organ (depot/palace). Conclusion. This study complements recent research delving into co-relations or interactions between mental status and other organs and systems.

Impact of increased digital use and internet gaming on nursing students' empathy: A cross-sectional study.

BACKGROUND: The COVID-19 restrictions and quarantines had led to increased dependence and usage of digital devices for various human activities and internet gaming to the extent of risking vulnerable individuals to develop addiction towards it. Little is known on such risks among populations of nursing students and its impact on their empathy skills or trait. OBJECTIVE: Determining the impact of digital use and internet gaming on empathy of nursing students undergoing remote learning during closure of learning institutions nationwide. DESIGN: Cross-sectional online survey was conducted from October to December 2020. SETTINGS: Two established public institutions located in Malaysia. PARTICIPANTS: A total of 345 nursing students pursuing diploma and bachelor nursing programs. METHODS: Toronto Empathy Questionnaire (TEQ), Digital Addiction Scale (DAS) and Internet Gaming Disorder Scale-Short form (IGDS9-SF) were self-administered via Google Form™. Following principal component analysis of TEQ using IBM-SPSS™ (V-27), path analyses was performed using SmartPLS™ (V-3). RESULTS: Despite the increased time spent on digital devices (∆ 2.8 h/day) and internet gaming (∆ 1 h/week) before and during the pandemic, the proportion of high digital users (1.4 %) and gamers (20.9 %) were low; and sizable ≈75 % had higher-than-normal empathy. Digital-related emotions and overuse of them were associated with lower empathy (ß = -0.111, -0.192; p values < 0.05) and higher callousness (ß = 0.181, 0.131; p values < 0.05); internet gaming addiction predicted callousness (ß = 0.265, p < 0.001) but digital dependence correlated with higher empathy (ß = 0.172, p = 0.009). CONCLUSIONS: Digital and internet gaming addiction potentially impact empathy. The negative impact of digital dependence can be attenuated by "digital empathy" - an emerging phenomenon becoming increasingly vital in digital health and communication.

Adrenal Crisis Mimicking COVID-19 Encephalopathy in a Teenager with Craniopharyngioma.

There is an increasing number of reported cases with neurological manifestations of COVID-19 in children. Symptoms include headache, general malaise, ageusia, seizure and alterations in consciousness. The differential diagnosis includes several potentially lethal conditions including encephalopathy, encephalitis, intracranial hemorrhage, thrombosis and adrenal crisis. We report the case of a 17-year-old boy with a positive antigen test of COVID-19 who presented with fever for one day, altered mental status and seizure, subsequently diagnosed with adrenal insufficiency. He had a history of panhypopituitarism secondary to a suprasellar craniopharyngioma treated with surgical resection; he was treated with regular hormone replacement therapy. After prompt administration of intravenous hydrocortisone, his mental status returned to normal within four hours. He recovered without neurologic complications. Adrenal insufficiency can present with neurological manifestations mimicking COVID-19 encephalopathy. Prompt recognition and treatment of adrenal insufficiency, especially in patients with brain tumors, Addison's disease or those recently treated with corticosteroids, can rapidly improve the clinical condition and prevent long-term consequences.

Selection of Pediatric Mental Health Quality Measures for Health System Improvement in British Columbia Based on a Modified Delphi Approach

Background The COVID-19 pandemic has highlighted the importance of mental wellbeing. The identification and implementation of quality measures can improve health outcomes and patient experience. The objective was to identify and define a core set of valid and relevant pediatric mental health quality measures that will support health system evaluation and quality improvement in British Columbia, Canada. Methods The study consisted of four phases. First, a comprehensive database search identified valid pediatric quality measures focused on mental health and substance use (MH/SU). Second, the identified quality measures were mapped to focus areas, which were then prioritized by two stakeholder groups consisting of 26 members. Third, up to two representative measures for each prioritized focus area were pre-selected by an expert panel (n = 9). And fourth, a three-step modified Delphi approach was employed to (1) assess each quality measure on a 7-point Likert scale against three relevance criteria (representative of a quality problem, value to intended audience and actionable), (2) discuss the results, and (3) select and rank the most relevant measures. Forty-eight stakeholders were invited to participate;of those 24 completed the round 1 survey, 21 participated in the round 2 discussion and 18 voted in the round 3 selection and ranking survey. For round 1, consensus was determined when at least 70% of the response rates were within the range of five to seven. For round 3, Kendall's coefficient of concordance W was used as an estimator of inter-rater reliability. Results One-hundred pediatric mental health quality measures were identified in the database search. Of those, 37 were mapped to ten focus areas. Pre-selection resulted in 19 representative measures moving forward to the Delphi study. Eleven measures met the consensus thresholds and were brought forward to the round 2 discussion. Round 3 ranking showed moderate to strong raters' agreement (Kendall's W = 0.595;p < 0.01) and resulted in the following five highest-ranked measures: level of satisfaction after discharge from inpatient admission due to MH/SU, number of patients experiencing seclusion or restraint, length of time from eating disorder referral to assessment, number of ED visits due to MH/SU, and number of readmissions to ED. Conclusion The selected core set of valid and relevant pediatric quality measures will support sustainable system change in British Columbia. The five top-ranked measures will be refined and tested for data collection feasibility before being implemented in the province.

