A Novel Framework Based on Deep Learning Architecture for Continuous Human Activity Recognition with Inertial Sensors.

Frameworks for human activity recognition (HAR) can be applied in the clinical environment for monitoring patients' motor and functional abilities either remotely or within a rehabilitation program. Deep Learning (DL) models can be exploited to perform HAR by means of raw data, thus avoiding time-demanding feature engineering operations. Most works targeting HAR with DL-based architectures have tested the workflow performance on data related to a separate execution of the tasks. Hence, a paucity in the literature has been found with regard to frameworks aimed at recognizing continuously executed motor actions. In this article, the authors present the design, development, and testing of a DL-based workflow targeting continuous human activity recognition (CHAR). The model was trained on the data recorded from ten healthy subjects and tested on eight different subjects. Despite the limited sample size, the authors claim the capability of the proposed framework to accurately classify motor actions within a feasible time, thus making it potentially useful in a clinical scenario.

Biomechanical Effects of Using a Passive Exoskeleton for the Upper Limb in Industrial Manufacturing Activities: A Pilot Study.

This study investigates the biomechanical impact of a passive Arm-Support Exoskeleton (ASE) on workers in wool textile processing. Eight workers, equipped with surface electrodes for electromyography (EMG) recording, performed three industrial tasks, with and without the exoskeleton. All tasks were performed in an upright stance involving repetitive upper limbs actions and overhead work, each presenting different physical demands in terms of cycle duration, load handling and percentage of cycle time with shoulder flexion over 80°. The use of ASE consistently lowered muscle activity in the anterior and medial deltoid compared to the free condition (reduction in signal Root Mean Square (RMS) -21.6% and -13.6%, respectively), while no difference was found for the Erector Spinae Longissimus (ESL) muscle. All workers reported complete satisfaction with the ASE effectiveness as rated on Quebec User Evaluation of Satisfaction with Assistive Technology (QUEST), and 62% of the subjects rated the usability score as very high (>80 System Usability Scale (SUS)). The reduction in shoulder flexor muscle activity during the performance of industrial tasks is not correlated to the level of ergonomic risk involved. This preliminary study affirms the potential adoption of ASE as support for repetitive activities in wool textile processing, emphasizing its efficacy in reducing shoulder muscle activity. Positive worker acceptance and intention to use ASE supports its broader adoption as a preventive tool in the occupational sector.

Reliability of IMU-Derived Gait Parameters in Foot Drop Patients.

Foot drop is a deficit in foot dorsiflexion causing difficulties in walking. Passive ankle-foot orthoses are external devices used to support the drop foot improving gait functions. Foot drop deficits and therapeutic effects of AFO can be highlighted using gait analysis. This study reports values of the major spatiotemporal gait parameters assessed using wearable inertial sensors on a group of 25 subjects suffering from unilateral foot drop. Collected data were used to assess the test-retest reliability by means of Intraclass Correlation Coefficient and Minimum Detectable Change. Excellent test-retest reliability was found for all the parameters in all walking conditions. The analysis of Minimum Detectable Change identified the gait phases duration and the cadence as the most appropriate parameters to detect changes or improvements in subject gait after rehabilitation or specific treatment.

Effects of Ankle-Foot Orthosis on Balance of Foot Drop Patients.

Ankle-Foot Orthoses (AFOs) are common non-surgical treatments used to support foot and ankle joint when their normal functioning is compromised. AFOs have relevant impact on gait biomechanics, while scientific literature about effects on static balance is less strong and confusing. This study aims to assess the effectiveness of a plastic semi-rigid AFO in improving static balance on foot drop patients. Results underline that no significant effects on static balance is obtained on the study population when the AFO is used on the impaired foot.

Using Features Extracted from Upper Limb Reaching Tasks to Detect Parkinson's Disease by means of Machine Learning Models.

