Search

Spontaneous formation of peer-to-peer agent-based data mining systems seems a plausible scenario in years to come. However, the emergence of peer-to-peer environments further exacerbates privacy and security concerns that arise when performing data mining tasks. We analyze potential threats to data privacy in a peer-to-peer agent-based distributed data mining scenario, and discuss inference attacks which could compromise data privacy in a peer-to-peer distributed clustering scheme known as KDEC.

Effect estimates from randomized trials and observational studies might not be directly comparable because of differences in study design, other than randomization, and in data analysis. We propose a 3-step procedure to facilitate meaningful comparisons of effect estimates from randomized trials and observational studies: 1) harmonization of the study protocols (eligibility criteria, treatment strategies, outcome, start and end of follow-up, causal contrast) so that the studies target the same causal effect, 2) harmonization of the data analysis to estimate the causal effect, and 3) sensitivity analyses to investigate the impact of discrepancies that could not be accounted for in the harmonization process. To illustrate our approach, we compared estimates of the effect of immediate with deferred initiation of antiretroviral therapy in individuals positive for the human immunodeficiency virus from the Strategic Timing of Antiretroviral Therapy (START) randomized trial and the observational HIV-CAUSAL Collaboration.

We present a novel algorithm for clustering streams of multidimensional points based on kernel density estimates of the data. The algorithm requires only one pass over each data point and a constant amount of space, which depends only on the accuracy of clustering. The algorithm recognizes clusters of nonspherical shapes and handles both inserted and deleted objects in the input stream. Querying the membership of a point in a cluster can be answered in constant time.

Several algorithms have been recently developed for distributed data clustering, which are applied when data cannot be concentrated on a single machine, for instance because of privacy reasons or due to net- work bandwidth limitations, or because of the huge amount of distributed data. Deployed and research Peer-to-Peer systems have proven to be able to man- age very large databases made up by thousands of personal computers resulting in a concrete solutions for the forthcoming new distributed database systems to be used in large grid computing networks and in clustering database management systems. Current distributed data clustering algorithms cannot be ap- plied to such kind of networks because they expect data be organized according to traditional distributed database management systems where the distribution of the relational schema is planned a-priori in the de- sign phase. In this paper we describe methods to cluster distributed data across peer-to-peer networks without requiring any costly reorganization of data, which would be infeasible in such a large and dynamic overlay networks, and without reducing their perfor- mance in message routing and query processing. We compare the data clustering quality and ef- ciency of three multi-dimensional peer-to-peer sys- tems according to two well-known clustering tech- niques.

Background Multiparametric magnetic resonance imaging (mpMRI) can guide the surgical plan during robot-assisted radical prostatectomy (RARP), and intraoperative frozen section (IFS) can facilitate real-time surgical margin assessment. Objective To assess a novel technique of IFS targeted to the index lesion by using augmented reality three-dimensional (AR-3D) models in patients scheduled for nerve-sparing RARP (NS-RARP). Design, setting, and participants Between March 2019 and July 2019, 20 consecutive prostate cancer patients underwent NS-RARP with IFS directed to the index lesion with the help of AR-3D models (study group). Control group consists of 20 patients matched with 1:1 propensity score for age, clinical stage, Prostate Imaging Reporting and Data System score v2, International Society of Urological Pathology grade, prostate volume, NS approach, and prostate-specific antigen in which RARP was performed by cognitive assessment of mpMRI. Surgical procedure In the study group, an AR-3D model was superimposed to the surgical field to guide the surgical dissection. Tissue sampling for IFS was taken in the area in which the index lesion was projected by AR-3D guidance. Measurements Chi-square test, Student t test, and Mann-Whitney U test were used to compare, respectively, proportions, means, and medians between the two groups. Results and limitations Patients in the AR-3D group had comparable preoperative characteristics and those undergoing the NS approach were referred to as the control group (all p ≥ 0.06). Overall, positive surgical margin (PSM) rates were comparable between the two groups; PSMs at the level of the index lesion were significantly lower in patients referred to AR-3D guided IFS to the index lesion (5%) than those in the control group (20%; p = 0.01). Conclusions The novel technique of AR-3D guidance for IFS analysis may allow for reducing PSMs at the level of the index lesion. Patient summary Augmented reality three-dimensional guidance for intraoperative frozen section analysis during robot-assisted radical prostatectomy facilitates the real-time assessment of surgical margins and may reduce positive surgical margins at the index lesion.

