Gli algoritmi sono procedimenti che consentono la risoluzione di un determinato problema attraverso un numero finito di istruzioni sequenziali semplici. In realtà, di semplice in un algoritmo c'è ben poco: lo sa bene chi per mestiere elabora queste istruzioni e le immette in un computer, sfidando ogni giorno la capacità di apprendimento delle macchine, con l'obiettivo di riprodurre sistemi di acquisizione automatica simili a quello umano. Siamo nel mondo del cosiddetto apprendimento supervisionato, una delle aree fondamentali dell'intelligenza artificiale.

The hippocampus is an important structural biomarker for Alzheimer's disease (AD) and has a primary role in the pathogenesis of other neurological and psychiatric diseases. This study presents a fully automated pattern recognition system for an accurate and reproducible segmentation of the hippocampus in structural Magnetic Resonance Imaging (MRI). The method was validated on a mixed cohort of 56 T1-weighted structural brain images, and consists of three processing levels: (a) Linear registration: all brain images were registered to a standard template and an automated method was applied to capture the global shape of the hippocampus. (b) Feature extraction: all voxels included in the previously selected volume were characterized by 315 features computed from local information. (c) Voxel classification: a Random Forest algorithm was used to classify voxels as belonging or not belonging to the hippocampus. In order to improve the classification performance, an adaptive learning method based on the use of the Pearson's correlation coefficient was developed. The segmentation results (Dice similarity index=0.81±0.03) compare well with other state-of-the art approaches. A validation study was conducted on an independent dataset of 100 T1-weighted brain images, achieving significantly better results than those obtained with FreeSurfer.

The automated identification of brain structure in Magnetic Resonance Imaging is very important both in neuroscience research and as a possible clinical diagnostic tool. In this study, a novel strategy for fully automated hippocampal segmentation in MRI is presented. It is based on a supervised algorithm, called RUSBoost, which combines data random undersampling with a boosting algorithm. RUSBoost is an algorithm specifically designed for imbalanced classification, suitable for large data sets because it uses random undersampling of the majority class. The RUSBoost performances were compared with those of ADABoost, Random Forest and the publicly available brain segmentation package, FreeSurfer. This study was conducted on a data set of 50 T1-weighted structural brain images. The RUSBoost-based segmentation tool achieved the best results with a Dice's index of (Formula presented.) ((Formula presented.)) for the left (right) brain hemisphere. An independent data set of 50 T1-weighted structural brain scans was used for an independent validation of the fully trained strategies. Again the RUSBoost segmentations compared favorably with manual segmentations with the highest performances among the four tools. Moreover, the Pearson correlation coefficient between hippocampal volumes computed by manual and RUSBoost segmentations was 0.83 (0.82) for left (right) side, statistically significant, and higher than those computed by Adaboost, Random Forest and FreeSurfer. The proposed method may be suitable for accurate, robust and statistically significant segmentations of hippocampi.

Neurodegenerative diseases are frequently associated with structural changes in the brain. Magnetic resonance imaging (MRI) scans can show these variations and therefore can be used as a supportive feature for a number of neurodegenerative diseases. The hippocampus has been known to be a biomarker for Alzheimer disease and other neurological and psychiatric diseases. However, it requires accurate, robust, and reproducible delineation of hippocampal structures. Fully automatic methods are usually the voxel based approach; for each voxel a number of local features were calculated. In this paper, we compared four different techniques for feature selection from a set of 315 features extracted for each voxel: (i) filter method based on the Kolmogorov-Smirnov test; two wrapper methods, respectively, (ii) sequential forward selection and (iii) sequential backward elimination; and (iv) embedded method based on the Random Forest Classifier on a set of 10 T1-weighted brain MRIs and tested on an independent set of 25 subjects. The resulting segmentations were compared with manual reference labelling. By using only 23 feature for each voxel (sequential backward elimination) we obtained comparable state-of-the-art performances with respect to the standard tool FreeSurfer.

