Data-Driven Based Approach to Aid Parkinson's Disease Diagnosis.

This article presents a machine learning methodology for diagnosing Parkinson's disease (PD) based on the use of vertical Ground Reaction Forces (vGRFs) data collected from the gait cycle. A classification engine assigns subjects to healthy or Parkinsonian classes. The diagnosis process involves four steps: data pre-processing, feature extraction and selection, data classification and performance evaluation. The selected features are used as inputs of each classifier. Feature selection is achieved through a wrapper approach established using the random forest algorithm. The proposed methodology uses both supervised classification methods including K-nearest neighbour (K-NN), decision tree (DT), random forest (RF), Naïve Bayes (NB), support vector machine (SVM) and unsupervised classification methods such as K-means and the Gaussian mixture model (GMM). To evaluate the effectiveness of the proposed methodology, an online dataset collected within three different studies is used. This data set includes vGRF measurements collected from eight force sensors placed under each foot of the subjects. Ninety-three patients suffering from Parkinson's disease and 72 healthy subjects participated in the experiments. The obtained performances are compared with respect to various metrics including accuracy, precision, recall and F-measure. The classification performance evaluation is performed using the leave-one-out cross validation. The results demonstrate the ability of the proposed methodology to accurately differentiate between PD subjects and healthy subjects. For the purpose of validation, the proposed methodology is also evaluated with an additional dataset including subjects with neurodegenerative diseases (Amyotrophic Lateral Sclerosis (ALS) and Huntington's disease (HD)). The obtained results show the effectiveness of the proposed methodology to discriminate PD subjects from subjects with other neurodegenerative diseases with a relatively high accuracy.

Effect of Utilizing Cheek Retractors on Patient Satisfaction and Trueness of Peripheral Borders in Maxillary Digital Scans for Totally Edentulous Patients: An In Vivo Study.

PURPOSE: To evaluate the effect of cheek retractors on the accuracy of capturing peripheral borders in totally edentulous digital scans by comparing the conventional impression technique to digital scans made using two different cheek retractors. MATERIAL AND METHODS: Sixteen edentulous maxillary impressions were made using three techniques: the conventional impression technique, using modeling thermoplastic compound and zinc oxide eugenol paste; the digital intraoral scanning technique using the DIO scan retractor (DIO); and using the Br.nemark lip retractor (BRAN). The control impressions of each patient were poured, scanned using a desktop scanner, then transferred into a three-dimensional analysis software. DIO and BRAN groups were scanned using an intraoral scanner, imported, and superimposed using best fit algorithm on the corresponding control. The root mean square for the whole surface and for particular interest regions were calculated to assess the degree of trueness. The patients' perceptions of the impression techniques were the secondary outcomes. Statistical analyses were performed using the one sample T-test and Wilcoxon test (α=.05). RESULTS: Significant discrepancies were found for BRAN and DIO compared to the control. No significant discrepancies were found when comparing RMS of BRAN and DIO at different regions. Scan retractors had a significant impact on patient satisfaction, with patients preferring DIO. CONCLUSIONS: Edentulous intraoral scans made using cheek retractors had similar deviations when compared to each other but diverged from the conventional impression in edentulous maxilla. Patient preferences for intraoral scans over conventional impressions were confirmed. CLINICAL IMPLICATIONS: The use of different retracting methods during intraoral scanning of totally edentulous maxillary arches does not affect the peripheral border registration.

Three-dimensional radiographic assessment of immediate implant placement in the posterior mandible: A novel parameter-based classification.