Introduction The purpose of this study was to evaluate the impact of 3-dimensional (3D) digital reconstructions of renal models on the arterial clamping approach during partial nephrectomy (PN). Patients and Methods Fifty-seven patients with T1 renal mass, referred for PN, were prospectively enrolled in 2 groups: Group 1 (n = 32) with revision of both 2-dimensional (2D) computed tomography (CT) imaging and 3D virtual model before surgery; Group 2 (n = 25) with revision of 2D CT imaging. Segmentation of the 3D models from preoperative high-quality CT scan was achieved using D2P software. In a sub-analysis of patients treated with PN with the on-clamp approach (n = 36), the effective intraoperative level of arterial clamping was compared with the preoperative planning. Results In the sub-group of patients referred to PN with the on-clamp approach, the intraoperative selective clamping was performed in 12 (57.1%) patients of Group 1 and in 2 (13.3%) cases of Group 2 (P = .01). The intraoperative management of the renal pedicle was done as preoperatively planned in 61.9% of patients in Group 1 and in 86.6% of cases in Group 2 (P = .1). Conclusion The 3D-guided plan of PN allows to perform selective clamping in higher proportion of patients compared with the standard 2D-guided approach without increasing intraoperative and postoperative complications.

Introduction to explore the feasibility of augmented reality (AR) to guide arterial clamping during robot-assisted partial nephrectomy (RAPN). Patients and Methods 15 consecutive patients with T1 renal mass underwent RAPN guided by AR. The 3D virtual model derived by computed tomography was superimposed on the actual view provided by the Da Vinci video stream thought AR technology. Preoperative plan of arterial clamping based on 2D conventional imaging, on 3D model and the effective intraoperative surgical approach guided by AR were compared using the McNeamar test. Results The plan of arterial clamping based on 2D preoperative imaging was recorded as follows: no clamping in 3 (20%), clamping of the main artery in 10 (66.7%) and selective clamping in 1 (6.7%) and super-selective clamping in 1 (6.7%) cases. After revision of the 3D model, the plan of clamping was modified as follows: no clamping in 1 (6.7%), clamping of the main artery in 2 (13.3%), selective clamping in 8 (53.3%) and super-selective clamping in 4 (26.7%) cases (p=0.03). The effective intraoperative clamping approach guided by AR-guidance was performed as planned in 13 (86.7%) patients. Conclusion AR for 3D guided renal surgery is useful to increase the adoption of selective clamping during RAPN.

Background: Augmented reality (AR) is a novel technology adopted in prostatic surgery. Objective: To evaluate the impact of a 3D model with AR (AR-3D model), to guide nerve sparing (NS) during robot-assisted radical prostatectomy (RARP), on surgical planning. Design, Setting, and Participants: Twenty-six consecutive patients with diagnosis of prostate cancer (PCa) and multiparametric magnetic resonance imaging (mpMRI) results available were scheduled for AR-3D NS RARP. Intervention: Segmentation of mpMRI and creation of 3D virtual models were achieved. To develop AR guidance, the surgical DaVinci video stream was sent to an AR-dedicated personal computer, and the 3D virtual model was superimposed and manipulated in real time on the robotic console. Outcome measurements and statistical analysis: The concordance of localisation of the index lesion between the 3D model and the pathological specimen was evaluated using a prostate map of 32 specific areas. A preliminary surgical plan to determinate the extent of the NS approach was recorded based on mpMRI. The final surgical plan was reassessed during surgery by implementation of the AR-3D model guidance. Results and limitations: The positive surgical margin (PSM) rate was 15.4% in the overall patient population; three patients (11.5%) had PSMs at the level of the index lesion. AR-3D technology changed the NS surgical plan in 38.5% of men on patient-based and in 34.6% of sides on side-based analysis, resulting in overall appropriateness of 94.4%. The 3D model revealed 70%, 100%, and 92% of sensitivity, specificity, and accuracy, respectively, at the 32-area map analysis. Conclusions: AR-3D guided surgery is useful for improving the real-time identification of the index lesion and allows changing of the NS approach in approximately one out of three cases, with overall appropriateness of 94.4%. Patient summary: Augmented reality three-dimensional guided robot-assisted radical prostatectomy allows identification of the index prostate cancer during surgery, to tailor the surgical dissection to the index lesion and to change the extent of nerve-sparing dissection. Augmented reality three-dimensional (3D) guided robotic prostatectomy is feasible to improve the real-time identification of index prostate cancer and to modulate the nerve-sparing approach targeted to the index lesion. Augmented reality 3D models revealed good concordance with the whole-mount pathology.

Background: Endovascular aneurysm repair (EVAR) has become the primary choice of treatment for abdominal aortic aneurysm (AAA) and preventive intraoperative AAA sac embolization (SE) is an effective way to reduce endoleaks after EVAR. Preoperative planning of AAA SE has not been standardized yet, and typically requires a quite long and operator-dependent procedure using standard EVAR planning software. Methods: We introduce &lsquo, EVAR-eaSE&rsquo, a software tool that we have developed to help vascular surgeons to plan the optimal SE procedure during EVAR, providing semi-automatic AAA segmentation and 3D modeling starting from patient scans, and then computing the aneurysm volumes. A preliminary accuracy evaluation on radiological phantoms, and a comparison with the commercial 3Mensio software for EVAR planning are carried out. Results: EVAR-eaSE software showed a good level of accuracy in aneurysm volume reconstruction and calculation (mean difference of 2.5% between real and segmented phantom volumes). A strong positive correlation (r = 0.940, r = 0.954, p &lt, 0.001) was found between EVAR-eaSE computation of AAA Lumen and AAA Total volumes and the same measurements obtained with 3Mensio, for a database of 26 AAA cases. Conclusions: EVAR-eaSE demonstrated to be an easy-to-use, objective and accurate tool to assist the vascular surgeons in preoperative planning of SE during EVAR.

