Biomechanical parameters assessment for the classification of Parkinson Disease using Bidirectional Long Short-Term Memory

Nowadays objective and efficient assessment of Parkinson Disease (PD) with machine learning techniques is a major focus for clinical management. This work presents a novel approach for classification of patients with PD (PwPD) and healthy controls (HC) using Bidirectional Long Short-Term Neural Network (BLSTM). In this paper, the SensHand and the SensFoot inertial wearable sensors for upper and lower limbs motion analysis were used to acquire motion data in thirteen tasks derived from the MDS-UPDRS III. Sixty-four PwPD and fifty HC were involved in this study. One hundred ninety extracted spatiotemporal and frequency parameters were applied as a single input against each subject to develop a recurrent BLSTM to discriminate the two groups. The maximum achieved accuracy was 82.4%, with the sensitivity of 92.3% and specificity of 76.2%. The obtained results suggest that the use of the extracted parameters for the development of the BLSTM contributed significantly to the classification of PwPD and HC.

Wearable Sensors for Prodromal Motor Assessment of Parkinson's Disease using Supervised Learning

Parkinson's disease (PD) is a common neurodegenerative disorder characterized by disabling motor and non-motor symptoms. Idiopathic hyposmia (IH), a reduced olfactory sensitivity, is a preclinical marker for the pathology and affects >95% of PD patients. In this paper, SensHand V1 and SensFoot V2, two inertial wearable sensors for upper and lower limbs, were developed to acquire motion data in ten tasks of the MDS-UPDRS III. Fifteen healthy subjects of control, 15 IH people, and 15 PD patients were enrolled. Seventy-one parameters per side were computed by spatiotemporal and frequency data analysis, and the most significant were selected to distinguish among the different classes. Performances of supervised learning algorithms (i.e., Support Vector Machine (SVM), and Random Forest (RF)) were compared on two-group and three-group classification and considering upper and lower limbs separately or together as a full system. Excellent results were obtained for healthy vs. patients classification (accuracy 1.00 for RF, and 0.97 for SVM), and good results were achieved by including IH subjects (0.92 F-measure with RF) within a three-group classification. Overall, the best performances were obtained using the full system with an RF classifier. The system is, thus, suitable to support an objective PD diagnosis. Furthermore, combining motion analysis with a validated olfactory screening test, people at risk for PD can be appropriately analyzed, and subtle changes in motor performance that characterize the prodromal phase and the early PD onset can be identified.

Fine Motor Assessment With Unsupervised Learning For Personalized Rehabilitation in Parkinson Disease

Parkinson disease (PD) is a common neurodegenerative disorders characterized by motor and non-motor impairments. Since the quality of life of PD patients becomes poor while pathology develops, it is imperative to improve the identification of personalized rehabilitation and treatments approaches based on the level of the neurodegeneration process. Objective and precise assessment of the severity of the pathology is crucial to identify the most appropriate treatments. In this context, this paper proposes a wearable system able to measure the motor performance of PD subjects. Two inertial devices were used to capture the motion of the lower and upper limbs respectively, while performing six motor tasks. Forty-one kinematic features were extracted from the inertial signals to describe the performance of each subjects. Three unsupervised learning algorithms (k-Means, Self-organizing maps (SOM) and hierarchical clustering) were applied with a blind approach to group the motor performance. The results show that SOM was the best classifier since it reached accuracy equal to 0.950 to group the instances in two classes (mild vs advanced), and 0.817 considering three classes (mild vs moderate vs severe). Therefore, this system enabled objective assessment of the PD severity through motion analysis, allowing the evaluation of residual motor capabilities and fostering personalized paths for PD rehabilitation and assistance.

Combining olfactory test and motion analysis sensors in Parkinson's disease preclinical diagnosis: a pilot study

OBJECTIVES

Preclinical diagnosis of Parkinson's disease (PD) is nowadays a topic of interest as the neuropathological process could begin years before the appearance of motor symptoms. Several symptoms, among them hyposmia, could precede motor features in PD. In the preclinical phase of PD, a subclinical reduction in motor skills is highly likely. In this pilot study, we investigate a step-by-step method to achieve preclinical PD diagnosis.

MATERIAL AND METHODS

We used the IOIT (Italian Olfactory Identification Test) to screen a population of healthy subjects. We identified 20 subjects with idiopathic hyposmia. Hyposmic subjects underwent an evaluation of motor skills, at baseline and after 1 year, using motion analysis sensors previously created by us.

RESULTS

One subject showed significant worsening in motor measurements. In this subject, we further conducted a dopaminergic challenge test monitored with the same sensors and, finally, he underwent [¹²³ I]-FP/CIT (DaTscan) SPECT brain imaging. The results show that he is probably affected by preclinical PD.

CONCLUSIONS

Our pilot study suggests that the combined use of an olfactory test and motor sensors for motion analysis could be useful for a screening of healthy subjects to identify those at a high risk of developing PD.

