Use Of Similarity Measures Research Articles

A Scoping Review of Outcomes Measured and Involvement of People With Intellectual Disabilities in Active Support Research.

Active Support is a staff practice that aims to increase engagement of people with intellectual disabilities. This study seeks to: (1) identify the outcomes of staff using Active Support and how these are measured; (2) identify how the views of people with intellectual disabilities have been included in Active Support research. A scoping review was conducted of peer-reviewed articles about Active Support published between 2009 and 2023. Sixteen articles were identified. Outcomes included engagement, the quality of staff support, choice and challenging behaviour. Structured observations were the most frequently used measurement method. Only two studies explored people with intellectual disabilities' perspectives about Active Support. Consistent use of similar measures is a strength of Active Support research. Including perspectives of people with intellectual disabilities may provide insight into dimensions of support not been measured and assist in further developing these measures and Active Support practice.

Serial Organization in Krzysztof Penderecki's Threnody to the Victims of Hiroshima

Krzysztof Penderecki and his renowned Threnody to the Victims of Hiroshima (1960) are typi cally associated with sound mass or sonorism. However, Penderecki's sonoristic approach does not necessarily exclude other post-war compositional techniques—Scott Murphy, Wolfram Schwinger, Nelson Mandrell, and Mariusz Kozak have all identified serial tech niques in Penderecki's music in general or in Threnody specifically. As the locus classicus of sound mass compositions, these studies of serialism in Threnody deconstruct rhetorized polarities of twentieth-century musical aesthetics and composition. Neither sound mass nor serial analyses of Threnody, however, have accounted for the role of serial procedures in organizing textural elements. In this analytical vignette, I analyze the organization of extended techniques in mm. 6–9 of Penderecki's Threnody using serial approaches, which yields two benefits. First, it reveals an isography between the or ganization in each measure and the overall organization across all four measures. Second, it encourages the use of similarity measures to consider the compositional goal of this organization. Ultimately, I argue that, much like serialized pitch can be used to obfuscate a tonal center, Penderecki's organization of extended techniques creates a mass of sound without a perceivable pattern—a composed-out alteatoricism.

A novel multi-parameter similarity measure of interval neutrosophic sets for medical diagnosis

The use of similarity measures in interval-valued neutrosophic sets (IVNSs) theory is essential for comparing and assessing the degree of IVNSs difference. However, existing similarity measures for IVNSs suffer from several issues such as lacking precise axiomatic definitions, counterintuitive results, division by zero errors, inability to distinguish between positive and negative differences, and failure to satisfy the ranking definition. To address these limitations, we propose a novel multi-parameter similarity measure for IVNSs based on the tangent function. We demonstrate that our measure satisfies the axiomatic definition and apply it to medical diagnosis, achieving accurate diagnostic results. Additionally, we consider the interactions between symptoms, adjust the proposed similarity measure using Choquet integrals, and provide analytical comparisons to demonstrate the advantages of our improved similarity measure, highlighting its stability and high confidence in the field of medical diagnosis.This study contributes to the advancement of similarity measures in IVNSs theory and provides valuable insights for the field of medical diagnosis.

Impact of COVID-19 Social Distancing Mandates on Gastrointestinal Pathogen Positivity: Secondary Data Analysis.

BackgroundAcute gastrointestinal (GI) illnesses are of the most common problems evaluated by physicians and some of the most preventable. There is evidence of GI pathogen transmission when people are in close contact. The COVID-19 pandemic led to the sudden implementation of widespread social distancing measures in the United States. There is strong evidence that social distancing measures impact the spread of SARS-CoV-2, and a growing body of research indicates that these measures also decrease the transmission of other respiratory pathogens.ObjectiveThis study aims to investigate the impact of COVID-19 social distancing mandates on the GI pathogen positivity rates.MethodsDeidentified GI Panel polymerase chain reaction test results from a routinely collected diagnostic database from January 1, 2019, through August 31, 2020, were analyzed for the GI pathogen positivity percentage. An interrupted time series analysis was performed, using social distancing mandate issue dates as the intervention date. The following 3 target organisms were chosen for the final analysis to represent different primary transmission routes: adenovirus F40 and 41, norovirus GI and GII, and Escherichia coli O157.ResultsIn total, 84,223 test results from 9 states were included in the final data set. With the exception of E coli O157 in Kansas, Michigan, and Nebraska, we observed an immediate decrease in positivity percentage during the week of social distancing mandates for all other targets and states. Norovirus GI and GII showed the most notable drop in positivity, whereas E coli O157 appeared to be least impacted by social distancing mandates. Although we acknowledge the analysis has a multiple testing problem, the majority of our significant results showed significance even below the .01 level.ConclusionsThis study aimed to investigate the impact of social distancing mandates for COVID-19 on GI pathogen positivity, and we discovered that social distancing measures in fact decreased GI pathogen positivity initially. The use of similar measures may prove useful in GI pathogen outbreaks. The use of a unique diagnostic database in this study exhibits the potential for its use as a public health surveillance tool.

