The Role of Deep Learning and Gait Analysis in Parkinson's Disease: A Systematic Review.

Parkinson's disease (PD) is the second most common movement disorder in the world. It is characterized by motor and non-motor symptoms that have a profound impact on the independence and quality of life of people affected by the disease, which increases caregivers' burdens. The use of the quantitative gait data of people with PD and deep learning (DL) approaches based on gait are emerging as increasingly promising methods to support and aid clinical decision making, with the aim of providing a quantitative and objective diagnosis, as well as an additional tool for disease monitoring. This will allow for the early detection of the disease, assessment of progression, and implementation of therapeutic interventions. In this paper, the authors provide a systematic review of emerging DL techniques recently proposed for the analysis of PD by using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The Scopus, PubMed, and Web of Science databases were searched across an interval of six years (between 2018, when the first article was published, and 2023). A total of 25 articles were included in this review, which reports studies on the movement analysis of PD patients using both wearable and non-wearable sensors. Additionally, these studies employed DL networks for classification, diagnosis, and monitoring purposes. The authors demonstrate that there is a wide employment in the field of PD of convolutional neural networks for analyzing signals from wearable sensors and pose estimation networks for motion analysis from videos. In addition, the authors discuss current difficulties and highlight future solutions for PD monitoring and disease progression.

Identification of a Gait Pattern for Detecting Mild Cognitive Impairment in Parkinson's Disease.

The aim of this study was to determine a gait pattern, i.e., a subset of spatial and temporal parameters, through a supervised machine learning (ML) approach, which could be used to reliably distinguish Parkinson's Disease (PD) patients with and without mild cognitive impairment (MCI). Thus, 80 PD patients underwent gait analysis and spatial-temporal parameters were acquired in three different conditions (normal gait, motor dual task and cognitive dual task). Statistical analysis was performed to investigate the data and, then, five ML algorithms and the wrapper method were implemented: Decision Tree (DT), Random Forest (RF), Naïve Bayes (NB), Support Vector Machine (SVM) and K-Nearest Neighbour (KNN). First, the algorithms for classifying PD patients with MCI were trained and validated on an internal dataset (sixty patients) and, then, the performance was tested by using an external dataset (twenty patients). Specificity, sensitivity, precision, accuracy and area under the receiver operating characteristic curve were calculated. SVM and RF showed the best performance and detected MCI with an accuracy of over 80.0%. The key features emerging from this study are stance phase, mean velocity, step length and cycle length; moreover, the major number of features selected by the wrapper belonged to the cognitive dual task, thus, supporting the close relationship between gait dysfunction and MCI in PD.

Circuital modelling in muscle tissue impedance measurements.

Electrical impedance spectroscopy (EIS) stands as a widely employed characterization technique for studying muscular tissue in both physio/pathological conditions. This methodology commonly involves modeling tissues through equivalent electrical circuits, facilitating a correlation between electrical parameters and physiological properties. Within existing literature, diverse equivalent electrical circuits have been proposed, varying in complexity and fitting properties. However, to date, none have definitively proven to be the most suiTable for tissue impedance measurements. This study aims to outline a systematic methodology for EIS measurements and to compare the performances of three widely used electrical circuits in characterizing both physiological and pathological muscle tissue conditions. Results highlight that, for optimal fitting with electrical parameters relevant to tissue characterization, the choice of the circuit to be fitted closely hinges on the specific measurement objectives, including measurement parameters and associated physiological features. Naturally, this necessitates a balance between simplicity and fitting accuracy.

Kinematic and Kinetic Gait Features Associated With Mild Cognitive Impairment in Parkinson's Disease.

Mild cognitive impairment (MCI) and gait deficits are commonly associated with Parkinson's disease (PD). Early detection of MCI associated with Parkinson's disease (PD-MCI) and its biomarkers is critical to managing disability in PD patients, reducing caregiver burden and healthcare costs. Gait is considered a surrogate marker for cognitive decline in PD. However, gait kinematic and kinetic features in PD-MCI patients remain unknown. This study was designed to explore the difference in gait kinematics and kinetics during single-task and dual-task walking between PD patients with and without MCI. Kinematic and kinetic data of 90 PD patients were collected using 3D motion capture system. Differences in gait kinematic and kinetic gait features between groups were identified by using: first, univariate statistical analysis and then a supervised machine learning analysis. The findings of this study showed that the presence of MCI in PD patients is coupled with kinematic and kinetic deviations of gait cycle which may eventually identify two different phenotypes of the disease. Indeed, as shown by the demographical and clinical comparison between the two groups, PD-MCI patients were older and more impaired. Moreover, PD-MCI kinematic results showed that cognitive dysfunction coexists with more severe axial symptoms and an increase postural flexion. A lack of physiological distal-to-proximal shift in joint kinetics was evidenced in the PD phenotype associated with cognitive impairments.

