Explainable Artificial Intelligence and Wearable Sensor-Based Gait Analysis to Identify Patients with Osteopenia and Sarcopenia in Daily Life

Balancing Performance and Interpretability in Medical Image Analysis: Case study of Osteopenia

Multiple studies within the medical field have highlighted the remarkable effectiveness of using convolutional neural networks for predicting medical conditions, sometimes even surpassing that of medical professionals. Despite their great performance, convolutional neural networks operate as black boxes, potentially arriving at correct conclusions for incorrect reasons or areas of focus. Our work explores the possibility of mitigating this phenomenon by identifying and occluding confounding variables within images. Specifically, we focused on the prediction of osteopenia, a serious medical condition, using the publicly available GRAZPEDWRI-DX dataset. After detection of the confounding variables in the dataset, we generated masks that occlude regions of images associated with those variables. By doing so, models were forced to focus on different parts of the images for classification. Model evaluation using F1-score, precision, and recall showed that models trained on non-occluded images typically outperformed models trained on occluded images. However, a test where radiologists had to choose a model based on the focused regions extracted by the GRAD-CAM method showcased different outcomes. The radiologists' preference shifted towards models trained on the occluded images. These results suggest that while occluding confounding variables may degrade model performance, it enhances interpretability, providing more reliable insights into the reasoning behind predictions. The code to repeat our experiment is available on the following link: https://github.com/mikulicmateo/osteopenia .

Bidirectional mediation of bone mineral density and brain atrophy on their associations with gait variability

This mediation analysis aimed to investigate the associations among areal bone mineral density, mobility-related brain atrophy, and specific gait patterns. A total of 595 participants from the Taizhou Imaging Study, who underwent both gait and bone mineral density measurements, were included in this cross-sectional analysis. We used a wearable gait tracking device to collect quantitative gait parameters and then summarized them into independent gait domains with factor analysis. Bone mineral density was measured in the lumbar spine, femoral neck, and total hip using dual-energy X-ray absorptiometry. Magnetic resonance images were obtained on a 3.0-Tesla scanner, and the volumes of brain regions related to mobility were computed using FreeSurfer. Lower bone mineral density was found to be associated with higher gait variability, especially at the site of the lumbar spine (β = 0.174, FDR = 0.001). Besides, higher gait variability was correlated with mobility-related brain atrophy, like the primary motor cortex (β = 0.147, FDR = 0.006), sensorimotor cortex (β = 0.153, FDR = 0.006), and entorhinal cortex (β = 0.106, FDR = 0.043). Bidirectional mediation analysis revealed that regional brain atrophy contributed to higher gait variability through the low lumbar spine bone mineral density (for the primary motor cortex, P = 0.018; for the sensorimotor cortex, P = 0.010) and the low lumbar spine bone mineral density contributed to higher gait variability through the primary motor and sensorimotor cortices (P = 0.026 and 0.010, respectively).

Exploration and demonstration of explainable machine learning models in prosthetic rehabilitation-based gait analysis

Quantitative gait analysis is important for understanding the non-typical walking patterns associated with mobility impairments. Conventional linear statistical methods and machine learning (ML) models are commonly used to assess gait performance and related changes in the gait parameters. Nonetheless, explainable machine learning provides an alternative technique for distinguishing the significant and influential gait changes stemming from a given intervention. The goal of this work was to demonstrate the use of explainable ML models in gait analysis for prosthetic rehabilitation in both population- and sample-based interpretability analyses. Models were developed to classify amputee gait with two types of prosthetic knee joints. Sagittal plane gait patterns of 21 individuals with unilateral transfemoral amputations were video-recorded and 19 spatiotemporal and kinematic gait parameters were extracted and included in the models. Four ML models-logistic regression, support vector machine, random forest, and LightGBM-were assessed and tested for accuracy and precision. The Shapley Additive exPlanations (SHAP) framework was applied to examine global and local interpretability. Random Forest yielded the highest classification accuracy (98.3%). The SHAP framework quantified the level of influence of each gait parameter in the models where knee flexion-related parameters were found the most influential factors in yielding the outcomes of the models. The sample-based explainable ML provided additional insights over the population-based analyses, including an understanding of the effect of the knee type on the walking style of a specific sample, and whether or not it agreed with global interpretations. It was concluded that explainable ML models can be powerful tools for the assessment of gait-related clinical interventions, revealing important parameters that may be overlooked using conventional statistical methods.