Formalized approaches to assessing the usability of the user interface of web applications

The paper proposes an approach based on the use of mathematical methods for modeling the behavior of the designed web application and user behavior. This approach allows an evaluation of the usability of this application. The paper discusses the models of user interfaces (UI), which are used to assess the convenience of site navigation. The possibility of using two classes of UI models is shown – structural-logical (ontological) and probabilistic. Models of the first class are used to assess the degree of comprehensibility of the proposed UI navigation on the search web resource in the accepted terms. For this, mathematically defined measures of similarity between the elements of the UI and their relationships and the elements of a hypothetical (but plausible) model of the user’s perception of the search area in a given subject area are used. In particular, the paper substantiates the possibility of using Jaccard and Levenshtein lexicographic distance metrics as such measures of similarity. Probabilistic models are based on Markov chains. With the help of these models, in the presence of reliable statistical data collected during the trial operation of a web application or its prototype, it is proposed to estimate the average number of user steps required to solve a specific problem of finding information about a subject area object. The paper provides some recommendations on the use of similarity measures to improve the usability of the UI of web applications. In addition, data on the technique for estimating the probabilities of transitions of Markov chains and semantic connections are presented. To clarify the features of the UI analysis models proposed in the article, the website of the RTU MIREA Admission Committee (https://priem.mirea.ru) is provided with a fairly simple organization that provides a search for information about admission and training at the University. The application of the approach described in the paper, along with traditional methods of testing the usability of UI, will increase the overall level of usability of applications, and thereby reduce the cost of identifying and correcting related errors.

Kuaa: A unified framework for design, deployment, execution, and recommendation of machine learning experiments

In this work, we propose Kuaa, a workflow-based framework that can be used for designing, deploying, and executing machine learning experiments in an automated fashion. This framework is able to provide a standardized environment for exploratory analysis of machine learning solutions, as it supports the evaluation of feature descriptors, normalizers, classifiers, and fusion approaches in a wide range of tasks involving machine learning. Kuaa also is capable of providing users with the recommendation of machine-learning workflows. The use of recommendations allows users to identify, evaluate, and possibly reuse previously defined successful solutions. We propose the use of similarity measures (e.g., Jaccard, Sørensen, and Jaro–Winkler) and learning-to-rank methods (LRAR) in the implementation of the recommendation service. Experimental results show that Jaro–Winkler yields the highest effectiveness performance with comparable results to those observed for LRAR, presenting the best alternative machine learning experiments to the user. In both cases, the recommendations performed are very promising and the developed framework might help users in different daily exploratory machine learning tasks.

Improving Detection of Arrhythmia Drug-Drug Interactions in Pharmacovigilance Data through the Implementation of Similarity-Based Modeling.

Identification of Drug-Drug Interactions (DDIs) is a significant challenge during drug development and clinical practice. DDIs are responsible for many adverse drug effects (ADEs), decreasing patient quality of life and causing higher care expenses. DDIs are not systematically evaluated in pre-clinical or clinical trials and so the FDA U. S. Food and Drug Administration relies on post-marketing surveillance to monitor patient safety. However, existing pharmacovigilance algorithms show poor performance for detecting DDIs exhibiting prohibitively high false positive rates. Alternatively, methods based on chemical structure and pharmacological similarity have shown promise in adverse drug event detection. We hypothesize that the use of chemical biology data in a post hoc analysis of pharmacovigilance results will significantly improve the detection of dangerous interactions. Our model integrates a reference standard of DDIs known to cause arrhythmias with drug similarity data. To compare similarity between drugs we used chemical structure (both 2D and 3D molecular structure), adverse drug side effects, chemogenomic targets, drug indication classes, and known drug-drug interactions. We evaluated the method on external reference standards. Our results showed an enhancement of sensitivity, specificity and precision in different top positions with the use of similarity measures to rank the candidates extracted from pharmacovigilance data. For the top 100 DDI candidates, similarity-based modeling yielded close to twofold precision enhancement compared to the proportional reporting ratio (PRR). Moreover, the method helps in the DDI decision making through the identification of the DDI in the reference standard that generated the candidate.