Wearable sensors for assessing disease severity and progression in Progressive Supranuclear Palsy.

INTRODUCTION: Progressive supranuclear palsy (PSP) is an atypical parkinsonism characterized by prominent gait and postural impairment. The PSP rating scale (PSPrs) is a clinician-administered tool to evaluate disease severity and progression. More recently, digital technologies have been used to investigate gait parameters. Therefore, object of this study was to implement a protocol using wearable sensors evaluating disease severity and progression in PSP. METHODS: Patients were evaluated with the PSPrs as well as with three wearable sensors located on the feet and lumbar area. Spearman coefficient was used to assess the relationship between PSPrs and quantitative measurements. Furthermore, sensor parameters were included in a multiple linear regression model to assess their ability in predicting the PSPrs total score and sub-scores. Finally, differences between baseline and three-month follow-up were calculated for PSPrs and each quantitative variable. The significance level in all analyses was set at ≤ 0.05. RESULTS: Fifty-eight evaluations from thirty-five patients were analyzed. Quantitative measurements showed multiple significant correlations with the PSPrs scores (r between 0.3 and 0.7; p < 0.05). Linear regression models confirmed the relationships. After three months visit, significant worsening from baseline was observed for cadence, cycle duration and PSPrs item 25, while PSPrs item 10 showed a significant improvement. CONCLUSION: We propose wearable sensors can provide an objective, sensitive quantitative evaluation and immediate notification of gait changes in PSP. Our protocol can be easily introduced in outpatient and research settings as a complementary tool to clinical measures as well as an informative tool on disease severity and progression in PSP.

Interplay between gait and neuropsychiatric symptoms in Parkinson's Disease.

Parkinson's Disease (PD) is a neurodegenerative disease which involves both motor and non-motor symptoms. Non-motor mental symptoms are very common among patients with PD since the earliest stage. In this context, gait analysis allows to detect quantitative gait variables to distinguish patients affected by non-motor mental symptoms from patients without these symptoms. A cohort of 68 PD subjects (divided in two groups) was acquired through gait analysis (single and double task) and spatial temporal parameters were analysed; first with a statistical analysis and then with a machine learning (ML) approach. Single-task variables showed that 9 out of 16 spatial temporal features were statistically significant for the univariate statistical analysis (p-value< 0.05). Indeed, a statistically significant difference was found in stance phase (p-value=0.032), swing phase (p-value=0.042) and cycle length (p-value=0.03) of the dual task. The ML results confirmed the statistical analysis, in particular, the Decision Tree classifier showed the highest accuracy (80.9%) and also the highest scores in terms of specificity and precision. Our findings indicate that patients with non-motor mental symptoms display a worse gait pattern, mainly dominated by increased slowness and dynamic instability.

Gait Analysis in Progressive Supranuclear Palsy Phenotypes.

The objective of the present study was to describe gait parameters of progressive supranuclear palsy (PSP) phenotypes at early stage verifying the ability of gait analysis in discriminating between disease phenotypes and between the other variant syndromes of PSP (vPSP) and Parkinson's disease (PD). Nineteen PSP (10 PSP-Richardson's syndrome, five PSP-parkinsonism, and four PSP-progressive gait freezing) and nine PD patients performed gait analysis in single and dual tasks. Although phenotypes showed similar demographic and clinical variables, Richardson's syndrome presented worse cognitive functions. Gait analysis demonstrated worse parameters in Richardson's syndrome compared with the vPSP. The overall diagnostic accuracy of the statistical model during dual task was almost 90%. The correlation analysis showed a significant relationship between gait parameters and visuo-spatial, praxic, and attention abilities in PSP-Richardson's syndrome only. vPSP presented worse gait parameters than PD. Richardson's syndrome presents greater gait dynamic instability since the earliest stages than other phenotypes. Computerized gait analysis can differentiate between PSP phenotypes and between vPSP and PD.

Duration of untreated illness as a predictor of treatment response and clinical course in generalized anxiety disorder.