Low-Dose SARS-CoV-2 S-Trimer with an Emulsion Adjuvant Induced Th1-Biased Protective Immunity.

During the sustained COVID-19 pandemic, global mass vaccination to achieve herd immunity can prevent further viral spread and mutation. A protein subunit vaccine that is safe, effective, stable, has few storage restrictions, and involves a liable manufacturing process would be advantageous to distribute around the world. Here, we designed and produced a recombinant spike (S)-Trimer that is maintained in a prefusion state and exhibits a high ACE2 binding affinity. Rodents received different doses of S-Trimer (0.5, 5, or 20 µg) antigen formulated with aluminum hydroxide (Alum) or an emulsion-type adjuvant (SWE), or no adjuvant. After two vaccinations, the antibody response, T-cell responses, and number of follicular helper T-cells (Tfh) or germinal center (GC) B cells were assessed in mice; the protective efficacy was evaluated on a Syrian hamster infection model. The mouse studies demonstrated that adjuvating the S-Trimer with SWE induced a potent humoral immune response and Th1-biased cellular immune responses (in low dose) that were superior to those induced by Alum. In the Syrian hamster studies, when S-Trimer was adjuvanted with SWE, higher levels of neutralizing antibodies were induced against live SARS-CoV-2 from the original lineage and against the emergence of variants (Beta or Delta) with a slightly decreased potency. In addition, the SWE adjuvant demonstrated a dose-sparing effect; thus, a lower dose of S-Trimer as an antigen (0.5 µg) can induce comparable antisera and provide complete protection from viral infection. These data support the utility of SWE as an adjuvant to enhance the immunogenicity of the S-Trimer vaccine, which is feasible for further clinical testing.

Salutogenesis and COVID-19 pandemic impacting nursing education across SEANERN affiliated universities: A multi-national study.

BACKGROUND: The COVID-19 pandemic has disrupted the lives of many. Particularly, nursing students experience greater stress as their normal curriculum is interrupted and some of them face the risk of being infected as frontline workers. Nursing faculty members may face similar struggles, in addition to developing teaching materials for online learning. Thus, it is important to examine the faculty members' and students' views on their ability to adapt during the pandemic to obtain a holistic view of how learning and training has been affected. DESIGN: The descriptive cross-sectional quantitative design was used. SETTINGS: Data were collected from Southeast and East Asian Nursing Education and Research Network (SEANERN) affiliated nursing institutions from January 2021 to August 2021. PARTICIPANTS: A total of 1897 nursing students and 395 faculty members from SEANERN-affiliated nursing institutions in Cambodia, Hong Kong, Indonesia, Japan, Laos, Malaysia, Philippines, Singapore, Thailand and Vietnam were recruited for this study. METHODS: Quantitative surveys were used to explore the satisfaction levels in education modalities, confidence levels, psychosocial well-being, sense of coherence and stress levels of nursing students and faculty members during the COVID-19 pandemic. RESULTS: Participants were mostly satisfied with the new education modalities, although most students felt that their education was compromised. Both groups showed positive levels of psychosocial well-being, despite scoring low to medium on the sense of coherence scale and experiencing great stress. The participants' sense of coherence was positively correlated with their psychosocial well-being and negatively correlated with stress levels. CONCLUSIONS: While the COVID-19 pandemic had negatively impacted the lives of nursing students and faculty members, most of them had a healthy level of psychosocial well-being. Having a strong sense of coherence was associated with better psychosocial health and lower stress levels. As such, it may be helpful to develop interventions aimed at improving the sense of coherence of nursing students and staff to help them manage stressors better.

Monoclonal antibody targeting the conserved region of the SARS-CoV-2 spike protein to overcome viral variants.

Most therapeutic mAbs target the receptor-binding domain (RBD) of the spike protein of SARS-CoV-2. Unfortunately, the RBD is a hot spot for mutations in SARS-CoV-2 variants, which will lead to loss of the neutralizing function of current therapeutic mAbs. Universal mAbs for different variants are necessary. We identified mAbs that recognized the S2 region of the spike protein, which is identical in different variants. The mAbs could neutralize SARS-CoV-2 infection and protect animals from SARS-CoV-2 challenge. After cloning the variable region of the light chain and heavy chain, the variable region sequences were humanized to select a high-affinity humanized mAb, hMab5.17. hMab5.17 protected animals from SARS-CoV-2 challenge and neutralized SARS-CoV-2 variant infection. We further identified the linear epitope of the mAb, which is not mutated in any variant of concern. These data suggest that a mAb recognizing the S2 region of the spike protein will be a potential universal therapeutic mAb for COVID-19.

