Search

Patients' data security and privacy is fundamental in the perspective of moving clinical data over the Cloud. Indeed, this concern has slowed down the adoption of Cloud services in the healthcare context. In fact, clinical operators are reluctant to open Hospital Information Systems (HIS) to external Cloud services. In this paper, we discuss system developed at the IRCCS “Bonino Pulejo” clinical and research centre (Italy) that is able to solve this concern. Such a system is based on two software components that are anonymizer and splitter. The first collects anonymize clinical data, whereas the second obfuscates and stores data in multiple Cloud storage providers. Thus, only authorized clinical operators can access data over the Cloud. A case of study considering real Magnetic Resonance Imaging (MRI) data is analysed in order to assess the performance of the whole system.

In spite of the worldwide financial and research efforts made, the pathophysiological mechanism at the basis of Alzheimer's disease (AD) is still poorly understood. Previous studies using electroencephalography (EEG) have focused on the slowing of oscillatory brain rhythms, coupled with complexity reduction of the corresponding time-series and their enhanced compressibility. These analyses have been typically carried out on single channels. However, limited investigations have focused on the possibility yielded by computational intelligence methodologies and novel machine learning approaches applied to multichannel schemes. The study at screening level on EEG recordings of subjects at risk could be useful to highlight the emergence of underlying AD progression (or at least support any further clinical investigation). In this work, the representational power of Deep Learning on Convolutional Neural Networks (CNN) is exploited to generate suitable sets of features that are then able to classify EEG patterns of AD from a prodromal version of dementia (Mild Cognitive Impairment, MCI) and from age-matched Healthy Controls (HC). The processing system here used enforces a series of convolutional-subsampling layers in order to derive a multivariate assembly of latent, novel patterns, finally used to categorize sets of EEG from different classes of subjects. The final processor here proposed is able to reach an averaged 80% of correct classification with good performance on both sensitivity and specificity by using a Multilayered Feedforward Perceptron (MLP) of the standard type as a final block of the procedure.

Alzheimer's disease (AD) is a degenerative neurological disorder characterized by a loss of functional connections between different areas of the brain. AD is considered a cortical dementia, thus Electroencephalography (EEG) has been used as a tool for diagnosing AD for the last two decades. Often, the hallmarks of EEG abnormality in AD patients are a shift of the power spectrum to lower frequencies and reduced coherences among cortical regions, however, it is still mostly unknown how these abnormalities evolve together with the disease progression. In this paper we proposed a longitudinal study of the EEG of three AD patients in order to study the disease progression, from the coherence point of view, over the four major EEG sub-bands: delta, theta, alpha and beta. The EEG was recorded at time T0 and then after three months (time T1). We proposed a coherence based hierarchical clustering method to group the electrodes together according to their mutual pairwise coherence, in order to evaluate how the brain connectivity changed along with the disease in the spectral domain. The results provide an in-depth view of the structure of electrode interconnection of every single patient in every sub-band at time T0 and time T1. This study endorsed the commonly shared belief that coherence reduces over time but it revealed that coherence spatial distribution changes in a different way, from patient to patient. The results also showed that a patient-specific brain connectivity analysis is possible and that a personalized analysis of the disease's progression might provide valuable diagnostic information. In the near future, the study will be extended to a larger dataset in order to validate the method statistically.

The degree of carotid stenosis is up to now considered one of the most important features for determining the risk of stroke. Ultrasound (US) is a non-invasive technique, relatively inexpensive, portable, which has an excellent temporal resolution. Computer-Aided Diagnosis (CAD) has became one of the major research field in medical imaging and diagnostic radiology. This work presents a CAD system, an automatic prototype software used for carotid plaque detection. We studied US images of 44 subjects, 22 with and 22 subjects without carotid artery stenosis. The CAD system automatically performed the following processing steps: registration of images, detection of lesions by a segmentation method using Watershed algorithm, extraction of dynamic and morphologic features, and classification of plaques through Cluster Analysis. Investigations has been carried out on 287 suspect regions including 61 plaques and 226 non-plaques cases. The area under the ROC curve was 0.89. The experimental results also show that the proposed methods is feasible, and have a good agreement with the expert neurologist. Without the need of any user-interaction, the proposed method generates a detection out-put that may be useful in second opinion.

Image segmentation plays a crucial role in medical imaging by facilitating the delineation of regions of interest. The ultimate aim of an automatic image segmentation system is to mimic the human visual system in order to provide a meaningful image subdivision. The watershed transform is a well established tool for the segmentation of images. We have considered Magnetic Resonance Images (MRI) of four Multiple Sclerosis (MS) patients provided by IRCCS Centro Neurolesi “Bonino-Pulejo” of Messina in their original format and tested on them the watershed algorithm implemented using MATLAB 7.6. In this paper we propose an algorithm that use Watershed variant encapsulating Cluster Analysis, then region merging and edge detection procedures were used. This method uses an analysis of variance approach to evaluate the distances between clusters to identify the lesion and to support clinicians in the diagnosis of MS. The algorithm is able to segment or extract desired parts of only gray-scale images and is applied the Cluster Analysis for solved the problem of undesirable oversegmentation results produced by the watershed technique. From our results, we have seen that several analyzed regions have similar characteristics to be grouped together in same class. In particular, we saw that at a distance equal to the level of 0.084, you can find the MS regions. Then the set of parameters considered provides a good description of the regions selected by watershed and then through the cluster analysis allows the distinction between normal and suspect regions.