1,721,053 research outputs found
A Comparison of Performances of Different Feature Selection Methods applied to Biomedical Data
Full text linkMigraine is a debilitating disease whose causes are not yet completely explained. Near-InfraRed Spectroscopy (NIRS) is a non-invasive technology commonly used for the assessment of the cerebral autoregulation during active stimuli.
Feature Selection (FS) allows dimensionality reduction of multivariate datasets, highlighting the most informative variables and deleting redundant and irrelevant information. Rough Set Theory (RST) is one of the most used tool for FS, enables to manage incomplete and imperfect knowledge without any assumption about data model.
This study involved a total of 80 subjects, divided in 3 groups: 15 healthy subjects taken as controls, 14 women suffered from migraine without aura and 51 women from migraine with aura. We apply three different methods of FS based on RST to a set of 26 parameters extracted from NIRS signals recorded in the subjects during breath-holding (BH) and hyperventilation (HYP). We compare the extracted subsets of features in the subjects’ classification by means of Artificial Neural Networks. The results show good performance for all subsets, with a percentage of correct classification above the 90%
Automatic carotid segmentation based on pixels classification
No full textCarotid artery (CA) intima-media thickness (IMT) is commonly deemed as one of the risk marker for cardiovascular diseases. The automatic estimation of the IMT on ultrasound images is based on the correct identification of the lumen-intima (LI) and media-adventitia (MA) interfaces. This task is complicated by noise, different morphology and pathology of the carotid artery. In a previous study, we applied four non-linear methods for feature selection on a set of variables extracted from ultrasound carotid images. The main aim was to select those parameters containing the highest amount of information useful to classify the image pixels in the carotid regions they belong to. In this study we present a pixel classifier based on the selected features. Once the pixels classification was correctly performed, the LI and MA interfaces were extracted and compared with two sets of manual-traced profile
Rough Set Based Approach for IMT Automatic Estimation
Full text linkCarotid artery (CA) intima-media thickness (IMT) is commonly deemed as one of the risk marker for cardiovascular diseases. The automatic estimation of the IMT on ultrasound images is based on the correct identification of the lumen-intima (LI) and media-adventitia (MA) interfaces. This task is complicated by noise, vessel morphology and pathology of the carotid artery. In a previous study we applied four non-linear methods for feature selection on a set of variables extracted from ultrasound carotid images. The main aim was to select those parameters containing the highest amount of information useful to classify the image pixels in the carotid regions they belong to. In this study we present a pixel classifier based on the selected features. Once the pixels classification was correctly performed, the IMT was evaluated and compared with two sets of manual-traced profiles. The results showed that the automatic IMTs are not statistically different from the manual one
FEATURE SELECTION APPLIED TO THE TIME-FREQUENCY REPRESENTATION OF MUSCLE NEAR-INFRARED SPECTROSCOPY (NIRS) SIGNALS: CHARACTERIZATION OF DIABETIC OXYGENATION PATTERNS
Full text linkDiabetic patients might present peripheral microcirculation impairment and might benefit from physical training. Thirty-nine diabetic patients underwent the monitoring of the tibialis anterior muscle oxygenation during a series of voluntary ankle flexo-extensions by near-infrared spectroscopy (NIRS). NIRS signals were acquired before and after training protocols. Sixteen control subjects were tested with the same protocol. Time-frequency distributions of the Cohen's class were used to process the NIRS signals relative to the concentration changes of oxygenated and reduced hemoglobin. A total of 24 variables were measured for each subject and the most discriminative were selected by using four feature selection algorithms: QuickReduct, Genetic Rough-Set Attribute Reduction, Ant Rough-Set Attribute Reduction, and traditional ANOVA. Artificial neural networks were used to validate the discriminative power of the selected features. Results showed that different algorithms extracted different sets of variables, but all the combinations were discriminative. The best classification accuracy was about 70%. The oxygenation variables were selected when comparing controls to diabetic patients or diabetic patients before and after training. This preliminary study showed the importance of feature selection techniques in NIRS assessment of diabetic peripheral vascular impairmen
A Multi-Agent System for Monitoring Patient Flow
No full textPatient flow within a healthcare facility may follow different and, sometimes, complicated paths. Each path phase is associated with the documentation of the activities carried out during it and may require the consultation of clinical guidelines, medical literature and the use of specific software and decision aid systems. In this study we present the design of a Patient Flow Management System (PFMS) based on MultiAgent Systems (MAS) methodology. System requirements were identified by means of process modeling tools and a MAS consisting of six agents was designed and is under construction. Its main goal is to support both the medical staff during the health care process and the hospital managers in assuring that all the required documentation is completed and available. Moreover, such a tool can be used for the assessment and comparison of different clinical pathways, in order to identify possible improvements and the optimum patient flo
Fuzzy logic applied to a Patient Classification System
No full textThe optimization of the clinical staff resources is a very complicated task that can be supported by a set of tools called Patient Classification Systems (PCS). These methods allow the evaluation of the correct number of nurses and healthcare workers needed in order to guarantee an appropriate care level.