INTRODUCTION: The aim of the present study was to investigate the impact of the duration of untreated illness (DUI)-defined as the time elapsing between the onset of generalized anxiety disorder (GAD) and the first adequate pharmacologic treatment-on treatment response and clinical course in a sample of subjects with GAD. METHODS: One hundred patients with GAD, diagnosed according to Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition-Text Revision criteria, were enrolled and their main demographic and clinical features collected. Patients were then treated with selective serotonin reuptake inhibitors or venlafaxine for 8 weeks in open-label conditions. Treatment response and other clinical variables were analyzed after dividing the sample into two groups according to DUI (DUI 12 months). RESULTS: When the DUI was computed with respect to the first antidepressant treatment (DUI-AD), a higher improvement (Clinical Global Impressions-Severity of Illness scale) after the pharmacologic treatment was found in the group with a shorter DUI (analysis of variance with repeated measures: time effect F=654.975, P<.001; group effect: F=4.369, P=.039). When computed with respect to the first treatment with benzodiazepines (DUI-BDZ), the two groups did not show any significant difference in treatment response (time effect: F=652.183, P<.001; group effect: F=0.009, P=.924). In addition, patients with a longer DUI (DUI-BDZ or DUI-AD) showed an earlier age at onset, a longer duration of illness and a higher rate of comorbid psychiatric disorders with onset later than GAD. CONCLUSION: Results from this preliminary study seem to suggest that a shorter DUI-AD may determine a better response to pharmacologic treatment in patients with GAD, and that a longer DUI (DUI-BDZ and DUI-AD) may be associated to a worse clinical course. Further investigation on the relationship between DUI and GAD is needed.

dSTAM expression pattern during wild type and mutant egg chamber development in D. melanogaster.

STAM (signal-transducing adaptor molecule) is a protein highly conserved from yeast to mammals. In Drosophila melanogaster the basic molecular architecture of the protein is comprised of a N-terminal VHS domain, an ubiquitin-interacting motif and a central Src homology-3 domain. In this paper we examine the expression pattern of the stam gene and the localisation of the STAM protein during D. melanogaster oogenesis. Its transcript is present throughout egg chamber development in all germ-line cells, including the oocyte. dSTAM is firstly detected in germarial region 2, where the protein is present in the newly formed germ-line cysts and is mainly accumulated into the oocyte. As oogenesis proceeds, dSTAM is enriched in the perinuclear region of the nurse cells and is also found in the somatic polar follicular cells. In the oocyte, the protein is more abundant posteriorly and becomes restricted to the posterior pole just before disappearing at stage 10b. We show that dSTAM localisation is unaffected in the oocyte of grk mutant egg chambers, indicating that it is not dependent on the polarity of the microtubule network. In contrast, dSTAM distribution is remarkably altered in cup mutant oocytes where the protein accumulates in a round central spot and never reaches the posterior pole.

Engaging psychiatrists in the diagnosis of psychogenic nonepileptic seizures. What can they contribute?

PURPOSE: To investigate if psychiatrists could predict the diagnosis of psychogenic nonepileptic seizures (PNES) by reviewing videos of seizures of various types and to compare the accuracy and the criteria leading to the diagnosis used by psychiatrists with those used by epileptologists. METHODS: Four board-certified psychiatrists were asked to review 23 videos capturing representative events of 21 unselected consecutive patients admitted to an epilepsy center for long-term video-EEG monitoring. All raters were blind to EEG and clinical information. They were requested to (1) rate the videos for quality and content; (2) choose among four diagnoses: (a) epileptic seizures; (b) PNES; (c) Other nonepileptic seizures (syncope, movement disorder, migraine, etc.); (d) "Cannot Say"; and (3) explain in their own words the main reasons leading to the diagnosis of choice. The results were compared to those of four blind epileptologists who independently reviewed the same cases. The inter-rater reliability was tested with the Kappa statistic. RESULTS: All psychiatrists were concordant and correct in 3/23 video-events, compared to 8/23 among epileptologists. Despite widespread disagreement among themselves and frequent failures as a group, individual psychiatrists scored a comparable number of correct diagnoses as did individual epileptologists. The comments provided to justify the diagnosis of choice differed from neurologists, varied among raters, and reflected considerable attention to body movements and body language. CONCLUSION: Psychiatrists, as a group, are less reliable than neurologists in differentiating seizure types on video but, as individuals, can be quite accurate in making the correct diagnosis because they are more attuned to capture the subtleties of human behaviour, of subjective experiences, as the effects of hidden internal conflicts and can contribute a new lexicon in defining PNES.

Clinical variables related to antidepressant-induced mania in bipolar disorder.

