Combining Inertial Sensors and Machine Learning to Predict vGRF and Knee Biomechanics during a Double Limb Jump Landing Task

Sensors, Год журнала: 2021, Номер 21(13), С. 4383 - 4383

Опубликована: Июнь 26, 2021

(1) Background: Biomechanics during landing tasks, such as the kinematics and kinetics of knee, are altered following anterior cruciate ligament (ACL) injury reconstruction. These variables recommended to assess prior clearance for return sport, but clinicians lack access current gold-standard laboratory-based assessment. Inertial sensors serve a potential solution provide clinically feasible means biomechanics augment sport testing. The purposes this study were (a) develop multi-sensor machine learning algorithms predicting (b) quantify accuracy each algorithm. (2) Methods: 26 healthy young adults completed 8 trials double limb jump task. Peak vertical ground reaction force, peak knee flexion angle, extension moment, sagittal power absorption assessed using 3D motion capture force plates. Shank- thigh- mounted inertial used collect data concurrently. submitted inputs single- multiple- feature linear regressions predict biomechanical in limb. (3) Results: Multiple-feature models, particularly when an accelerometer gyroscope together, valid predictors (R2 = 0.68–0.94, normalized root mean square error 4.6–10.2%). Single-feature models had decreased performance 0.16–0.60, 10.0–16.2%). (4) Conclusions: combination provides prediction This is both clinical real-world settings outside traditional laboratory.

Язык: Английский

---

AI-Based Task Classification With Pressure Insoles for Occupational Safety

IEEE Access, Год журнала: 2024, Номер 12, С. 21347 - 21357

Опубликована: Янв. 1, 2024

Pressure insoles allow for the collection of real time pressure data inside and outside a laboratory setting as they are non-intrusive can be simply integrated into industrial environments occupational health safety monitoring purposes. Activity detection is important wellbeing workers, present study aims to employ detect type industry-related task an individual performing by using random forest, artificial intelligence-based classification technique. Twenty subjects wore loadsol® performed five specific tasks associated with typical workflow: standing, walking, pick place, assembly, manual handling. For each activity, statistical morphological features were extracted create training dataset. The classifier accuracy over 82%, ten-fold cross-validation, window 5 seconds, showing potential in edge-AI applications smart manufacturing environments. A re-analysis focused on most influential obtained 83% accuracy. combination forest in-depth feature analysis (SHAP) provided insights importance impact their value class. Such understanding aid reducing misclassifications purposes design that optimized impactful features. achieved comparable similar studies but benefit added explainability, which increases transparency and, thereby, trust decisions.

Язык: Английский

---

The Effect of Sensor Feature Inputs on Joint Angle Prediction across Simple Movements

Sensors, Год журнала: 2024, Номер 24(11), С. 3657 - 3657

Опубликована: Июнь 5, 2024

The use of wearable sensors, such as inertial measurement units (IMUs), and machine learning for human intent recognition in health-related areas has grown considerably. However, there is limited research exploring how IMU quantity placement affect movement prediction (HMIP) at the joint level. objective this study was to analyze various combinations input signals maximize accuracy multiple simple movements. We trained a Random Forest algorithm predict future angles across these movements using sensor features. hypothesized that angle would increase with addition IMUs attached adjacent body segments non-adjacent not accuracy. results indicated current inputs did significantly (RMSE 1.92° vs. 3.32° ankle, 8.78° 12.54° knee, 5.48° 9.67° hip). Additionally, including 5.35° 5.55° 20.29° 20.71° 14.86° 13.55° These demonstrated during improve alongside inputs.

Язык: Английский

---

How Accurately Can Wearable Sensors Assess Low Back Disorder Risks during Material Handling? Exploring the Fundamental Capabilities and Limitations of Different Sensor Signals

Sensors, Год журнала: 2023, Номер 23(4), С. 2064 - 2064

Опубликована: Фев. 12, 2023

Low back disorders (LBDs) are a leading occupational health issue. Wearable sensors, such as inertial measurement units (IMUs) and/or pressure insoles, could automate and enhance the ergonomic assessment of LBD risks during material handling. However, much remains unknown about which sensor signals to use how accurately sensors can estimate injury risk. The objective this study was address two open questions: (1) How we risk when combining trunk motion under-the-foot force data (simulating IMU insoles used together)? (2) greater is accuracy than using only alone)? We developed data-driven simulation randomized lifting tasks, machine learning algorithms, validated tool. found that motion-based estimates were not strongly correlated (r range: 0.20–0.56) with ground truth risk, but adding yielded 0.93–0.98). These results raise questions adequacy single for handling suggest an on trained algorithms may be able assess risks.

Язык: Английский

---

Spine patient care with wearable medical technology: state-of-the-art, opportunities, and challenges: a systematic review

The Spine Journal, Год журнала: 2023, Номер 23(7), С. 929 - 944

Опубликована: Март 7, 2023

Язык: Английский

---