In this study a PCS tool called MAP is presented, able both to support the staff allocation and to assess the complexity level of each single patient. This method, applicable in all clinical fields, is based on the analysis of the patient state by means of a set of physiological variables characterizing his clinical conditions and his environment.
Moreover, we introduce an evolution of MAP based on Fuzzy Logic, in order to produce an instrument more suitable to the daily clinical applications
Surface Electromyography Applied to Gait Analysis: How to Improve Its Impact in Clinics?
Full text linkSurface electromyography (sEMG) is the main non-invasive tool used to record the electrical activity of muscles during dynamic tasks. In clinical gait analysis, a number of techniques have been developed to obtain and interpret the muscle activation patterns of patients showing altered locomotion. However, the body of knowledge described in these studies is very seldom translated into routine clinical practice. The aim of this work is to analyze critically the key factors limiting the extensive use of these powerful techniques among clinicians. A thorough understanding of these limiting factors will provide an important opportunity to overcome limitations through specific actions, and advance toward an evidence-based approach to rehabilitation based on objective findings and measurements
Developing Medical Device Software in compliance with regulations
No full textIn the last decade, the use of information technology (IT) in healthcare has taken a growing role. In fact, the adoption of an increasing number of computer tools has led to several benefits related to the process of patient care and allowed easier access to social and health care resources. At the same time this trend gave rise to new challenges related to the implementation of these new technologies. Software used in healthcare can be classified as medical devices depending on the way they are used and on their functional characteristics. If they are classified as medical devices they must satisfy specific regulations. The aim of this work is to present a software development framework that can allow the production of safe and high quality medical device software and to highlight the correspondence between each software development phase and the appropriate standard and/or regulatio
Comparison of Hierarchical and Partitional Clustering in Multi-Source Phonocardiography
Full text linkPhonocardiography (PCG) has proved a valuable tool over the years to monitor the status of at-risk patients for some cardiovascular diseases. Its multi-source version, consisting of the simultaneous recording of multiple acoustic signals from different points of the patient’s chest, is currently under research as a solution to develop wearable devices based on PCG and bring PCG to the patient’s domicile. When a high number of PCG signals are available, to define the most suitable auscultation area, depending on the clinical question, clustering comes into the picture. In this work, we applied agglomerative hierarchical clustering and k-means to multi-source PCG recordings. A similarity metrics based on the correlation of the signals was used to compare the signals based on their morphological characteristics. The two clustering methods resulted in a Rand Index averagely higher than 0.84, showing a high level of agreement and validating the usage of clustering for the application of interest. Hierarchical clustering allowed for obtaining a better trade-off between the intra-cluster variability and the inter-cluster distance. Adding to its deterministic nature, it should be considered as preferrable with respect to k-means. This work moves one step further to the development a reliable wearable device based on digital auscultation for the monitoring of the patient at its domicile
Specificity improvement of a CAD system for multiparametric MR prostate cancer using texture features and artificial neural networks
Full text linkProstate cancer (PCa) is the most common cancer afflicting men in USA. Multiparametric Magnetic Resonance imaging is recently emerging as a powerful tool for PCa diagnosis, but its analysis and interpretation is time-consuming and affected by the radiologist experience. Computer aided detection (CAD) systems have been developed to overcome this limitation and to support radiologists in the PCa diagnosis. Although several studies proposed CAD systems with very high performances in terms of sensitivity, the analysis of false positive (FP) areas is usually not clearly presented. The aim of this study is to improve the performance of a CAD system in term of reduction of FPs findings, without affecting the sensitivity. To this scope, we developed a classifier composed by 3 Artificial Neural Networks (ANN) able to distinguish between malignant and healthy areas through a voting strategy. In this method, we exploit the role of the Gray Level Co-occurrence Matrix, the Gray Level Difference Method and Gray Level Run Length Method Matrix in differentiating tumoural from healthy tissues. We first extract 64 textural features from T2-weighted (T2w) images and the apparent diffusion coefficient (ADC) maps, then we discretized them to reduce the data variability. A features selection method, based on the correlation matrix, is finally applied to remove redundant variables, that are those highly correlated with others. The remaining set of features is fed into the three ANNs and a post-processing step is applied to remove very small areas. Results applied on a dataset of 58 patients showed a significant decrease of FPs (20 vs 12; p-value < 0.0001) and an increase of the precision of PCa segmentation (0.62 vs 0.71; p-value < 0.0001). Having less FPs is helpful to increase the performance of CAD systems in terms of specificity and to decrease the reporting time of radiologists. Moreover, having more precise PCa segmentation areas could be useful if a step of PCa characterization will be added to the CAD system
- …