Ulcerative colitis (UC) and Crohn's disease (CD) share some pathogenetic features. To provide new steps on the role of altered gene expression, and the involvement of gene networks, in the pathogenesis of these diseases, we performed a genome-wide analysis in 15 patients with CD and 14 patients with UC by comparing the RNA from inflamed and noninflamed colonic mucosa. Methods: Two hundred ninety-eight differentially expressed genes in CD and 520 genes in UC were identified. By bioinformatic analyses, 34 pathways for CD, 6 of them enriched in noninflamed and 28 in inflamed tissues, and 19 pathways for UC, 17 in noninflamed and 2 in inflamed tissues, were also highlighted. Results: In CD, the pathways included genes associated with cytokines and cytokine receptors connection, response to external stimuli, activation of cell proliferation or differentiation, cell migration, apoptosis, and immune regulation. In UC, the pathways were associated with genes related to metabolic and catabolic processes, biosynthesis and interconversion processes, leukocyte migration, regulation of cell proliferation, and epithelial-to-mesenchymal transition. Conclusions: In UC, the pattern of inflammation of colonic mucosa is due to a complex interaction network between host, gut microbiome, and diet, suggesting that bacterial products or endogenous synthetic/catabolic molecules contribute to impairment of the immune response, to breakdown of epithelial barrier, and to enhance the inflammatory process. In patients with CD, genes encoding a large variety of proteins, growth factors, cytokines, chemokines, and adhesion molecules may lead to uncontrolled inflammation with ensuing destruction of epithelial cells, inappropriate stimulation of antimicrobial and T cells differentiation, and inflammasome events.

The knowledge on dolphins and whales, their habitat suitability and conservation status in the Mediterranean Sea is still rather heterogeneous and defective although these aspects represent fundamental currency for evaluating the anthropogenic impacts on the ecosystem functioning. Concerning the Ionian Sea (Central-eastern Mediterranean Sea), starting from 2009 an intense research activity was carried out by the Jonian Dolphin Conservation (JDC) and the Department of Biology University of Bari (UNIBA Local Research Unit CoNISMa), reducing the shortcomings in the basic scientific information in the Gulf of Taranto (Northern Ionian Sea). Sightings data were collected during standardized vessel-based surveys confirming the presence of 8 different species of cetaceans (Dimatteo et al., 2011; Fanizza et al., 2014; Carlucci et al., 2015 and 2016). Innovative field of research were opened from 2015 collaborating with the Muséum National d'Histoire Naturelle UMR-Paris, for the genetic study on the social structure in the striped dolphin Stenella coeruleoalba and the bottlenose dolphin Tursiops truncatus, as well as with the Institute of Intelligent Systems for Automation ISSIA-CNR, for an automatic system of photo-identification for the bottlenose dolphin, the Risso's dolphin Grampus griseus and the sperm whale Physeter macrocephalus. Lastly, a new study on the acoustic inventory of the striped dolphin and the Risso's dolphin was started from 2016 collaborating with the Institute for Coastal Marine Environment IAMC-CNR of Torretta Granitola. Finally, an intense education activity was carried out from 2009 by JDC as a structured project of Citizen Science involving school-children, students as well tourists on whale and dolphin watching boarding on 3 different vessels equipped for scientific surveys and monitoring programs. Moreover, a national monitoring along the Italian costs was coordinated by UNIBA on board of the training ships Palinuro and Vespucci in the framework of the dual use activity carried out in collaboration with the Italian Navy.