Using In-Shoe Inertial Measurement Unit Sensors to Understand Daily-Life Gait Characteristics in Patients With Distal Radius Fractures During 6 Months of Recovery: Cross-Sectional Study

BACKGROUND

A distal radius fracture (DRF) is a common initial fragility fracture among women in their early postmenopausal period, which is associated with an increased risk of subsequent fractures. Gait assessments are valuable for evaluating fracture risk; inertial measurement units (IMUs) have been widely used to assess gait under free-living conditions. However, little is known about long-term changes in patients with DRF, especially concerning daily-life gait. We hypothesized that, in the long term, the daily-life gait parameters in patients with DRF could enable us to reveal future risk factors for falls and fractures.

OBJECTIVE

This study assessed the spatiotemporal characteristics of patients with DRF at 4 weeks and 6 months of recovery.

METHODS

We recruited 16 women in their postmenopausal period with DRF as their first fragility fracture (mean age 62.3, SD 7.0 years) and 28 matched healthy controls (mean age 65.6, SD 8.0 years). Daily-life gait assessments and physical assessments, such as hand grip strength (HGS), were performed using an in-shoe IMU sensor. Participants' results were compared with those of the control group, and their recovery was assessed for 6 months after the fracture.

RESULTS

In the fracture group, at 4 weeks after DRF, lower foot height in the swing phase (P=.049) and higher variability of stride length (P=.03) were observed, which improved gradually. However, the dorsiflexion angle in the fracture group tended to be lower consistently during 6 months (at 4 weeks: P=.06; during 6 months: P=.07). As for the physical assessments, the fracture group showed lower HGS at all time points (at 4 weeks: P<.001; during 6 months: P=.04), despite significant improvement at 6 months (P<.001).

CONCLUSIONS

With an in-shoe IMU sensor, we discovered the recovery of spatiotemporal gait characteristics 6 months after DRF surgery without the participants' awareness. The consistently unchanged dorsiflexion angle in the swing phase and lower HGS could be associated with fracture risk, implying the high clinical importance of appropriate interventions for patients with DRF to prevent future fractures. These results could be applied to a screening tool for evaluating the risk of falls and fractures, which may contribute to constructing a new health care system using wearable devices in the near future.

Sarcopenia classification model for musculoskeletal patients using smart insole and artificial intelligence gait analysis

BACKGROUND

The relationship between physical function, musculoskeletal disorders and sarcopenia is intricate. Current physical function tests, such as the gait speed test and the chair stand test, have limitations in eliminating subjective influences. To overcome this, smart devices utilizing inertial measurement unit sensors and artificial intelligence (AI)-based methods are being developed.

METHODS

We employed cutting-edge technologies, including the smart insole device and pose estimation based on AI, along with three classification models: random forest (RF), support vector machine and artificial neural network, to classify control and sarcopenia groups. Patient data of 83 individuals were divided into train and test sets, with approximately 67% allocated for training. Classification models were implemented using RStudio, considering individual and combined variables obtained through pose estimation and smart insole measurements.

RESULTS

Performance evaluation of the classification models utilized accuracy, precision, recall and F1-score indicators. Using only pose estimation variables, accuracy ranged from 0.92 to 0.96, with F1-scores of 0.94-0.97. Key variables identified by the RF model were 'Hip_dif', 'Ankle_dif' and 'Hipankle_dif'. Combining variables from both methods increased accuracy to 0.80-1.00, with F1-scores of 0.73-1.00.

