A convenient approach for knee kinematics assessment using wearable inertial sensors during home-based rehabilitation: Validation with an optoelectronic system.

Rehabilitation services are among the most severely impacted by the COVID-19 pandemic. This has increased the number of people not receiving the needed rehabilitation care. Home-based rehabilitation becomes alternative support to face this greater need. However, monitoring kinematics parameters during rehabilitation exercises is critical for an effective recovery. This work proposes a detailed framework to estimate knee kinematics using a wearable Magnetic and Inertial Measurement Unit (MIMU). That allows at-home monitoring for knee rehabilitation progress. Two MIMU sensors were attached to the shank and thigh segments respectively. First, the absolute orientation of each sensor was estimated using a sensor fusion algorithm. Second, these sensor orientations were transformed to segments orientations using a functional sensor-to-segment (STS) alignment. Third, the relative orientation between segments, i.e., knee joint angle, was computed and the relevant kinematics parameters were extracted. Then, the validity of our approach was evaluated with a gold-standard optoelectronic system. Seven participants completed three to five Timed-Up-and-Go (TUG) tests. The estimated knee angle was compared to the reference angle. Root-mean-square error (RMSE), correlation coefficient, and Bland-Altman analysis were considered as evaluation metrics. Our results showed reasonable accuracy (RMSE < 8°), strong to very-strong correlation (correlation coefficient > 0.86), a mean difference within 1.1°, and agreement limits from -16° to 14°. In addition, no significant difference was found (p-value > 0.05) in extracted kinematics parameters between both systems. The proposed approach might represent a suitable alternative for the assessment of knee rehabilitation progress in a home context.

Comparing CT scan and chest X-ray imaging for COVID-19 diagnosis.

People suspected of having COVID-19 need to know quickly if they are infected, so they can receive appropriate treatment, self-isolate, and inform those with whom they have been in close contact. Currently, the formal diagnosis of COVID-19 requires a laboratory test (RT-PCR) on samples taken from the nose and throat. The RT-PCR test requires specialized equipment and takes at least 24 h to produce a result. Chest imaging has demonstrated its valuable role in the development of this lung disease. Fast and accurate diagnosis of COVID-19 is possible with the chest X-ray (CXR) and computed tomography (CT) scan images. Our manuscript aims to compare the performances of chest imaging techniques in the diagnosis of COVID-19 infection using different convolutional neural networks (CNN). To do so, we have tested Resnet-18, InceptionV3, and MobileNetV2, for CT scan and CXR images. We found that the ResNet-18 has the best overall precision and sensitivity of 98.5% and 98.6%, respectively, the InceptionV3 model has achieved the best overall specificity of 97.4%, and the MobileNetV2 has obtained a perfect sensitivity for COVID-19 cases. All these performances have occurred with CT scan images.