The use of privacy-protected computer vision to measure the quality of healthcare worker hand hygiene.

OBJECTIVES: (i) To demonstrate the feasibility of automated, direct observation and collection of hand hygiene data, (ii) to develop computer visual methods capable of reporting compliance with moment 1 (the performance of hand hygiene before touching a patient) and (iii) to report the diagnostic accuracy of automated, direct observation of moment 1. DESIGN: Observation of simulated hand hygiene encounters between a healthcare worker and a patient. SETTING: Computer laboratory in a university. PARTICIPANTS: Healthy volunteers. MAIN OUTCOME MEASURES: Sensitivity and specificity of automatic detection of the first moment of hand hygiene. METHODS: We captured video and depth images using a Kinect camera and developed computer visual methods to automatically detect the use of alcohol-based hand rub (ABHR), rubbing together of hands and subsequent contact of the patient by the healthcare worker using depth imagery. RESULTS: We acquired images from 18 different simulated hand hygiene encounters where the healthcare worker complied with the first moment of hand hygiene, and 8 encounters where they did not. The diagnostic accuracy of determining that ABHR was dispensed and that the patient was touched was excellent (sensitivity 100%, specificity 100%). The diagnostic accuracy of determining that the hands were rubbed together after dispensing ABHR was good (sensitivity 83%, specificity 88%). CONCLUSIONS: We have demonstrated that it is possible to automate the direct observation of hand hygiene performance in a simulated clinical setting. We used cheap, widely available consumer technology and depth imagery which potentially increases clinical application and decreases privacy concerns.

An integration engineering framework for machine learning in healthcare.

Background and Objectives: Machine Learning offers opportunities to improve patient outcomes, team performance, and reduce healthcare costs. Yet only a small fraction of all Machine Learning models for health care have been successfully integrated into the clinical space. There are no current guidelines for clinical model integration, leading to waste, unnecessary costs, patient harm, and decreases in efficiency when improperly implemented. Systems engineering is widely used in industry to achieve an integrated system of systems through an interprofessional collaborative approach to system design, development, and integration. We propose a framework based on systems engineering to guide the development and integration of Machine Learning models in healthcare. Methods: Applied systems engineering, software engineering and health care Machine Learning software development practices were reviewed and critically appraised to establish an understanding of limitations and challenges within these domains. Principles of systems engineering were used to develop solutions to address the identified problems. The framework was then harmonized with the Machine Learning software development process to create a systems engineering-based Machine Learning software development approach in the healthcare domain. Results: We present an integration framework for healthcare Artificial Intelligence that considers the entirety of this system of systems. Our proposed framework utilizes a combined software and integration engineering approach and consists of four phases: (1) Inception, (2) Preparation, (3) Development, and (4) Integration. During each phase, we present specific elements for consideration in each of the three domains of integration: The Human, The Technical System, and The Environment. There are also elements that are considered in the interactions between these domains. Conclusion: Clinical models are technical systems that need to be integrated into the existing system of systems in health care. A systems engineering approach to integration ensures appropriate elements are considered at each stage of model design to facilitate model integration. Our proposed framework is based on principles of systems engineering and can serve as a guide for model development, increasing the likelihood of successful Machine Learning translation and integration.

Developing a Screening Procedure During the COVID-19 Pandemic: Process and Challenges Faced by a Low-Incidence Area.