The hippocampus has a key role in a number of neurodegenerative diseases, such as Alzheimer's Disease. Here we present a novel method for the automated segmentation of the hippocampus from structural magnetic resonance images (MRI), based on a combination of multiple classifiers. The method is validated on a cohort of 50 T1 MRI scans, comprehending healthy control, mild cognitive impairment, and Alzheimer's Disease subjects. The preliminary release of the EADC-ADNI Harmonized Protocol training labels is used as gold standard. The fully automated pipeline consists of a registration using an affine transformation, the extraction of a local bounding box, and the classification of each voxel in two classes (background and hippocampus). The classification is performed slice-by-slice along each of the three orthogonal directions of the 3D-MRI using a Random Forest (RF) classifier, followed by a fusion of the three full segmentations. Dice coefficients obtained by multiple RF (0.87 ± 0.03) are larger than those obtained by a single monolithic RF applied to the entire bounding box, and are comparable to state-of-the-art. A test on an external cohort of 50 T1 MRI scans shows that the presented method is robust and reliable. Additionally, a comparison of local changes in the morphology of the hippocampi between the three subject groups is performed. Our work showed that a multiple classification approach can be implemented for the segmentation for the measurement of volume and shape changes of the hippocampus with diagnostic purposes.

One of the major problems in genomics and medicine is the identification of gene networks and pathwaysderegulated in complex and polygenic diseases, like cancer. In this paper, we address the problem ofassessing the variability of results of pathways analysis identified in different and independent genomewide expression studies, in which the same phenotypic conditions are assayed. To this end, we assessedthe deregulation of 1891 curated gene sets in four independent gene expression data sets of subjectsaffected by colorectal cancer (CRC). In this comparison we used two well-founded statistical modelsfor evaluating deregulation of gene networks. We found that the results of pathway analysis in expressionstudies are highly reproducible. Our study revealed 53 pathways identified by the two methods inall the four data sets analyzed with high statistical significance and strong biological relevance withthe pathology examined. This set of pathways associated to single markers as well as to whole biologicalprocesses altered constitutes a signature of the disease which sheds light on the genetics bases of CRC.

Several applications aim to identify rare events from very large data sets. Classification algorithms may present great limitations on large data sets and show a performance degradation due to class imbalance. Many solutions have been presented in literature to deal with the problem of huge amount of data or imbalancing separately. In this paper we assessed the performances of a novel method, Parallel Selective Sampling (PSS), able to select data from the majority class to reduce imbalance in large data sets. PSS was combined with the Support Vector Machine (SVM) classification. PSS-SVM showed excellent performances on synthetic data sets, much better than SVM. Moreover, we showed that on real data sets PSS-SVM classifiers had performances slightly better than those of SVM and RUSBoost classifiers with reduced processing times. In fact, the proposed strategy was conceived and designed for parallel and distributed computing. In conclusion, PSS-SVM is a valuable alternative to SVM and RUSBoost for the problem of classification by huge and imbalanced data, due to its accurate statistical predictions and low computational complexity.

We live in the era of the fourth industrial revolution, where everything - from small objects to entire factories - is smart and connected, and we are also strongly accustomed to comforts and services, but emergent questions are arising. What are the consequences of human activities on terrestrial and aquatic/marine systems? And how does the loss of biodiversity alter the integrity and functioning of ecosystems? It is reasonable to assert that there are correlations between the anthropic pressure and degradation of natural habitats and loss in biodiversity. In fact, the alteration of ecosystem structure affects ecosystem services and resilience, the level of perturbation that an ecosystem can withstand without shifting to an alternative status providing fewer benefits to humans [1]. To that regards, the research studies on cetacean species distribution and conservation status along with their habitats can give an idea of the current impact of human pressure on marine biodiversity and its ecosystem services, being both dolphins and whales key species in the marine food webs. However, although the inherent complexity of food-web dynamics often makes difficult to investigate and quantify the role of marine mammals in the ecosystem [2], the challenge to investigate their ecological significance is leading and highly informative when facing human induced environmental changes from local to global scales. For this reason, dedicated research activities have been performed in the last years to standardize the best practices for sampling and collecting scientific relevant information on the cetaceans in the Gulf of Taranto (Northern Ionian Sea in the Central-Eastern Mediterranean Sea) [3, 4, 5, 6]. Standardized scientific protocols and technological innovations have been brought by integrating interdisciplinary approaches: a genetic study on dolphin's social structure, an automated photo-identification, assisted by intelligent unsupervised algorithms and the study of acoustic signals. Finally, education and citizen science were applied as fundamental to raise awareness on the need of marine environmental protection among the active population, from children to adults.