While in the literature there is much interest in investigating lower limbs gait of patients affected by neurological diseases, such as Parkinson's Disease (PD), fewer publications involving upper limbs movements are available. In previous studies, 24 motion signals (the so-called reaching tasks) of the upper limbs of PD patients and Healthy Controls (HCs) were used to extract several kinematic features through a custom-made software; conversely, the aim of our paper is to investigate the possibility to build models - using these features - for distinguishing PD patients from HCs. First, a binary logistic regression and, then, a Machine Learning (ML) analysis was performed by implementing five algorithms through the Knime Analytics Platform. The ML analysis was performed twice: first, a leave-one out-cross validation was applied; then, a wrapper feature selection method was implemented to identify the best subset of features that could maximize the accuracy. The binary logistic regression achieved an accuracy of 90.5%, demonstrating the importance of the maximum jerk during subjects upper limb motion; the Hosmer-Lemeshow test supported the validity of this model (p-value=0.408). The first ML analysis achieved high evaluation metrics by overcoming 95% of accuracy; the second ML analysis achieved a perfect classification with 100% of both accuracy and area under the curve receiver operating characteristics. The top-five features in terms of importance were the maximum acceleration, smoothness, duration, maximum jerk and kurtosis. The investigation carried out in our work has proved the predictive power of the features, extracted from the reaching tasks involving the upper limbs, to distinguish HCs and PD patients.

Gait asymmetry in stroke patients with unilateral spatial neglect.

The recovery of independent gait represents one of the main functional goals of the rehabilitative interventions after stroke but it can be hindered by the presence of unilateral spatial neglect (USN). The aim of the paper is to study if the presence of USN in stroke patients affects lower limb gait parameters between the two body sides, differently from what could be expected by the motor impairment alone, and to explore whether USN is associated to specific gait asymmetry. Thirty-five stroke patients (right or left lesion and ischemic or hemorrhagic etiology) who regained independent gait were assessed for global cognitive functioning and USN. All patients underwent a gait analysis session by using a wearable inertial system, kinematic parameters were computed. Enrolled patients presented altered motion parameters. Stroke patients with USN showed specific asymmetries in the following parameters: stance phase, swing phase, and knee range of motion. No differences in the clinical scores were found as the presence of USN. The presence of USN was associated with a specific form of altered gait symmetry. These findings may help clinicians to develop more tailored rehabilitative training to enhance gait efficacy of patients with motor defects complicated by the presence of selected cognitive impairments. Overview of the experiment setup. The workflow shows: diagnosis of unilateral spatial neglect by the neuropsychologist, sensors placement, gait analysis protocol and evaluation of the gait asymmetry together with the statistically significant features.

Machine learning can predict mild cognitive impairment in Parkinson's disease.

Background: Clinical markers of cognitive decline in Parkinson's disease (PD) encompass several mental non-motor symptoms such as hallucinations, apathy, anxiety, and depression. Furthermore, freezing of gait (FOG) and specific gait alterations have been associated with cognitive dysfunction in PD. Finally, although low cerebrospinal fluid levels of amyloid-ß42 have been found to predict cognitive decline in PD, hitherto PET imaging of amyloid-ß (Aß) failed to consistently demonstrate the association between Aß plaques deposition and mild cognitive impairment in PD (PD-MCI). Aim: Finding significant features associated with PD-MCI through a machine learning approach. Patients and methods: Patients were assessed with an extensive clinical and neuropsychological examination. Clinical evaluation included the assessment of mental non-motor symptoms and FOG using the specific items of the MDS-UPDRS I and II. Based on the neuropsychological examination, patients were classified as subjects without and with MCI (noPD-MCI, PD-MCI). All patients were evaluated using a motion analysis system. A subgroup of PD patients also underwent amyloid PET imaging. PD-MCI and noPD-MCI subjects were compared with a univariate statistical analysis on demographic data, clinical features, gait analysis variables, and amyloid PET data. Then, machine learning analysis was performed two times: Model 1 was implemented with age, clinical variables (hallucinations/psychosis, depression, anxiety, apathy, sleep problems, FOG), and gait features, while Model 2, including only the subgroup performing PET, was implemented with PET variables combined with the top five features of the former model. Results: Seventy-five PD patients were enrolled (33 PD-MCI and 42 noPD-MCI). PD-MCI vs. noPD-MCI resulted in older and showed worse gait patterns, mainly characterized by increased dynamic instability and reduced step length; when comparing amyloid PET data, the two groups did not differ. Regarding the machine learning analyses, evaluation metrics were satisfactory for Model 1 overcoming 80% for accuracy and specificity, whereas they were disappointing for Model 2. Conclusions: This study demonstrates that machine learning implemented with specific clinical features and gait variables exhibits high accuracy in predicting PD-MCI, whereas amyloid PET imaging is not able to increase prediction. Additionally, our results prompt that a data mining approach on certain gait parameters might represent a reliable surrogate biomarker of PD-MCI.

