Francesco Infarinato

Background: One of the main goals of rehabilitation after stroke is the restoration of motor function. Understanding movement patterns and compensatory strategies is essential to optimize therapy. This study analyzes upper limb kinematics during the Box and Block Test (BBT) to identify and quantify typical post-stroke compensation strategies. Methods: Thirty-one sub-acute stroke participants and thirty-one healthy controls were included. Kinematic data were collected using a 7-IMU system. Joint angles were analyzed with MATLAB R2023a, and 3D trajectories were reconstructed from calibrated quaternions and anthropometric data. Group differences were assessed with the Mann–Whitney test. Compensation strategies were quantified in percentage terms relative to healthy subjects. Results: Significant intergroup differences were observed in mean joint angles and ranges of motion. On the paretic side, participants overused the wrist and shoulder to compensate for reduced elbow and trunk activity. Similar overuse was also observed on the unaffected side. Quantification showed that 83.9% and 80.6% compensate, respectively, with wrist and trunk and 67.7% with the shoulder. Conclusions: Using IMUs during the BBT, this study identified specific compensation strategies that may hinder recovery. It also contributed to developing a quantification scale, supporting more personalized rehabilitation and improved quality of life.

In recent years, the fusion of the medical and computer science domains has gained significant traction in the scientific research landscape. Progress in both fields has enabled the generation of a vast amount of data used for making predictions and identifying interesting clusters and pathways. The Machine Learning (ML) model’s application in the medical domain is one of the most compelling and challenging topics to explore, bridging the gap between Artificial Intelligence (AI) and healthcare. The combination of AI and medical information offers the possibility to create tools that can benefit both healthcare providers and physicians. This enables the enhancement of rehabilitation therapy and patient care. In the rehabilitation context, this work provides an alternative perspective: prediction of patients’ home discharge upon completing the rehabilitation protocol. Demographic and clinical data were collected on 7282 inpatients from electronic Medical Record, each record was categorized into Neurological Patients (NP, N = 3222) or Orthopedic Patients (OP, N = 4060). To identify the most suitable machine learning model, an extensive data preprocessing phase was conducted. This process involved variables recoding, scaling, and the evaluation of different dataset balancing methods to optimize model performance. Following a thorough review and comparison of algorithms commonly employed in the clinical-rehabilitative field, the Random Over Sampling (ROS) technique, in combination with the Random Forest (RF) machine learning model, was selected. Subsequently, a comprehensive hyperparameter tuning phase was performed using a grid search approach. The optimized model achieved an average accuracy of 98% for OP and 96% for NP, based on 10-fold cross-validation applied to the balanced training set (unrealistic scenario). When tested on the unbalanced dataset (real-world condition), the RF model maintained strong generalization performance, achieving 90% accuracy for OP and 83% for NP. This work points out the increasing importance of AI in medicine, especially in the realm of personalized rehabilitation. The use of such approaches could signify a transformative shift in healthcare. The integration of machine learning not only enhances the precision of treatment but also opens new possibilities for patient-centered care, improving outcomes and quality of care for individuals undergoing rehabilitation.

Frailty is a geriatric multidimensional syndrome characterized by a loss of physiologic reserves and disproportionate vulnerability to external stressors and associated with increased risk of multiple negative health outcomes. Since frailty can be prevented, controlled, and even reverted in its early stages, identifying the main factors involved in its development is crucial to implement preventive and/or restorative interventions. The aim of this study was to assess the impact of a broad range of parameters, including host factors, lifestyle, diet, and environmental and occupational conditions, on the development of frailty in later life. A cross-sectional study was conducted on 221,896 individuals aged 60 and over classified as non-frail (119,332, 53.8%), pre-frail (93,180, 42.0%), and frail (9384, 4.2%) according to the frailty phenotype. Using principal component analysis and machine learning to streamline the data, significant associations were found between frailty risk and air quality, diet, smoking, working conditions, and heavy alcohol consumption. Early-life factors, including breastfed as a baby and maternal smoking around birth, also emerged as predictors of frailty, which was further characterized by clinical indicators like polypharmacy, levels of C-reactive protein and other biomarkers of inflammageing. This study provided robust and original evidence on the association between a large battery of potential risk factors, from early to later stages of life, and the occurrence of frailty in older age. These results will contribute to the development of effective prevention strategies and facilitate the early detection of individuals at high risk of developing frailty.

