ABSTRACT
Novel Coronavirus 2019 disease or COVID-19 is a viral disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The use of chest X-rays (CXRs) has become an important practice to assist in the diagnosis of COVID-19 as they can be used to detect the abnormalities developed in the infected patients' lungs. With the fast spread of the disease, many researchers across the world are striving to use several deep learning-based systems to identify the COVID-19 from such CXR images. To this end, we propose an inverted bell-curve-based ensemble of deep learning models for the detection of COVID-19 from CXR images. We first use a selection of models pretrained on ImageNet dataset and use the concept of transfer learning to retrain them with CXR datasets. Then the trained models are combined with the proposed inverted bell curve weighted ensemble method, where the output of each classifier is assigned a weight, and the final prediction is done by performing a weighted average of those outputs. We evaluate the proposed method on two publicly available datasets: the COVID-19 Radiography Database and the IEEE COVID Chest X-ray Dataset. The accuracy, F1 score and the AUC ROC achieved by the proposed method are 99.66%, 99.75% and 99.99%, respectively, in the first dataset, and, 99.84%, 99.81% and 99.99%, respectively, in the other dataset. Experimental results ensure that the use of transfer learning-based models and their combination using the proposed ensemble method result in improved predictions of COVID-19 in CXRs.
ABSTRACT
BACKGROUND: Previous studies reported the beneficial effects of walking in individual with mild to moderate knee osteoarthritis (OA). The current study aimed to compare the effect of 6-week retro versus forward walking program versus control group on pain, functional disability, quadriceps muscle strength and physical performance in individuals with knee OA. METHODS: A three-arm single-blinded, randomized, controlled trial and intention-to-treat analysis was conducted in outpatient physiotherapy department, King Saud University, Saudi Arabia. Sixty-eight individuals (mean age, 55.6 years; 38 female) with knee OA participated. The participants in the retro or forward walking group completed 10 min of supervised retro or forward walking training in addition to usual care, 3 days/week for 6 weeks. The control group received a routine physiotherapy program. This program comprises a combination of closed and open kinematic chain exercises, including straight leg raising, isometric quadriceps, isometric hip adduction, terminal knee extension, semi-squat, and leg press. The primary outcomes were mean pain and knee function score measured by the numerical rating scale and the Western Ontario and McMaster Universities Osteoarthritis Index, respectively. The secondary outcomes were mean score of quadriceps muscle strength and timed up and go test scores. All the outcomes were analyzed at baseline and week 6. RESULTS: In total, 68 subjects participated in this 6-week randomized, controlled trial. The completion rates of the primary and secondary outcome measures at week 6 were 91, 87, and 82% in the retro walking, forward walking, and control groups, respectively. In the intention-to-treat analysis, the retro walking group had a greater reduction in pain intensity (mean changes, 1.8 versus 1; p = 0.01) and functional disability (mean changes, 4.8 versus 2.2; p = 0.008) than the control group. Similarly, the retro walking group had a greater improvement in the quadriceps muscle strength (mean changes, 1.7 kg versus 0.7 kg; p = 0.008) and the timed up and go test (mean changes, 0.6 s versus 0.1 s; p = 0.003) than the control group. CONCLUSIONS: The 6-week retro walking program compared with forward walking or control groups resulted in greater reduction in pain and functional disability and improved quadriceps muscle strength and performance in individuals with knee OA. TRIAL REGISTRATION: Controlled Trials ISRCTN12850845 , Registered 26 January 2015.
