This study delves into decoding hand gestures using surface electromyography (EMG) signals collected via a precision Myo-armband sensor, leveraging machine learning algorithms. The research entails rigorous data preprocessing to extract features and labels from raw EMG data. Following partitioning into training and testing sets, four traditional machine learning models are scrutinized for their efficacy in classifying finger movements across seven distinct gestures. The analysis includes meticulous parameter optimization and five-fold cross-validation to evaluate model performance. Among the models assessed, the Random Forest emerges as the top performer, consistently delivering superior precision, recall, and F1-score values across gesture classes, with ROC-AUC scores surpassing 99%. These findings underscore the Random Forest model as the optimal classifier for our EMG dataset, promising significant advancements in healthcare rehabilitation engineering and enhancing human–computer interaction technologies.

Diabetic Retinopathy is a vision impairment caused by blood vessel degeneration in the retina. It is becoming more widespread as it is linked to diabetes. Diabetic retinopathy can lead to blindness. Early detection of diabetic retinopathy by an ophthalmologist can help avoid vision loss and other complications. Diabetic retinopathy is currently diagnosed by visually recognizing irregularities on fundus pictures. This procedure, however, necessitates the use of ophthalmic imaging technologies to acquire fundus images as well as a detailed visual analysis of the stored photos, resulting in a costly and time-consuming diagnosis. The fundamental goal of this project is to create an easy-to-use machine learning model tool that can accurately predict diabetic retinopathy using pre-recorded digital fundus images. To create the suggested classifier model, we gathered annotated fundus images from publicly accessible data repositories and used two machine learning methods, support vector machine (SVM) and deep neural network (DNN). On test data, the proposed SVM model had a mean area under the receiver operating characteristic curve (AUC) of 97.11%, whereas the DNN model had a mean AUC of 99.15%.

A mathematical model of the knee in the sagittal plane was used to analyse the effect of tibial surface geometry and ligament orientations on the estimates of cruciate ligament forces. An elementary mechanical analysis was used to calculate the ligament forces L during a simulated antero-posterior (A-P) laxity test (passive test), and during isometric quadriceps contraction (IQE). Three sets of anatomical parameters (sites of bony attachments and lengths of the ligaments) were used from the literature. Flat, concave and convex shapes were considered for the tibial plateau articulating with a compatible convex femoral condyle. Also, the effects of position and posterior tilt of the tibial plateau were analysed for the concave and flat surfaces. The analysis showed that the ligament forces rose steeply as they approached collinearity with the contact force, C. For a flat tibia, the direction of C remained independent of flexion angle or position of the tibial plateau. For the concave tibia, the direction of C varied with flexion as well as with position of the centre of curvature, O, of the tibial plateau. C had an anterior component synergistic with the posterior cruciate ligament (PCL) when the point of contact, F, was anterior to O. On the other hand, C had a posterior component synergistic with the anterior cruciate ligament (ACL) when F was posterior to O. Also, posterior tilt of the flat or concave surfaces resulted in an increase in the anterior component of C. The effects of the convex tibial surface were opposite to those of the concave surface. During the A-P laxity test, the ligament forces arising from the concave tibia were very sensitive to the A-P position of O and to the posterior tilt of the plateau which allowed C and L to approach collinearity. Proximo-distal movement of O had a relatively insignificant effect on the ligament force. Similarly, compared to the concave surface, the posterior tilt of the flat tibia had a lesser effect on the ligament forces. Compared to the passive test, use of the muscle forces during IQE resulted in larger ligament forces, though the trends in force behaviour over the flexion range remained the same. The analysis emphasizes the importance of correct graft placement during ligament reconstruction, and the need to avoid steep inclination (more than 70° to tibial plateau). Increasing the inclination from 70 to 80° doubles the ligament force. Further, the analysis suggests that, in bicompartmental knee replacement, the efficacy of concave surfaces in replacing cruciate ligament function depends critically on the horizontal placement of the tibial component and its orientation about the mediolateral axis.

The orientations and moment arms of the knee extensor and flexor muscle tendons are evaluated with increasing values of muscle force during simulated isometric exercises. A four-bar linkage model of the knee in the sagittal plane was used to define the motion of the joint in the unloaded state during 0–120° flexion. The cruciate and collateral ligaments were represented by arrays of elastic fibres, which were recruited sequentially under load or remained buckled when slack. A bi-articular model of the patello-femoral joint was used. Simple straight-line representation was used for the lines of action of the forces transmitted by the model muscle tendons. The effects of tissue deformation with increasing muscle force were considered. During quadriceps contraction resisted by an external flexing load, the maximum change in moment arm of the patellar tendon was found to be 2 per cent at 0° flexion when the quadriceps force was increased tenfold, from 250 to 2500 N. The corresponding maximum change in orientation of the tendon was 3° at 120° flexion. During hamstrings contraction resisted by an external extending load, the maximum change in moment arm of the hamstrings tendon was 8 per cent at 60° flexion when the hamstrings force was increased tenfold, from 100 to 1000 N. During gastrocnemious contraction, the corresponding maximum change for the gastrocnemious tendon was 3 per cent at 0°. The orientations of the flexor muscle tendons in this range of force either remained constant or changed by 1° or less at any flexion angle. The general trend at any flexion angle was that, as the muscle force was increased, the moment arms and the orientations approached nearly constant values, showing asymptotic behaviour. It is concluded that experimental simulations of knee muscle action with low values of the externally applied load, of the order of 50 N, can provide reliable estimates of the relationships between muscle forces and external loads during activity.