A Logistic Regression Model for Biomechanical Risk Classification in Lifting Tasks.

Lifting is one of the most potentially harmful activities for work-related musculoskeletal disorders (WMSDs), due to exposure to biomechanical risk. Risk assessment for work activities that involve lifting loads can be performed through the NIOSH (National Institute of Occupational Safety and Health) method, and specifically the Revised NIOSH Lifting Equation (RNLE). Aim of this work is to explore the feasibility of a logistic regression model fed with time and frequency domains features extracted from signals acquired through one inertial measurement unit (IMU) to classify risk classes associated with lifting activities according to the RNLE. Furthermore, an attempt was made to evaluate which are the most discriminating features relating to the risk classes, and to understand which inertial signals and which axis were the most representative. In a simplified scenario, where only two RNLE variables were altered during lifting tasks performed by 14 healthy adults, inertial signals (linear acceleration and angular velocity) acquired using one IMU placed on the subject's sternum during repeated rhythmic lifting tasks were automatically segmented to extract several features in the time and frequency domains. The logistic regression model fed with significant features showed good results to discriminate "risk" and "no risk" NIOSH classes with an accuracy, sensitivity and specificity equal to 82.8%, 84.8% and 80.9%, respectively. This preliminary work indicated that a logistic regression model-fed with specific inertial features extracted by signals acquired using a single IMU sensor placed on the sternum-is able to discriminate risk classes according to the RNLE in a simplified context, and therefore could be a valid tool to assess the biomechanical risk in an automatic way also in more complex conditions (e.g., real working scenarios).

A machine learning approach to characterize patients with asthma exacerbation attending an acute care setting.

BACKGROUND: One of the main problems in poorly controlled asthma is the access to the Emergency Department (ED). Using a machine learning (ML) approach, the aim of our study was to identify the main predictors of severe asthma exacerbations requiring hospital admission. METHODS: Consecutive patients with asthma exacerbation were screened for inclusion within 48 hours of ED discharge. A k-means clustering algorithm was implemented to evaluate a potential distinction of different phenotypes. K-Nearest Neighbor (KNN) as instance-based algorithm and Random Forest (RF) as tree-based algorithm were implemented in order to classify patients, based on the presence of at least one additional access to the ED in the previous 12 months. RESULTS: To train our model, we included 260 patients (31.5% males, mean age 47.6 years). Unsupervised ML identified two groups, based on eosinophil count. A total of 86 patients with eosinophiles ≥370 cells/µL were significantly older, had a longer disease duration, more restrictions to daily activities, and lower rate of treatment compared to 174 patients with eosinophiles <370 cells/µL. In addition, they reported lower values of predicted FEV1 (64.8±12.3% vs. 83.9±17.3%) and FEV1/FVC (71.3±9.3 vs. 78.5±6.8), with a higher amount of exacerbations/year. In supervised ML, KNN achieved the best performance in identifying frequent exacerbators (AUROC: 96.7%), confirming the importance of spirometry parameters and eosinophil count, along with the number of prior exacerbations and other clinical and demographic variables. CONCLUSIONS: This study confirms the key prognostic value of eosinophiles in asthma, suggesting the usefulness of ML in defining biological pathways that can help plan personalized pharmacological and rehabilitation strategies.

A Machine Learning Approach to Predict the Rehabilitation Outcome in Convalescent COVID-19 Patients.