Purpose: To summarize the imaging results of COVID-19 pneumonia and develop a computerized tomography (CT) screening procedure for patients at our institution with malignant tumors. Methods: Following epidemiological investigation, 1,429 patients preparing to undergo anti-tumor-treatment underwent CT scans between February 17 and April 16, 2020. When CT findings showed suspected COVID-19 pneumonia after the supervisor radiologist and the thoracic experience radiologist had double-read the initial CT images, radiologists would report the result to our hospital infection control staff. Further necessary examinations, including the RT-PCR test, in the assigned hospital was strongly recommended for patients with positive CT results. The CT examination room would perform sterilization for 30 min to 1 h. If the negative results of any suspected COVID-19 pneumonia CT findings were identified, the radiologists would upload the results to our Hospital Information Systems and inform clinicians within 2 h. Results: Fifty (0.35%, 50/1,429) suspected pneumonia cases, including 29 males and 21 females (median age: 59.5 years old; age range 27-79 years), were identified. A total of 34.0% (17/50) of the patients had a history of lung cancer and 54.0 (27/50) underwent chemotherapy or targeted therapy. Forty-six patients (92.0%) had prior CT scans, and 35 patients (76.1%) with suspected pneumonia were newly seen (median interval time: 62 days). Sub-pleura small patchy or strip-like lesions most likely due to fibrosis or hypostatic pneumonia and cluster of nodular lesions were the two main signs of suspected cases on CT images (34, 68.0%). Twenty-seven patients (54.0%) had, at least once, follow-up CT scan (median interval time: 18.0 days). Only one patient had an increase in size (interval time: 8 days), the immediately RT-PCR test result was negative. Conclusion: CT may be useful as a screening tool for COVID-19 based on imaging features. But the differential diagnosis between COVID-19 and other pulmonary infection and/or non-infectious disease is very difficult due to its overlapping imaging features.The confirmed diagnosis of the COVID-19 infection should be based on the etiologic eventually. The cancer patients at a low-incidence area would continue treatment by screening carefully before admission.

Women's experiences of health care utilization in cases of early pregnancy loss: A scoping review.

BACKGROUND: Approximately 25% of pregnancies end in miscarriage, most occurring within the first trimester (<13 weeks). For many women early pregnancy loss has implications for short- and long- term mental health, and women's well-being following early pregnancy loss is impacted by their experiences within the healthcare setting. To improve quality of care, it is crucial to understand women's' experiences within the healthcare system in cases of early pregnancy loss. QUESTIONS: What does the research literature tell us about the experiences of early pregnancy loss within healthcare settings? Are these experiences positive or negative? 'How can care improve for those experiencing early pregnancy loss?' METHODS: A scoping review of the research literature was undertaken. Three research databases were searched for relevant articles published in English since 2009, with key words related to 'Experience', 'Healthcare' and 'Early Pregnancy Loss'. A thematic analysis was undertaken to identify and summarize key findings emerging from the research literature. FINDINGS: Twenty-seven (27) articles met our inclusion criteria. Three main themes were identified: (1) issues related to communication, (2) challenges within care environments, and (3) inadequacies in aftercare. DISCUSSION: The literature suggests that women's experiences related to healthcare for early pregnancy loss are largely negative, particularly within emergency departments. Recommendations to improve women's experiences should extend beyond attempts to improve existing care structures, to include emerging environments and providers. CONCLUSION: Women's experiences identified within the literature provide further insights on what women are seeking from their care, and how care models can be improved.

How Do Health Professionals Maintain Compassion Over Time? Insights From a Study of Compassion in Health.

Although compassion in healthcare differs in important ways from compassion in everyday life, it provides a key, applied microcosm in which the science of compassion can be applied. Compassion is among the most important virtues in medicine, expected from medical professionals and anticipated by patients. Yet, despite evidence of its centrality to effective clinical care, research has focused on compassion fatigue or barriers to compassion and neglected to study the fact that most healthcare professionals maintain compassion for their patients. In contributing to this understudied area, the present report provides an exploratory investigation into how healthcare professionals report trying to maintain compassion. In the study, 151 professionals were asked questions about how they maintained compassion for their patients. Text responses were coded, with a complex mixture of internal vs. external, self vs. patient, and immediate vs. general strategies being reported. Exploratory analyses revealed reliable individual differences in the tendency to report strategies of particular types but no consistent age-related differences between older and younger practitioners emerged. Overall, these data suggest that while a range of compassion-maintaining strategies were reported, strategies were typically concentrated in particular areas and most professionals seek to maintain care using internal strategies. A preliminary typology of compassion maintaining strategies is proposed, study limitations and future directions are discussed, and implications for the study of how compassion is maintained are considered.