Objectives: The aim of the study was to evaluate differences in immunovirological response to combination antiretroviral therapy (cART) in migrant and native men and women within a European collaboration of HIV cohorts Collaboration of Observational HIV Epidemiological Research in Europ (COHERE) in EuroCoord, 2004-2013. Methods: Migrants were defined as those with geographical origin (GO) different from the reporting country and were grouped as originating from Western Europe and Western Countries (WEWC), Eastern Europe (EE), North Africa and the Middle East (NAME), sub-Saharan Africa (SSA), Latin America (LA), Caribbean (CRB) and Asia/Oceania (ASIA/OCE). Native (NAT) individuals were defined as those originating from the reporting country. CD4 cell counts were modelled using piecewise linear mixed-effects models with two slopes, whereas models to estimate subdistribution hazard ratios (sHRs) were used for time to virological response (VR) (i.e. time from cART initiation to the first of two successive HIV RNA measurements < 400 HIV-1 RNA copies/ml). Results: Of 32 817 individuals, 25 799 (78.6%) were men. The percentage of migrants was higher in women (48.9%) than in men (21.2%) and migrants from SSA accounted for the largest migrant group (29.9% in men and 63.3% in women). Migrant men and women from SSA started at lower CD4 cell counts than NAT individuals, which remained lower over time. VR was ≥ 85% at 12 months for all groups except CRB women (77.7%). Compared with NAT men and women, lower VR was experienced by NAME [sHR 0.91; 95% confidence interval (CI) 0.86-0.97] and SSA (sHR 0.88; 95% CI 0.82-0.95) men and CRB (sHR 0.77; 85% CI 0.67-0.89) women, respectively. Conclusions: Immunovirological response to cART in Western Europe varies by GO and sex of patients. ART benefits are not equal for all, underlining the point that efforts need to prioritize those most in need.

Background: CD4 cell recovery following first-line combination ART (cART) is poorer in HIV-2+ than in HIV-1+ patients. Only large comparisons may allow adjustments for demographic and pretreatment plasma viral load (pVL). Methods: ART-naive HIV+ adults from two European multicohort collaborations, COHERE (HIV-1 alone) and ACHIeV2e (HIV-2 alone), were included, if they started first-line cART (without NNRTIs or fusion inhibitors) between 1997 and 2011. Patients without at least one CD4 cell count before start of cART, without a pretreatment pVL and with missing a priori-defined covariables were excluded. Evolution of CD4 cell count was studied using adjusted linear mixed models. Results: We included 185 HIV-2+ and 30321 HIV-1+ patients with median age of 46 years (IQR 36-52) and 37 years (IQR 31-44), respectively. Median observed pretreatment CD4 cell counts/mm(3) were 203 (95% CI 100-290) in HIV-2+ patients and 223 (95% CI 100-353) in HIV-1+ patients. Mean observed CD4 cell count changes from start of cART to 12months were + 105 (95% CI 77-134) in HIV-2+ patients and + 202 (95% CI 199-205) in HIV-1+ patients, an observed difference of 97 cells/mm(3) in 1 year. In adjusted analysis, the mean CD4 cell increase was overall 25 CD4 cells/mm(3)/year lower (95% CI 5-44; P=0.0127) in HIV-2+ patients compared with HIV-1+ patients. Conclusions: A poorer CD4 cell increase during first-line cART was observed in HIV-2+ patients, even after adjusting for pretreatment pVL and other potential confounders. Our results underline the need to identify more potent therapeutic regimens or strategies against HIV-2

That the already vast and ever-increasing amounts of data still do present formidable challenges to effective and efficient acquisition of knowledge is by no means an exaggeration. The knowledge discovery process entails more than just the application of data mining strategies. There are many other aspects including, but not limited to: planning, data pre-processing, data integration, evaluation and presentation. The human-vision channel is capable of recognizing and understanding data at an instant. Effective visual strategies can be used to tap the outstanding human visual channel in extracting useful information from data. Unlike is the case with most research efforts, the exploitation should be employed not just at the beginning or at the end of the knowledge discovery process but across the entire discovery process. In essence, this calls for the development of an effective user/visual component, the development of an overall framework that can support the entire discovery process/all discovery phases, and the strategic placement of the visual component in that framework. Key issues of this component will be the open architecture, allowing extensions and adaptations to specific mining environments, and the precise semantics and syntax, allowing an optimal integration between the presentation and the computation.