Periodic inspection of large tonnage vessels is critical to assess integrity and prevent structural failures that could have catastrophic consequences for people and the environment. Currently, inspection operations are undertaken by human surveyors, often in extreme conditions. This paper presents an innovative system for the automatic visual inspection of ship hull surfaces, using a Magnetic Climbing Robot (MARC) equipped with a low-cost monocular camera.

Simultaneous analysis of the transcripts of thousands of genes by cDNA microarrays allows the identification of genetic regulatory mechanisms involved in disease pathophysiology. The circadian clock circuitry controls essential cell processes and the functioning of organ systems, which are characterized by rhythmic variations with 24-hour periodicity. The derangement of these processes is involved in the basic mechanisms of inflammatory, metabolic, degenerative and neoplastic diseases. We evaluated by genome-wide cDNA microarray analysis the transcriptome of endoscopic mucosal biopsies of patients with inflammatory bowel diseases (IBD) focusing on the expression of circadian genes in Crohn's disease (CD) and ulcerative colitis (UC). Twenty-nine IBD patients (15 with CD and 14 with UC) were enrolled and mucosal biopsies were sampled at either inflamed or adjacent non-inflamed areas of the colon. A total of 150 circadian genes involved in pathways controlling crucial cell processes and tissue functions were investigated. In CD specimens 50 genes were differentially expressed, and 21 genes resulted up-regulated when compared to healthy colonic mucosa. In UC specimens 50 genes were differentially expressed, and 27 genes resulted up-regulated when compared to healthy colonic mucosa. Among the core clock genes ARNTL2 and RORA were up-regulated, while CSNK2B, NPAS2, PER1 and PER3 were down-regulated in CD specimens. Conversely, ARNTL2, CRY1, CSNK1E, RORA and TIPIN were up-regulated, while NR1D2 and PER3 were down-regulated in UC. In conclusion, in CD and UC patients there are differences in the expression of circadian genes between normal and diseased intestinal mucosa. The deregulated genes evidenced by transcriptome analysis in the major IBDs may play a crucial role in the pathophysiological mechanisms and may suggest novel therapeutic approaches.

Little is known about the immunoediting process in precancerous lesions. We explored this aspect of benign colorectal adenomas with a descriptive analysis of the immune pathways and immune cells whose regulation is linked to the morphology and size of these lesions. Two series of polypoid and nonpolypoid colorectal adenomas were used in this study: 1) 84 samples (42 lesions, each with matched samples of normal mucosa) whose gene expression data were used to quantify the tumor morphology- and size-related dysregulation of immune pathways collected in the Molecular Signature Database, using Gene Set Enrichment Analysis; 2) 40 other lesions examined with immunohistochemistry to quantify the presence of immune cells in the stromal compartment. In the analysis of transcriptomic data, 429 immune pathways displayed significant differential regulation in neoplasms of different morphology and size. Most pathways were significantly upregulated or downregulated in polypoid lesions versus nonpolypoid lesions (regardless of size). Differential pathway regulation associated with lesion size was observed only in polypoid neoplasms. These findings were mirrored by tissue immunostaining with CD4, CD8, FOXP3, MHC-I, CD68, and CD163 antibodies: stromal immune cell counts (mainly T lymphocytes and macrophages) were significantly higher in polypoid lesions. Certain markers displayed significant size-related differences regardless of lesion morphology. Multivariate analysis of variance showed that the marker panel clearly discriminated between precancerous lesions of different morphologies and sizes. Statistical analysis of immunostained cell counts fully support the results of the transcriptomic data analysis: the density of infiltration of most immune cells in the stroma of polypoid precancerous lesions was significantly higher than that observed in nonpolypoid lesions. Large neoplasms also have more immune cells in their stroma than small lesions. Immunoediting in precancerous colorectal tumors may vary with lesion morphology and stage of development, and this variability could influence a given lesion's trajectory to cancer.

A poster about the use of marsupial robots and vehicles for next-genaration of missions in polar regions