Nursing students' and faculty members' experiences of online education during COVID-19 across Southeast Asia: A Photovoice study.

BACKGROUND: The COVID-19 pandemic has caused extraordinary disruptions to education systems globally, forcing a rapid switch from conventional to online education. Although some qualitative studies have been carried out exploring the online education experiences of nursing students and faculty members during the COVID-19 pandemic, to our knowledge, no study has used the Photovoice approach. OBJECTIVES: To explore the experiences of nursing students and faculty members as related to online education during the COVID-19 pandemic. DESIGN: A descriptive qualitative design using Photovoice was adopted. SETTING: The study took place across five countries and one city in Southeast Asia (Indonesia, Malaysia, the Philippines, Thailand, Vietnam, and Hong Kong). PARTICIPANTS: Fifty-two nursing students and twenty-eight nursing faculty members who participated in online education during the COVID-19 pandemic. METHODS: Each participant submitted one photo substantiated with written reflections. Data were analysed using thematic analysis. Ethical approval was obtained from institution-specific ethics boards. RESULTS: Three themes and eleven sub-themes emerged from the data. The three main themes were: 1) Psychological roadblocks to online education; 2) Developing resilience despite adversities; and 3) Online education: What worked and what did not. CONCLUSION: Through Photovoice, the reflections revealed that nursing students and faculty members were generally overwhelmed with the online education experience. At the same time, participants were satisfied with the flexibility and convenience, opportunities for professional and personal development and safety afforded by online education. However, concerns over academic integrity, practical skills and clinical competencies, engagement and participation, the duality of technology and social isolation out-shadowed the advantages. It is worthwhile to explore the concerns raised to enhance online education across the nursing curriculum.

The successful rehabilitation of a 75-year-old female with debilitating long COVID: A case report.

A 75-year-old previously healthy female became severely ill, functionally dependent, and required long-term home oxygen therapy, after recovery from coronavirus disease 2019 (COVID-19) with acute respiratory failure and extensive pulmonary fibrosis. After two months of respiratory muscle training and a comprehensive cardiopulmonary rehabilitation program, her dyspnea, physical performance, pulmonary function parameters, and activities of daily living rapidly improved. This Case highlights the importance of a timely active rehabilitation program for COVID-19 survivors experiencing the long-term effects of coronavirus (long COVID).

Patients' Technology Readiness and eHealth Literacy: Implications for Adoption and Deployment of eHealth in the COVID-19 Era and Beyond.

The COVID-19 pandemic has rerouted the healthcare ecosystem by accelerating digital health, and rapid adoption of eHealth is partly influenced by eHealth literacy (eHL). This study aims to examine patients' eHL in relation to their "technology readiness"-an innate attitude that is underexplored in clinical research. A total of 276 adult inpatients with hypertension, diabetes mellitus, and coronary heart disease were surveyed cross-sectionally in 2019 using self-reported questionnaires: eHealth Literacy Scale and Technology Readiness Index (2.0). The study found moderate eHL (mean, 27.38) and moderate technology readiness (mean, 3.03) among patients. The hierarchical regression model shows that lower eHL scores were associated with patients of minor ethnicity (Malaysian Chinese), with an unemployed status, and having >1 cardiovascular risk (ß = -0.136 to -0.215, R2 = 0.283, Ps < .005). Technology readiness is a strong determinant of eHL (ΔR2 = 0.295, P < .001) with its subdomains (optimism, innovativeness, and discomfort) significantly influencing eHL (|ß| = 0.28-0.40, Ps < .001), except for the insecurity subdomain. Deployment of eHealth interventions that incorporate assessment of patients' eHL and technology readiness will enable targeted strategies, especially in resource-limited settings hit hard by the pandemic crisis.

Development of the International Classification of Diseases Ontology (ICDO) and its application for COVID-19 diagnostic data analysis.

