Identification of Diseases Based on the Use of Inertial Sensors: A Systematic Review

Ponciano, Vasco; Pires, Ivan Miguel; Ribeiro, Fernando; Marques, Gonçalo; Villasana, María Vanessa; García, Nuno M.; Zdravevski, Eftim; Spinsante, Susanna

doi:10.3390/electronics9050778

Cited by 11 publications

(10 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Hence, they can be implemented as wearable devices (e.g., smartwatches, fitness bands, and smart clothing [9,10]). Because human health problems are most often expressed as measurable behaviors [11], IMUs are more suitable for daily activity data collection than other sensors. Hence, many IMUbased HAR studies have been accomplished [3][4][5]9].…”

Section: Introductionmentioning

confidence: 99%

Wearable IMU-Based Human Activity Recognition Algorithm for Clinical Balance Assessment Using 1D-CNN and GRU Ensemble Model

Kim

Kyunglim

Jeong

et al. 2021

Sensors

View full text Add to dashboard Cite

In this study, a wearable inertial measurement unit system was introduced to assess patients via the Berg balance scale (BBS), a clinical test for balance assessment. For this purpose, an automatic scoring algorithm was developed. The principal aim of this study is to improve the performance of the machine-learning-based method by introducing a deep-learning algorithm. A one-dimensional (1D) convolutional neural network (CNN) and a gated recurrent unit (GRU) that shows good performance in multivariate time-series data were used as model components to find the optimal ensemble model. Various structures were tested, and a stacking ensemble model with a simple meta-learner after two 1D-CNN heads and one GRU head showed the best performance. Additionally, model performance was enhanced by improving the dataset via preprocessing. The data were down sampled, an appropriate sampling rate was found, and the training and evaluation times of the model were improved. Using an augmentation process, the data imbalance problem was solved, and model accuracy was improved. The maximum accuracy of 14 BBS tasks using the model was 98.4%, which is superior to the results of previous studies.

show abstract

Section: Introductionmentioning

confidence: 99%

Wearable IMU-Based Human Activity Recognition Algorithm for Clinical Balance Assessment Using 1D-CNN and GRU Ensemble Model

Kim

Kyunglim

Jeong

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Entropy and statistical parameters such as mean value, standard deviation and variance are other sensible metrics [ 28 , 31 ]. Several features, extracted from both the time and frequency domain and different machine learning (ML) algorithms have been employed to classify FOG and pre-FOG events [ 36 ].…”

Section: Introductionmentioning

confidence: 99%

A Multi-Modal Analysis of the Freezing of Gait Phenomenon in Parkinson’s Disease

Mesin

Porcu

Russu

et al. 2022

Sensors

View full text Add to dashboard Cite

Background: Freezing of Gait (FOG) is one of the most disabling motor complications of Parkinson’s disease, and consists of an episodic inability to move forward, despite the intention to walk. FOG increases the risk of falls and reduces the quality of life of patients and their caregivers. The phenomenon is difficult to appreciate during outpatients visits; hence, its automatic recognition is of great clinical importance. Many types of sensors and different locations on the body have been proposed. However, the advantages of a multi-sensor configuration with respect to a single-sensor one are not clear, whereas this latter would be advisable for use in a non-supervised environment. Methods: In this study, we used a multi-modal dataset and machine learning algorithms to perform different classifications between FOG and non-FOG periods. Moreover, we explored the relevance of features in the time and frequency domains extracted from inertial sensors, electroencephalogram and skin conductance. We developed both a subject-independent and a subject-dependent algorithm, considering different sensor subsets. Results: The subject-independent and subject-dependent algorithms yielded accuracies of 85% and 88% in the leave-one-subject-out and leave-one-task-out test, respectively. Results suggest that the inertial sensors positioned on the lower limb are generally the most significant in recognizing FOG. Moreover, the performance impairment experienced when using a single tibial accelerometer instead of the optimal multi-modal configuration is limited to 2–3%. Conclusions: The achieved results disclose the possibility of getting a good FOG recognition using a minimally invasive set-up made of a single inertial sensor. This is very significant in the perspective of implementing a long-term monitoring of patients in their homes, during activities of daily living.

show abstract

“…However, it has been demonstrated that remote mobility measurement systems requiring the usage of a wearable sensor can have dropout rates of up to 32% after 6 months and 50% after a year [ 10 ]. Smartphones, ubiquitous, portable, user-friendly and affordable devices with one or several embedded inertial sensors, offer a promising new avenue for the analysis of the mobility performance of PD patients outside clinical environments, complementing the capacity measures obtained in-consultation and yielding rater-independent, quantitative outcomes with higher ecological validity [ 11 , 12 ]. The validity of smartphone applications for this purpose has been shown in healthy subjects and in PD patients, albeit mostly in supervised environments, requiring the device to be worn in a predetermined location and orientation, and not accounting for the heterogeneity of smartphone usage among the population [ 13 , 14 , 15 ].…”

Section: Introductionmentioning

confidence: 99%

Smartphone-Based Body Location-Independent Functional Mobility Analysis in Patients with Parkinson’s Disease: A Step towards Precise Medicine

Vila-Viçosa¹,

Leitão

Bouça‐Machado

et al. 2022

JPM

View full text Add to dashboard Cite

Ecological evaluation of gait using mobile technologies provides crucial information regarding the evolution of symptoms in Parkinson’s disease (PD). However, the reliability and validity of such information may be influenced by the smartphone’s location on the body. This study analyzed how the smartphone location affects the assessment of PD patients’ gait in a free-living environment. Twenty PD patients (mean ± SD age, 64.3 ± 10.6 years; 9 women (45%) performed 3 trials of a 250 m outdoor walk using smartphones in 5 different body locations (pants pocket, belt, hand, shirt pocket, and a shoulder bag). A method to derive gait-related metrics from smartphone sensors is presented, and its reliability is evaluated between different trials as well as its concurrent validity against optoelectronic and smartphone criteria. Excellent relative reliability was found with all intraclass correlation coefficient values above or equal to 0.85. High absolute reliability was observed in 21 out of 30 comparisons. Bland-Altman analysis revealed a high level of agreement (LoA between 4.4 and 17.5%), supporting the use of the presented method. This study advances the use of mobile technology to accurately and reliably quantify gait-related metrics from PD patients in free-living walking regardless of the smartphone’s location on the body.

show abstract

Identification of Diseases Based on the Use of Inertial Sensors: A Systematic Review

Cited by 11 publications

References 69 publications

Wearable IMU-Based Human Activity Recognition Algorithm for Clinical Balance Assessment Using 1D-CNN and GRU Ensemble Model

Wearable IMU-Based Human Activity Recognition Algorithm for Clinical Balance Assessment Using 1D-CNN and GRU Ensemble Model

A Multi-Modal Analysis of the Freezing of Gait Phenomenon in Parkinson’s Disease

Smartphone-Based Body Location-Independent Functional Mobility Analysis in Patients with Parkinson’s Disease: A Step towards Precise Medicine

Contact Info

Product

Resources

About