BACKGROUND: The development of mania or hypomania during antidepressant treatment is a serious complication of the clinical management of bipolar disorder (BP). The primary aim of this study was to evaluate the clinical variables related to antidepressant-induced mania or hypomania (AIM) in patients with BP. METHODS: DSM-IV BP-I or BP-II patients who had had at least one depressive episode treated with antidepressants were considered. Patients were subdivided into two groups according to the presence (n = 30) or absence (n = 106) of manic or hypomanic episodes occurring during antidepressant treatment. Possible predictive clinical variables of AIM were considered: gender, diagnostic subtype, age at onset, duration of illness, duration of untreated illness, type of antidepressant administered, number of previous spontaneous hypomanic or manic episodes, number of previous depressive episodes, presence of lifetime suicide attempts, presence of mood stabilizer treatments, presence of psychotic symptoms during spontaneous episodes, family history for psychiatric disorders in first degree relatives. Data were compared between the two groups, with (AIM+) and without (AIM-) antidepressant-induced mania, using Student's t tests and chi-square tests. RESULTS: The lack of mood stabilizer treatments during antidepressant therapy (chi-square = 37.602, df = 1, p < 0.001) and the exposure to tricyclic antidepressants (chi-square = 4.901, df = 1, p < 0.05) resulted significantly associated to the development of AIM. LIMITATIONS: This study was not done under controlled conditions and the relatively small sample studied warrants further replications. CONCLUSIONS: These results point out the risk of mania induction associated to the use of tricyclic antidepressants in BP patients, mainly in absence of adequate mood stabilizers.

Cigarette smoke condensate causes a decrease of the gene expression of Cu-Zn superoxide dismutase, Mn superoxide dismutase, glutathione peroxidase, catalase, and free radical-induced cell injury in SH-SY5Y human neuroblastoma cells.

Cigarette smoking condensate (CSC) contains oxidant compounds able to generate superoxide. The aim of the present study was to investigate the effect of the exposure to CSC on: (1) free radical production, (2) the gene expression of the antioxidant enzymes Cu-Zn superoxide dismutase (SOD1), Mn superoxide dismutase (SOD2), Glutathione Peroxidase (GPx), and catalase (CAT), and (3) cell survival in human neuroblastoma SH-SY5Y cells. The results showed that exposure (24 h) to different concentrations (10-150 µg/ml) of CSC caused a dose dependent cell injury that was coupled to the maximal increase of free radical production. These events were prevented by the addition to the incubation medium of the scavenger Vitamin E (50 µM). Furthermore, CSC exposure caused a reduction of the gene expression of the antioxidant enzymes SOD1, SOD2, GPx, and CAT that was counteracted by Vitamin E (50 µM). These results suggest that CSC exposure can induce a free radical overcharge that may be responsible for the inhibition of antioxidant enzymes expression and cell injury in SH-SY5Y human neuroblastoma cells. In fact the scavenger vitamin E can block both cell injury and inhibition of SOD1, SOD2, GPx, and CAT induced by CSC exposure.

Imatinib-mesylate blocks recombinant T-type calcium channels expressed in human embryonic kidney-293 cells by a protein tyrosine kinase-independent mechanism.

The 2-phenylaminopyrimidine derivative imatinib-mesylate, a powerful protein tyrosine kinase (PTK) inhibitor that targets abl, c-kit, and the platelet-derived growth factor receptors, is rapidly gaining a relevant role in the treatment of several types of neoplasms. Because first generation PTK inhibitors affect the activity of a large number of voltage-dependent ion channels, the present study explored the possibility that imatinib-mesylate could interfere with the activity of T-type channels, a class of voltage-dependent Ca2+ channels that take part in the chain of events elicited by PTK activation. The effect of the drug on T-type channel activity was examined using the whole-cell patch-clamp technique with Ba2+ (10 mM) as the permeant ion in human embryonic kidney-293 cells, stably expressing the rat Ca(V)3.3 channels. Imatinib-mesylate concentrations, ranging from 30 to 300 microM, reversibly decreased Ca(V)3.3 current amplitude with an IC(50) value of 56.9 microM. By contrast, when imatinib-mesylate (500 microM) was intracellularly dialyzed with the pipette solution, no reduction in Ba2+ current density was observed. The 2-phenylaminopyrimidine derivative modified neither the voltage dependence of activation nor the steady-state inactivation of Ca(V)3.3 channels. The decrease in extracellular Ba2+ concentration from 10 to 2 mM and the substitution of Ca2+ for Ba2+ increased the extent of 30 microM imatinib-mesylate-induced percentage of channel blockade from 25.9 +/- 2.4 to 36.3 +/- 0.9% in 2 mM Ba2+ and 44.2 +/- 2.3% in 2 mM Ca2+. In conclusion, imatinib-mesylate blocked the cloned Ca(V)3.3 channels by a PTK-independent mechanism. Specifically, the drug did not affect the activation or the inactivation of the channel but interfered with the ion permeation process.