OBJECTIVES: To assess the implant position (IP), the interradicular septum width (ISW) and radiographic bone to implant contact (BIC), when simulating an immediate implant placement in first and second mandibular molars. STUDY DESIGN: 75 patients (150 sites) were studied using cone beam computed tomography scans (CBCT) and computer aided design software. Implants were placed in a prosthetically driven position; subsequently, IP and BIC were digitally calculated. Linear ISW was also analyzed at 2, 4 and 6mm apically to the highest septal bony peak. Multiple linear regressions were performed to assess relationships between BIC and the different predictive variables. Additionally, the receiver operating characteristics (ROC) curve was used to create a model for BIC based on the ISW at 2mm. RESULTS: BIC in implants replacing first molars was the highest at the septal (S) position when compared to those in septal-mesial (S-M) position (p-value 0.001). As for the second molar, the highest percentage of BIC was recorded at the septal (S) position, followed by those in S-M and mesial (M) positions (p<0.001). CONCLUSION: According to the proposed classification, clinician must consider the ISW and IP when placing immediate implant in the first and second mandibular molar sites. When tackling first molars, S position is predominant, while S-M position is the most common in the second molars. ISW at 2mm should be at least respectively 2mm and 2.5mm at the first and second molar sites to achieve 50% of BIC.

Three-dimensional radiographic assessment of the mandibular retromolar donor site in different vertical facial growth types.

PURPOSE: To assess volumetrically, the impact of vertical facial growth types (VGFT) on the retromolar area as a bone donor site MATERIAL AND METHODS: 60 cone beam computed tomography (CBCT) scans of adult individuals were classified in three groups according to their SN-GoGn angle: hypodivergent group (hG) (33.33%), normodivergent group (NG) (30%) and hyperdivergent group (HG) (36.67%). Total harvestable bone volume and surface (TBV and TBS respectively), total cortical and cancellous bone volume (TCBV and TcBV respectively) and the percentage of cortical and cancellous bone volume (CBV and cBV respectively) were evaluated. RESULTS: The whole sample showed a mean TBV of 1220.99±448.81mm³ and a mean TBS of 940.29±259.93mm². Statistically significant differences were found between the different outcome variables and the vertical growth patterns (p<0.001). TBS differs for the different vertical growth patterns with the highest mean of TBS observed in the hG group. TBV also significantly differs between the different vertical growth patterns (p<0.001) with the highest mean observed for the hG individuals. Significant differences in percentages of cBV and CBV were present between the hyper-divergent groups and the other groups (p<0.001) with the hyper-divergent group having the lowest percentage of CBV and the highest percentage of cBV. CONCLUSION: hypodivergent individuals tend to have thicker bone blocks that can be used in onlay technique while thinner bone blocks harvested from hyperdivergent and normodivergent individuals can be used in three-dimensional grafting approach.

Genetically­modified stem cells in treatment of human diseases: Tissue kallikrein (KLK1)­based targeted therapy (Review).

The tissue kallikrein­kinin system (KKS) is an endogenous multiprotein metabolic cascade which is implicated in the homeostasis of the cardiovascular, renal and central nervous system. Human tissue kallikrein (KLK1) is a serine protease, component of the KKS that has been demonstrated to exert pleiotropic beneficial effects in protection from tissue injury through its anti­inflammatory, anti­apoptotic, anti­fibrotic and anti­oxidative actions. Mesenchymal stem cells (MSCs) or endothelial progenitor cells (EPCs) constitute populations of well­characterized, readily obtainable multipotent cells with special immunomodulatory, migratory and paracrine properties rendering them appealing potential therapeutics in experimental animal models of various diseases. Genetic modification enhances their inherent properties. MSCs or EPCs are competent cellular vehicles for drug and/or gene delivery in the targeted treatment of diseases. KLK1 gene delivery using adenoviral vectors or KLK1 protein infusion into injured tissues of animal models has provided particularly encouraging results in attenuating or reversing myocardial, renal and cerebrovascular ischemic phenotype and tissue damage, thus paving the way for the administration of genetically modified MSCs or EPCs with the human tissue KLK1 gene. Engraftment of KLK1­modified MSCs and/or KLK1­modified EPCs resulted in advanced beneficial outcome regarding heart and kidney protection and recovery from ischemic insults. Collectively, findings from pre­clinical studies raise the possibility that tissue KLK1 may be a novel future therapeutic target in the treatment of a wide range of cardiovascular, cerebrovascular and renal disorders.