CONCLUSIONS

In our study, a classification model that integrates smart insole and pose estimation technology was assessed. The RF model showed impressive results, particularly in the case of the Hip and Ankle variables. The growth of advanced measurement technologies suggests a promising avenue for identifying and utilizing additional digital biomarkers in the management of various disorders. The convergence of AI technologies with diagnostics and treatment approaches a promising future for enhanced interventions in conditions like sarcopenia.

Machine Learning Applications in Sarcopenia Detection and Management: A Comprehensive Survey

This extensive review examines sarcopenia, a condition characterized by a loss of muscle mass, stamina, and physical performance, with a particular emphasis on its detection and management using contemporary technologies. It highlights the lack of global agreement or standardization regarding the definition of sarcopenia and the various techniques used to measure muscle mass, stamina, and physical performance. The distinctive criteria employed by the European Working Group on Sarcopenia in Older People (EWGSOP) and the Asian Working Group for Sarcopenia (AWGSOP) for diagnosing sarcopenia are examined, emphasizing potential obstacles in comparing research results across studies. The paper delves into the use of machine learning techniques in sarcopenia detection and diagnosis, noting challenges such as data accessibility, data imbalance, and feature selection. It suggests that wearable devices, like activity trackers and smartwatches, could offer valuable insights into sarcopenia progression and aid individuals in monitoring and managing their condition. Additionally, the paper investigates the potential of blockchain technology and edge computing in healthcare data storage, discussing models and systems that leverage these technologies to secure patient data privacy and enhance personal health information management. However, it acknowledges the limitations of these models and systems, including inefficiencies in handling large volumes of medical data and the lack of dynamic selection capability. In conclusion, the paper provides a comprehensive summary of current sarcopenia research, emphasizing the potential of modern technologies in enhancing the detection and management of the condition while also highlighting the need for further research to address challenges in standardization, data management, and effective technology use.

Assessing physical abilities of sarcopenia patients using gait analysis and smart insole for development of digital biomarker

The aim of this study is to compare variable importance across multiple measurement tools, and to use smart insole and artificial intelligence (AI) gait analysis to create variables that can evaluate the physical abilities of sarcopenia patients. By analyzing and comparing sarcopenia patients with non sarcopenia patients, this study aims to develop predictive and classification models for sarcopenia and discover digital biomarkers. The researchers used smart insole equipment to collect plantar pressure data from 83 patients, and a smart phone to collect video data for pose estimation. A Mann-Whitney U was conducted to compare the sarcopenia group of 23 patients and the control group of 60 patients. Smart insole and pose estimation were used to compare the physical abilities of sarcopenia patients with a control group. Analysis of joint point variables showed significant differences in 12 out of 15 variables, but not in knee mean, ankle range, and hip range. These findings suggest that digital biomarkers can be used to differentiate sarcopenia patients from the normal population with improved accuracy. This study compared musculoskeletal disorder patients to sarcopenia patients using smart insole and pose estimation. Multiple measurement methods are important for accurate sarcopenia diagnosis and digital technology has potential for improving diagnosis and treatment.

DPReLU: Dynamic Parametric Rectified Linear Unit and Its Proper Weight Initialization Method

Activation functions are essential in deep learning, and the rectified linear unit (ReLU) is the most widely used activation function to solve the vanishing gradient problem. However, owing to the dying ReLU problem and bias shift effect, deep learning models using ReLU cannot exploit the potential benefits of negative values. Numerous ReLU variants have been proposed to address this issue. In this study, we propose Dynamic Parametric ReLU (DPReLU), which can dynamically control the overall functional shape of ReLU with four learnable parameters. The parameters of DPReLU are determined by training rather than by humans, thereby making the formulation more suitable and flexible for each model and dataset. Furthermore, we propose an appropriate and robust weight initialization method for DPReLU. To evaluate DPReLU and its weight initialization method, we performed two experiments on various image datasets: one using an autoencoder for image generation and the other using the ResNet50 for image classification. The results show that DPReLU and our weight initialization method provide faster convergence and better accuracy than the original ReLU and the previous ReLU variants.

Intelligent Biosignal Processing in Wearable and Implantable Sensors

Wearable technology including sensors, sensor networks, and the associated devices have opened up space in a variety of applications [...].