BACKGROUND: After the acute disease, convalescent coronavirus disease 2019 (COVID-19) patients may experience several persistent manifestations that require multidisciplinary pulmonary rehabilitation (PR). By using a machine learning (ML) approach, we aimed to evaluate the clinical characteristics predicting the effectiveness of PR, expressed by an improved performance at the 6-min walking test (6MWT). METHODS: Convalescent COVID-19 patients referring to a Pulmonary Rehabilitation Unit were consecutively screened. The 6MWT performance was partitioned into three classes, corresponding to different degrees of improvement (low, medium, and high) following PR. A multiclass supervised classification learning was performed with random forest (RF), adaptive boosting (ADA-B), and gradient boosting (GB), as well as tree-based and k-nearest neighbors (KNN) as instance-based algorithms. RESULTS: To train and validate our model, we included 189 convalescent COVID-19 patients (74.1% males, mean age 59.7 years). RF obtained the best results in terms of accuracy (83.7%), sensitivity (84.0%), and area under the ROC curve (94.5%), while ADA-B reached the highest specificity (92.7%). CONCLUSIONS: Our model enables a good performance in predicting the rehabilitation outcome in convalescent COVID-19 patients.

Statistical Analysis and Kinematic Assessment of Upper Limb Reaching Task in Parkinson's Disease.

The impact of neurodegenerative disorders is twofold; they affect both quality of life and healthcare expenditure. In the case of Parkinson's disease, several strategies have been attempted to support the pharmacological treatment with rehabilitation protocols aimed at restoring motor function. In this scenario, the study of upper limb control mechanisms is particularly relevant due to the complexity of the joints involved in the movement of the arm. For these reasons, it is difficult to define proper indicators of the rehabilitation outcome. In this work, we propose a methodology to analyze and extract an ensemble of kinematic parameters from signals acquired during a complex upper limb reaching task. The methodology is tested in both healthy subjects and Parkinson's disease patients (N = 12), and a statistical analysis is carried out to establish the value of the extracted kinematic features in distinguishing between the two groups under study. The parameters with the greatest number of significances across the submovements are duration, mean velocity, maximum velocity, maximum acceleration, and smoothness. Results allowed the identification of a subset of significant kinematic parameters that could serve as a proof-of-concept for a future definition of potential indicators of the rehabilitation outcome in Parkinson's disease.

Reducing the Healthcare-Associated Infections in a Rehabilitation Hospital under the Guidance of Lean Six Sigma and DMAIC.

The reduction of healthcare-associated infections (HAIs) is one of the most important issues in the healthcare context for every type of hospital. In three operational units of the Scientific Clinical Institutes Maugeri SpA SB, a rehabilitation hospital in Cassano delle Murge (Italy), some corrective measures were introduced in 2017 to reduce the occurrence of HAIs. Lean Six Sigma was used together with the Define, Measure, Analyze, Improve, Control (DMAIC) roadmap to analyze both the impact of such measures on HAIs and the length of hospital stay (LOS) in the Rehabilitative Cardiology, Rehabilitative Neurology, Functional Recovery and Rehabilitation units in the Medical Center for Intensive Rehabilitation. The data of 2415 patients were analyzed, considering the phases both before and after the introduction of the measures. The hospital experienced a LOS reduction in both patients with and without HAIs; in particular, Cardiology had the greatest reduction for patients with infections (-7 days). The overall decrease in HAIs in the hospital was 3.44%, going from 169 to 121 cases of infections. The noteworthy decrease in LOS implies an increase in admissions and in the turnover indicator of the hospital, which has a positive impact on the hospital management as well as on costs.

The E-Textile for Biomedical Applications: A Systematic Review of Literature.

The use of e-textile technologies spread out in the scientific research with several applications in both medical and nonmedical world. In particular, wearable technologies and miniature electronics devices were implemented and tested for medical research purposes. In this paper, a systematic review regarding the use of e-textile for clinical applications was conducted: the Scopus and Pubmed databases were investigate by considering research studies from 2010 to 2020. Overall, 262 papers were found, and 71 of them were included in the systematic review. Of the included studies, 63.4% focused on information and communication technology studies, while the other 36.6% focused on industrial bioengineering applications. Overall, 56.3% of the research was published as an article, while the remainder were conference papers. Papers included in the review were grouped by main aim into cardiological, muscular, physical medicine and orthopaedic, respiratory, and miscellaneous applications. The systematic review showed that there are several types of applications regarding e-textile in medicine and several devices were implemented as well; nevertheless, there is still a lack of validation studies on larger cohorts of subjects since the majority of the research only focuses on developing and testing the new device without considering a further extended validation.

Extracting Features from Poincaré Plots to Distinguish Congestive Heart Failure Patients According to NYHA Classes.

