AIoT-Enabled Rehabilitation Recognition System—Exemplified by Hybrid Lower-Limb Exercises

Lai, Yi-Chun; Kan, Yao-Chiang; Lin, Yu-Chiang; Lin, Hsueh-Chun

doi:10.3390/s21144761

Cited by 13 publications

(8 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further SVM (Support Vector Machine) method is mostly used to perform the classification of the data because of its unique properties such as effectiveness in high dimensional spaces, clear margin of separation between classes and number of dimensions is more than the number of samples as shown in Lai et al [24]. We have two labels SSLR and LSLR in this research so a 1-vs-1 strategy was chosen in which using respective data from both the classes the SVMs are formed.…”

Section: Methodsmentioning

confidence: 99%

Classification of lower limb rehabilitation exercises with multiple and individual inertial measurement units

Tannna¹,

Vithalani

2022

IJEECS

View full text Add to dashboard Cite

Straight leg raise rehabilitation exercises (for both lying and seated position) for lower limb injuries play a critical role in terms of stress on joints after the injury. The primary objective of the paper is to find how accurately and efficiently a single and a two IMU sensor-based system could classify SSLR (Seated straight leg raise) and LSLR (Lying straight leg raise) exercises using machine learning. Inertial Measurement Units (IMUs) that include accelerometer and gyroscope were calibrated and tested, individual and combined, for classified seating as well as lying exercise and for different demanded personalities. Individual IMUs achieved about 96 % accuracy in binary classification. However, the combined (two) IMUs achieved about 96.8 % accuracy. The merits of the proposed IMU based sensor system are that it is easy to install, cost effective and very useful for telemedical operations in pandemic situations like COVID19. On the basis of these results, it could be concluded that the accuracy of a single IMU sensor system and a two IMU sensor-based system is approximately 96% and both were efficiently able to classify SSLR and LSLR exercises as well as identify the individual performing the exercise.

show abstract

Section: Methodsmentioning

confidence: 99%

Classification of lower limb rehabilitation exercises with multiple and individual inertial measurement units

Tannna¹,

Vithalani

2022

IJEECS

View full text Add to dashboard Cite

show abstract

“…Despite the promising outcomes achieved so far, challenges still remain for the low efficiency of coordination between patients and professionals in IoT-assisted telerehabilitation, where substantial time delays for intervention planning could occur in data management and analytics that rely on professionals (Sarmento et al, 2019 ). In this respect, AIoT has emerged for more efficient IoT operations via enhanced data management and analytics with artificial intelligence (AI) algorithms (Lai et al, 2021 ). Despite the wide application of AIoT in smart cities, smart retail, and smart appliances, little has been done for implementing AIoT in telerehabilitation management, possibly owing to the immaturity of current theranostic systems for self-help rehabilitation (Lai et al, 2021 ).…”

Section: Automation In Rehabilitation Treatments With Coordination Be...mentioning

confidence: 99%

“…In this respect, AIoT has emerged for more efficient IoT operations via enhanced data management and analytics with artificial intelligence (AI) algorithms (Lai et al, 2021 ). Despite the wide application of AIoT in smart cities, smart retail, and smart appliances, little has been done for implementing AIoT in telerehabilitation management, possibly owing to the immaturity of current theranostic systems for self-help rehabilitation (Lai et al, 2021 ). For future automation in rehabilitation treatment, AIoT could be incorporated with novel point-of-care diagnostics and treatment devices to augment the efficiency of telerehabilitation management for real-time decision making in intervention planning.…”

Section: Automation In Rehabilitation Treatments With Coordination Be...mentioning

confidence: 99%

Automatic theranostics for long-term neurorehabilitation after stroke

Zhou,

Zhang,

Chen

et al. 2023

Front. Aging Neurosci.

View full text Add to dashboard Cite

“…For example, Lai et al achieved 99% accuracy in recognizing 6 lower limb exercises using one Inertial Measurement Unit (IMU). The IMU was attached to the knee for 4 exercises and instep for the other two [ 26 ]. García-de-Villa et al classified 8 exercises (5 lower limbs) with 96.2% accuracy [ 27 ].…”

Section: Introductionmentioning

confidence: 99%

Enhancing automated lower limb rehabilitation exercise task recognition through multi-sensor data fusion in tele-rehabilitation

Ettefagh,

Roshan Fekr

2024

BioMed Eng OnLine

View full text Add to dashboard Cite

Background Tele-rehabilitation is the provision of physiotherapy services to individuals in their own homes. Activity recognition plays a crucial role in the realm of automatic tele-rehabilitation. By assessing patient movements, identifying exercises, and providing feedback, these platforms can offer insightful information to clinicians, thereby facilitating an improved plan of care. This study introduces a novel deep learning approach aimed at identifying lower limb rehabilitation exercises. This is achieved through the integration of depth data and pressure heatmaps. We hypothesized that combining pressure heatmaps and depth data could improve the model’s overall performance. Methods In this study, depth videos and body pressure data from an accessible online dataset were used. This dataset comprises data from 30 healthy individuals performing 7 lower limb rehabilitation exercises. To accomplish the classification task, three deep learning models were developed, all based on an established 3D-CNN architecture. The models were designed to classify the depth videos, sequences of pressure data frames, and combination of depth videos and pressure frames. The models’ performance was assessed through leave-one-subject-out and leave-multiple-subjects-out cross-validation methods. Performance metrics, including accuracy, precision, recall, and F1 score, were reported for each model. Results Our findings indicated that the model trained on the fusion of depth and pressure data showed the highest and most stable performance when compared with models using individual modality inputs. This model could effectively identify the exercises with an accuracy of 95.71%, precision of 95.83%, recall of 95.71%, and an F1 score of 95.74%. Conclusion Our results highlight the impact of data fusion for accurately classifying lower limb rehabilitation exercises. We showed that our model could capture different aspects of exercise movements using the visual and weight distribution data from the depth camera and pressure mat, respectively. This integration of data provides a better representation of exercise patterns, leading to higher classification performance. Notably, our results indicate the potential application of this model in automatic tele-rehabilitation platforms.

show abstract

AIoT-Enabled Rehabilitation Recognition System—Exemplified by Hybrid Lower-Limb Exercises

Cited by 13 publications

References 52 publications

Classification of lower limb rehabilitation exercises with multiple and individual inertial measurement units

Classification of lower limb rehabilitation exercises with multiple and individual inertial measurement units

Automatic theranostics for long-term neurorehabilitation after stroke

Enhancing automated lower limb rehabilitation exercise task recognition through multi-sensor data fusion in tele-rehabilitation

Contact Info

Product

Resources

About