BACKGROUND: The 10th and 9th revisions of the International Statistical Classification of Diseases and Related Health Problems (ICD10 and ICD9) have been adopted worldwide as a well-recognized norm to share codes for diseases, signs and symptoms, abnormal findings, etc. The international Consortium for Clinical Characterization of COVID-19 by EHR (4CE) website stores diagnosis COVID-19 disease data using ICD10 and ICD9 codes. However, the ICD systems are difficult to decode due to their many shortcomings, which can be addressed using ontology. METHODS: An ICD ontology (ICDO) was developed to logically and scientifically represent ICD terms and their relations among different ICD terms. ICDO is also aligned with the Basic Formal Ontology (BFO) and reuses terms from existing ontologies. As a use case, the ICD10 and ICD9 diagnosis data from the 4CE website were extracted, mapped to ICDO, and analyzed using ICDO. RESULTS: We have developed the ICDO to ontologize the ICD terms and relations. Different from existing disease ontologies, all ICD diseases in ICDO are defined as disease processes to describe their occurrence with other properties. The ICDO decomposes each disease term into different components, including anatomic entities, process profiles, etiological causes, output phenotype, etc. Over 900 ICD terms have been represented in ICDO. Many ICDO terms are presented in both English and Chinese. The ICD10/ICD9-based diagnosis data of over 27,000 COVID-19 patients from 5 countries were extracted from the 4CE. A total of 917 COVID-19-related disease codes, each of which were associated with 1 or more cases in the 4CE dataset, were mapped to ICDO and further analyzed using the ICDO logical annotations. Our study showed that COVID-19 targeted multiple systems and organs such as the lung, heart, and kidney. Different acute and chronic kidney phenotypes were identified. Some kidney diseases appeared to result from other diseases, such as diabetes. Some of the findings could only be easily found using ICDO instead of ICD9/10. CONCLUSIONS: ICDO was developed to ontologize ICD10/10 codes and applied to study COVID-19 patient diagnosis data. Our findings showed that ICDO provides a semantic platform for more accurate detection of disease profiles.

The COVID-19 Global Pandemic and Its Impact on the Mental Health of Nurses in Malaysia.

The Coronavirus disease 2019 (COVID-19) global pandemic since its onset has had a dramatic and often devastating impact, both physical and psychological, on all healthcare workers. This study aimed to assess the impact of psychological distress that COVID-19 has on nurses, as well as the coping strategies that they employed. This is a cross-sectional national online survey. A total of 859 nurses actively involved in caring for patients with suspected or confirmed COVID-19 in Malaysia participated in the study. More than three-quarters of the nurses experienced stress (77.2%). A total of 88.7% and 7.2% of nurses revealed a moderate and high stress level, respectively. Approximately one in eight (12.1%) nurses reported feeling depressed. Nurses working in the outpatient departments reported significantly higher stress levels than nurses working in inpatient care departments. Nurses having chronic health problems reported significantly higher depression levels than nurses with no chronic health problem. Highly stressed or depressed nurses tend to adopt avoidance coping strategies while religion and emotional support were used regardless of the stress or depression levels experienced. The findings of the study provide insight into the mental health and coping strategies of nurses actively involved in caring for patients with suspected or confirmed COVID-19 in Malaysia. This would be of tremendous help to nursing administrators in implementing mental health services for nurses during and following the COVID-19 global pandemic.

Benchmarking of eight recurrent neural network variants for breath phase and adventitious sound detection on a self-developed open-access lung sound database-HF_Lung_V1.

A reliable, remote, and continuous real-time respiratory sound monitor with automated respiratory sound analysis ability is urgently required in many clinical scenarios-such as in monitoring disease progression of coronavirus disease 2019-to replace conventional auscultation with a handheld stethoscope. However, a robust computerized respiratory sound analysis algorithm for breath phase detection and adventitious sound detection at the recording level has not yet been validated in practical applications. In this study, we developed a lung sound database (HF_Lung_V1) comprising 9,765 audio files of lung sounds (duration of 15 s each), 34,095 inhalation labels, 18,349 exhalation labels, 13,883 continuous adventitious sound (CAS) labels (comprising 8,457 wheeze labels, 686 stridor labels, and 4,740 rhonchus labels), and 15,606 discontinuous adventitious sound labels (all crackles). We conducted benchmark tests using long short-term memory (LSTM), gated recurrent unit (GRU), bidirectional LSTM (BiLSTM), bidirectional GRU (BiGRU), convolutional neural network (CNN)-LSTM, CNN-GRU, CNN-BiLSTM, and CNN-BiGRU models for breath phase detection and adventitious sound detection. We also conducted a performance comparison between the LSTM-based and GRU-based models, between unidirectional and bidirectional models, and between models with and without a CNN. The results revealed that these models exhibited adequate performance in lung sound analysis. The GRU-based models outperformed, in terms of F1 scores and areas under the receiver operating characteristic curves, the LSTM-based models in most of the defined tasks. Furthermore, all bidirectional models outperformed their unidirectional counterparts. Finally, the addition of a CNN improved the accuracy of lung sound analysis, especially in the CAS detection tasks.