Heart-rate variability has proved a valid tool in prognosis definition of patients with congestive heart failure (CHF). Previous research has documented Poincaré plot analysis as a valuable approach to study heart-rate variability performance among different subjects. In this paper, we explored the possibility to feed machine-learning (ML) algorithms using unconventional quantitative parameters extracted from Poincaré plots (generated from 24-h electrocardiogram recordings) to classify patients with CHF belonging to different New York Heart Association (NYHA) classes. We performed in sequence the following investigations: first, a statistical analysis was carried out on 9 morphological parameters, automatically measured from Poincaré plots. Subsequently, a feature selection through a wrapper with a 10-fold cross-validation method was performed to find the best subset of features which maximized the classification accuracy for each considered ML algorithm. Finally, patient classification was assessed through a ML analysis using AdaBoost of Decision Tree, k-Nearest Neighbors and Naive Bayes algorithms. A univariate statistical analysis proved 5 out of 9 parameters presented statistically significant differences among patients of distinct NYHA classes; similarly, a multivariate logistic regression confirmed the importance of the parameter ρy in the separability between low-risk and high-risk classes. The ML analysis achieved promising results in terms of evaluation metrics (especially the Naive Bayes algorithm), with accuracies greater than 80% and Area Under the Receiver Operating Curve indices greater than 0.7 for the overall three algorithms. The study indicates the proposed features have a predictive power to discriminate the NYHA classes, to which the features seem evenly correlated. Despite the NYHA classification being subjective and easily recognized by cardiologists, the potential relevance in the clinical cardiology of the proposed features and the promising ML results implies the methodology could be a valuable approach to automatically classify CHF. Future investigations on enriched datasets may further confirm the presented evidence.

Positive impact of short-term gait rehabilitation in Parkinson patients: a combined approach based on statistics and machine learning.

Parkinson's disease is the second most common neurodegenerative disorder in the world. Assumed that gait dysfunctions represent a major motor symptom for the pathology, gait analysis can provide clinicians quantitative information about the rehabilitation outcome of patients. In this scenario, wearable inertial systems for gait analysis can be a valid tool to assess the functional recovery of patients in an automatic and quantitative way, helping clinicians in decision making. Aim of the study is to evaluate the impact of the short-term rehabilitation on gait and balance of patients with Parkinson's disease. A cohort of 12 patients with Idiopathic Parkinson's disease performed a gait analysis session instrumented by a wearable inertial system for gait analysis: Opal System, by APDM Inc., with spatial and temporal parameters being analyzed through a statistic and machine learning approach. Six out of fourteen motion parameters exhibited a statistically significant difference between the measurements at admission and at discharge of the patients, while the machine learning analysis confirmed the separability of the two phases in terms of Accuracy and Area under the Receiving Operating Characteristic Curve. The rehabilitation treatment especially improved the motion parameters related to the gait. The study shows the positive impact on the gait of a short-term rehabilitation in patients with Parkinson's disease and the feasibility of the wearable inertial devices, that are increasingly spreading in clinical practice, to quantitatively assess the gait improvement.

Work-Related Risk Assessment According to the Revised NIOSH Lifting Equation: A Preliminary Study Using a Wearable Inertial Sensor and Machine Learning.

Many activities may elicit a biomechanical overload. Among these, lifting loads can cause work-related musculoskeletal disorders. Aspiring to improve risk prevention, the National Institute for Occupational Safety and Health (NIOSH) established a methodology for assessing lifting actions by means of a quantitative method based on intensity, duration, frequency and other geometrical characteristics of lifting. In this paper, we explored the machine learning (ML) feasibility to classify biomechanical risk according to the revised NIOSH lifting equation. Acceleration and angular velocity signals were collected using a wearable sensor during lifting tasks performed by seven subjects and further segmented to extract time-domain features: root mean square, minimum, maximum and standard deviation. The features were fed to several ML algorithms. Interesting results were obtained in terms of evaluation metrics for a binary risk/no-risk classification; specifically, the tree-based algorithms reached accuracies greater than 90% and Area under the Receiver operating curve characteristics curves greater than 0.9. In conclusion, this study indicates the proposed combination of features and algorithms represents a valuable approach to automatically classify work activities in two NIOSH risk groups. These data confirm the potential of this methodology to assess the biomechanical risk to which subjects are exposed during their work activity.