Mucus of the human olfactory epithelium contains the insulin-like growth factor-I system which is altered in some neurodegenerative diseases

Growth factors are believed to be involved in the mitotic regulation of the animal olfactory epithelium (OE). We investigated mucus covering the human OE area to see if it contained the insulin-like growth factor-I (IGF-I) and its binding proteins (IGFBPs) and to examine their behaviour in neurodegenerative diseases. Thirty patients with idiopathic late onset cerebellar ataxia (ILOCA), Parkinson's disease, and amyotrophic lateral sclerosis (ALS) and 30 age- and sex-matched healthy subjects were studied. In 10 controls, we also analyzed the mucus of the respiratory mucosa of the nose and tears. We detected IGF-I in the mucus covering the OE and Western ligand blot analysis (WLB) showed IGFBPs with an apparent Mr of 41, 500/38,500, 34,000 and 24,000, which were immunoprecipitated by specific antisera to IGFBP-3, -2 and -4, respectively. Their levels were higher than those observed in the respiratory mucosa of the nose or in tears. Mucus of the OE of the patients contained significantly reduced levels of IGF-I in comparison with those of controls. The intensity of all the IGFBPs-related bands were reduced in the ILOCA, while the remaining patients had a loss in the amounts of IGFBP-3. Plasma IGF-I and IGFBPs levels were similar in patients and controls. In conclusion, our data show that mucus covering the human OE contains IGF-I and IGFBPs, suggesting that these factors have a role in the activity of the OE. The amounts are reduced in the patients' mucus, possibly reflecting a dysfunction of the OE itself.

Electrophoretic pattern of physiological human nasal secretions

Various proteins have been detected in human nasal mucus, but their electrophoretic pattern has not been extensively investigated. Therefore, we have studied this pattern in nasal mucus samples from a group of 40 healthy subjects (20 males and 20 females). The electrophoretic separations have been performed under both native and denaturing conditions. The electrophoretic pattern of nasal mucus obtained under denaturing conditions and stained with Coomassie brilliant blue R-250, reveals 17 distinguishable areas into which protein bands are divided, with no significant variations from one individual to another. Eight of these areas contain high concentrations of nasal mucus proteins, whereas lower concentrations are seen in the other nine areas. The electrophoretic patterns of nasal mucus differs from that of eye mucus. The method used for sampling is rapid, simple, requires no local anaesthetic, and supplies an adequate quantity of proteins for laboratory testing. Electrophoresis under denaturing conditions proves to be the better of the two techniques, since it permits good separation of proteins. Among the protein bands revealed in our study there are some already known proteins (i.a. pre-albumim, albumin, lysozymne), but the presence of other proteins which have yet to be identified cannot be excluded. Hence, a study is underway for the purification and identification, by amino-acid sequencing, of the various proteins which make up each band. A through definition of the protein pattern of nasal mucus might prove useful for detecting anomalies in its composition, for example, in cases of olfactory disorders.

Clozapine in Huntington's chorea

In an open-label trial, we evaluated the efficacy of clozapine on abnormal involuntary movements in five patients with Huntington's chorea. We administered clozapine at increasing doses of 25, 50, and 150 mg/d for 3 weeks. Subjective self-evaluation of all patients reported reduction of chorea and improvement of daily living activities. At the end of the trial, all patients requested to continue with clozapine. Objective evaluation with the Abnormal Involuntary Movements Scale demonstrated in all patients moderate-to-marked reduction of abnormal involuntary movements without any significant side effects; the improvement was dose-dependent and markedly decreased 1 week after drug withdrawal.

[Dexamethasone therapy in Huntington chorea: preliminary results]

Neuroleptic drugs represent the current therapy for Huntington's chorea (HC). However neuroleptics can improve involuntary movements, but not functional performance and disease progression. Several clinical and experimental data suggest the existence of functional relationship between corticosteroids and extrapyramidal system. We administered dexamethasone to six choreics, all female. Dexamethasone was given i.m. at dose of 4 mg/die for 20 days and 8 mg/die for 20 days more. Dexamethasone at both the doses used, determined significant improvement (p less than 0.05) of dyskinesia, evaluated by AIMS, and manual dexterity, evaluated by Tapping test. Although at present it is not clear which mechanism are responsible for this of dexamethasone favourable effect, it might open new perspectives in HC therapy.

Descriptive epidemiologic study of epilepsy syndromes in a district of northwest Tuscany, Italy

To evaluate the prevalence of epilepsy syndromes in a district in Northwest Tuscany (Vecchiano, 9,952 inhabitants) we identified all suspected cases (probable epilepsy) from several information sources: files of local doctors, prescriptions of antiepileptic drugs (AEDs), hospital files from the center for epilepsy at the University of Pisa, medical files from university and hospital departments in Pisa, and files of the school doctors and social workers in the district of Vecchiano. All persons suspected of having epilepsy were examined; 51 cases of epilepsy were found, i.e., a prevalence rate of 5.1 in 1,000. During a previous study performed in the same district with the "house-to-house" method, a lower rate was found. Our study confirms that use of different epidemiologic investigational methods can produce different results.