In the era of long COVID, can we seek new techniques for better rehabilitation?

Abstract

Since the coronavirus disease 2019 (COVID-19) outbreak has become a pandemic, medical staff and researchers have devotedly managed the disease in terms of pathogens, prevention, and treatment. Even so, the virus continues to wreak havoc in people's lives. Recent evidence shows that patients with severe acute respiratory syndrome coronavirus 2 show distinct symptoms ranging from asymptomatic or mild infection to fatal disease. Moreover, this virus not only provokes an acute inflammatory response but could also cause a range of persistent symptoms after the phase of acute infection. The National Institute for Health and Care Excellence defines this phenomenon as long COVID.1 Currently, there is no definite definition of long COVID. However, the basic argument, that patients who have recovered from acute infection present with persistent symptoms that cannot be explained by another diagnosis beyond 3–4 weeks postinfection, remains the same. Academic publications have estimated that 10%–20% of patients with COVID-19 have some complaints after COVID-19.2 Most patients with COVID-19 present with respiratory dysfunction, fatigue, and psychological disorders during and after acute infection, and their lives are greatly affected. Under this circumstance, pulmonary rehabilitation (PR) has become a powerful weapon to deal with those symptoms in the context of quantities of evidence of its efficiency.3 However, restricting quarantine to minimize viral spread may limit the utility of conventional PR programs. Thus, new techniques and innovative programs or approaches must be discussed and estimated. In this case, we could obtain a more comprehensive and critical perspective on the frontiers of medicine. The pandemic necessitated a transformation of traditional PR. More physiotherapists have begun to move their attention from face-to-face to remote rehabilitation and evaluate its accessibility and safety. Remote rehabilitation, also named telerehabilitation, involves various techniques and could eliminate the limitations of distance. The techniques range from cell phone messages to videoconferencing and Internet platforms. It may not only provide more access to PR but also help maintain the original outcomes of traditional PR. Accordingly, it facilitates many new techniques to conduct home-based assessments to guarantee the safety of PR and improve the prescription of PR programs. During the pandemic, artificial intelligence (AI) technology has been used to support diagnosis, treatment and vaccine discovery, epidemiological modeling, patient outcome–related tasks, and infodemiology.4, 5 Currently, an increasing number of AI or robotic technologies are used to provide training to students and healthcare professionals, find appropriate remote rehabilitation approaches, and implement a standard PR program. Initially, the environment and assessment are essential to support engagement and motivation. Thus, Vourganas et al.6 demonstrated a patient-centric individualized home-based rehabilitation support system with the help of accountability, responsibility, and transparency (ART) AI. They used the timed up and go (TUG) and five-time sit to stand (FTSTS) tests to evaluate patients’ daily living activity performance in the presence or development of comorbidities. To satisfy the requirements of individualization, interpretability, and ART design, they used a hybrid learning approach. As a result, the model reaches up to 100% accuracy for both FTSTS and TUG in predicting medical conditions, 100% for FTSTS and 83.13% for TUG in predicting the area of difficulty, and is much more accurate and individualized than the state-of-the-art approaches in TUG and FTSTS AI. Moreover, the author advocates that if the condition can be more finely classified, such as by severity, the model can be improved. Lemhöfer et al.7 developed an integrative survey questionnaire, The COVID-19 Rehabilitation Needs Survey (C19-RehabNeS) for COVID-19 patients assesses function