Gait analysis may distinguish progressive supranuclear palsy and Parkinson disease since the earliest stages.

Progressive supranuclear palsy (PSP) is a rare and rapidly progressing atypical parkinsonism. Albeit existing clinical criteria for PSP have good specificity and sensitivity, there is a need for biomarkers able to capture early objective disease-specific abnormalities. This study aimed to identify gait patterns specifically associated with early PSP. The study population comprised 104 consecutively enrolled participants (83 PD and 21 PSP patients). Gait was investigated using a gait analysis system during normal gait and a cognitive dual task. Univariate statistical analysis and binary logistic regression were used to compare all PD patients and all PSP patients, as well as newly diagnosed PD and early PSP patients. Gait pattern was poorer in PSP patients than in PD patients, even from early stages. PSP patients exhibited reduced velocity and increased measures of dynamic instability when compared to PD patients. Application of predictive models to gait data revealed that PD gait pattern was typified by increased cadence and longer cycle length, whereas a longer stance phase characterized PSP patients in both mid and early disease stages. The present study demonstrates that quantitative gait evaluation clearly distinguishes PSP patients from PD patients since the earliest stages of disease. First, this might candidate gait analysis as a reliable biomarker in both clinical and research setting. Furthermore, our results may offer speculative clues for conceiving early disease-specific rehabilitation strategies.

A multiple linear regression approach to extimate lifted load from features extracted from inertial data.

SUMMARY: Work-related musculoskeletal disorders are among the main occupational health problems. Substantial evidence has shown that work-related physical risk factors are the main source of low back complaints, particularly affecting heavy and repetitive manual lifting activities. The aim of the study is, during load lifting tasks, to explore the correlation between the time domain features extracted from the acceleration and angular velocity signals of the performing subject and the load lifted, and to explore the feasibility of a multiple linear regression model to predict the lifted load. The acceleration and angular velocity signals were acquired along the three directions of space by means of an inertial sensor placed on the subject's chest, during lifting activities with load gradually increased by 1 kg from 0 kg to 18 kg. Successively three time-domain features (Root Mean Square, Standard Deviation and MinMax value) were extracted from the acquired signals. First a correlation analysis was carried out between each individual feature and the load lifted (calculating r); then the time-domain features that proved most representative (strong correlation) were used to create a multiple linear regression model (calculating R-square). The statistical analysis was carried out by means of the Pearson correlation and multiple linear regression model was fed with the most informative time-domain features according to the correlation analysis. The correlation analysis showed a strong correlation (r > 0,7) between six features (three extracted from z-axes acceleration and three extracted from y-axes angular velocity) and the lifted load. The predictive multiple linear regression model, fed with these six features achieved a Rsquare greater than 0,9.The study demonstrated that the proposed combination of kinematic features and a multiple regression model represents a valid approach to automatically calculate the load lifted based on raw signals obtained by means of an inertial sensor placed on the chest. The results confirm the potential application of this methodology to indirectly monitor the load lifted by workers during their activity.

Machine learning to predict mortality after rehabilitation among patients with severe stroke.

Stroke is among the leading causes of death and disability worldwide. Approximately 20-25% of stroke survivors present severe disability, which is associated with increased mortality risk. Prognostication is inherent in the process of clinical decision-making. Machine learning (ML) methods have gained increasing popularity in the setting of biomedical research. The aim of this study was twofold: assessing the performance of ML tree-based algorithms for predicting three-year mortality model in 1207 stroke patients with severe disability who completed rehabilitation and comparing the performance of ML algorithms to that of a standard logistic regression. The logistic regression model achieved an area under the Receiver Operating Characteristics curve (AUC) of 0.745 and was well calibrated. At the optimal risk threshold, the model had an accuracy of 75.7%, a positive predictive value (PPV) of 33.9%, and a negative predictive value (NPV) of 91.0%. The ML algorithm outperformed the logistic regression model through the implementation of synthetic minority oversampling technique and the Random Forests, achieving an AUC of 0.928 and an accuracy of 86.3%. The PPV was 84.6% and the NPV 87.5%. This study introduced a step forward in the creation of standardisable tools for predicting health outcomes in individuals affected by stroke.