The present exploratory study tested the hypothesis that computerized cognitive training (CCT) in home telemonitoring may beneficially affect eyes-closed resting-state electroencephalographic (rsEEG) rhythms in Parkinson's disease patients with cognitive deficits (PDCD). A Eurasian database provided clinical-demographic-rsEEG datasets in 40 PDCD patients, 29 PD patients without cognitive deficits (PDNCD), 40 Alzheimer’s disease patients with cognitive deficits (ADCD), and 40 cognitively normal older adults (Healthy). Sixteen of the 40 PDCD patients completed a cross-over unsupervised CCT program of simple, serious video games (versus a sham program) consisting of 14 daily sessions of approximately 20 minutes each in the patients' homes. Compared to the Healthy, PDNCD, and ADCD groups, the PDCD group was characterized by greater diffuse rsEEG delta (about 2-4 Hz) and theta (about 4-7 Hz) source activity. The PDCD patients who underwent the CCT program showed an improvement in video game performance and a reduction in these delta-theta source activities after this program compared to the control condition. In conclusion, these results suggest that the 2-week CCT program in home telemonitoring may mitigate the abnormal “slowing” of rsEEG rhythms in PDCD patients, possibly enhancing the regulation of brain arousal and quiet vigilance.

Background: In the last few years, new noninvasive strategies have emerged as rehabilitative treatments for patients with stroke. Action observation treatment (AOT) is a rehabilitation approach based on the properties of the mirror neuron system with a positive impact on modifying cortical activation patterns and improving the upper limb kinematics. AOT involves the dynamic process of observing purposeful actions with the intention of imitating and then practicing those actions. In recent years, several clinical studies suggested the effectiveness of AOT in patients with stroke to improve motor recovery and autonomy in activities of daily living. However, a deeper knowledge of the behavior of the sensorimotor cortex during AOT seems to be essential.Objective: The aim of this clinical trial, conducted in 2 neurorehabilitation centers and in patients’ homes, is to investigate the effectiveness of AOT in patients with stroke, confirming the translational power of a tailored treatment. Particular emphasis will be placed on the predictive value of neurophysiological biomarkers. In addition, the feasibility and impact of a home-based AOT program will be investigated.Methods: A 3-arm, assessor-blinded, randomized controlled trial will be performed by enrolling patients with stroke in the chronic stage. A total of 60 participants will be randomly allocated to receive 15 sessions of AOT with different protocols (AOT at the hospital, AOT at home, and sham AOT), 3 sessions per week. The primary outcome will be assessed using the Fugl-Meyer Assessment-Upper Extremity scores. Secondary outcomes will be clinical, biomechanical, and neurophysiological assessment.Results: The study protocol is part of a project (project code GR-2016–02361678) approved and funded by the Italian Ministry of Health. The study began with the recruitment phase in January 2022, and enrollment was expected to end in October 2022. Recruitment is now closed (December 2022). The results of this study are expected to be published in spring 2023. Upon completion of the analyses, we will examine the preliminary effectiveness of the intervention and neurophysiological outcomes.Conclusions: This study will be used to evaluate the effectiveness of 2 different AOT scenarios (ie, AOT at the hospital and AOT at home) in patients with chronic stroke and to assess the predictive value of neurophysiological biomarkers. Specifically, we will attempt to induce the functional modification of the cortical components by exploiting the features of the mirror neuron system, demonstrating relevant clinical, kinematic, and neurophysiological changes after AOT. With our study, we also want to provide, for the first time in Italy, the AOT home-based program while assessing its feasibility and impact.Trial Registration: ClinicalTrials.gov NCT04047134; https://clinicaltrials.gov/ct2/show/NCT04047134International Registered Report Identifier (IRRID): DERR1-10.2196/42094