Data compatibility in the addiction sciences: An examination of measure commonality

The need for comprehensive analysis to compare and combine data across multiple studies in order to validate and extend results is widely recognized. This paper aims to assess the extent of data compatibility in the substance abuse and addiction (SAA) sciences through an examination of measure commonality, defined as the use of similar measures, across grants funded by the National Institute on Drug Abuse (NIDA) and the National Institute on Alcohol Abuse and Alcoholism (NIAAA). Data were extracted from applications of funded, active grants involving human-subjects research in four scientific areas (epidemiology, prevention, services, and treatment) and six frequently assessed scientific domains. A total of 548 distinct measures were cited across 141 randomly sampled applications. Commonality, as assessed by density (range of 0–1) of shared measurement, was examined. Results showed that commonality was low and varied by domain/area. Commonality was most prominent for (1) diagnostic interviews (structured and semi-structured) for substance use disorders and psychopathology (density of 0.88), followed by (2) scales to assess dimensions of substance use problems and disorders (0.70), (3) scales to assess dimensions of affect and psychopathology (0.69), (4) measures of substance use quantity and frequency (0.62), (5) measures of personality traits (0.40), and (6) assessments of cognitive/neurologic ability (0.22). The areas of prevention (density of 0.41) and treatment (0.42) had greater commonality than epidemiology (0.36) and services (0.32). To address the lack of measure commonality, NIDA and its scientific partners recommend and provide common measures for SAA researchers within the PhenX Toolkit.

WTO Case Law in 2013

This is an analytical survey of the WTO case law for 2013. This survey is carried out within the framework of the PhD program in International Law and Economics of the PhD school of Bocconi University in Milan. Professor Giorgio Sacerdoti of Bocconi University and the Board of Editors, a former member of the WTO Appellate Body (2001-2009), has coordinated the individual reviews of WTO case law. They have been contributed by current PhD candidates and other young scholars associated with the PhD program in International Law and Economics; their names are listed above and at the end of each review. It was a slow year for WTO case law in the sense that the only Appellate Body decisions to appear were the “twin reports” Canada – Renewable Energy and Canada – Feed-In Tariffs, which focus on the same renewable energy measures in the Canadian province of Ontario. In addition, two unappealed Panel Reports on antidumping measures, China – X-Ray Equipment and China – Broiler Products were adopted by the Dispute Settlement Body (DSB) in 2013. Despite few Reports, 2013 was in several respects a landmark year. In particular, the disputes against Canada marked the first time the Appellate Body addressed the WTO-compatibility of renewable energy regulations, currently a contentious trade issue. The US and China have filed complaints against India and the EU respectively for the use of similar measures. The outcomes thus provided an important precedent. While making clear that local content requirements for generation equipment are not compatible with WTO obligations, the decisions leave open questions about what types of green subsidies are permissible. The rest of the dispute activity for the year focused on China, specifically two Panel decisions mentioned above and reviewed here after, concluding that it had misused trade remedies against alleged anti-competitive practices. Furthermore, in 2013, another antidumping dispute, concluded in 2012, China – GOES, which focused on similar claims, went to arbitration under Article 21.3 to determine a reasonable period of time for implementation; a review of the award is also included here. A number of disputes were set in motion, with 17 requests for consultation initiated. This is a return to average after 2012, an outlier year which saw the initiation of 27 new disputes. Russia marked its first full year of WTO Membership with a high level of dispute activity. Russia filed its first complaint over EU anti-dumping measures. Its recycling fees for motor vehicles also became the target of complaints from Japan and the EU. It was also an active year for arbitration and compliance Panels. Besides the China – GOES arbitration, the Appellate Body reports from the 2012 in the disputes against the US under the TBT Agreement failed to be implemented to the satisfaction of complaining governments. Compliance panels have been composed to evaluate whether the US has implemented the Appellate Body decisions in US – Tuna II and US – COOL. In US – Clove Cigarettes, Indonesia proceeded directly to a request to impose trade remedies under Article 22.6 without first seeking a decision about whether the measures were in compliance under Article 21.5. This is the first time this has happened since EC – Bananas. Finally, the two compliance panels in EC – Large Civil Aircraft and US – Large Civil Aircraft (both composed in late 2012) have been unable to complete their work in 2013, and continued working on their reports into 2014.

The Calculation of Molecular Structural Similarity: Principles and Practice

Measures of structural similarity play an important role in chemoinformatics for applications such as similarity searching, database clustering and molecular diversity analysis. A similarity measure comprises three components: a structure representation; a weighting scheme; and a similarity coefficient. The paper introduces these components and describes methods for comparing different measures. The use of similarity measures in chemoinformatics research is illustrated by recent projects in the author's laboratory on: the interactions between a weighting scheme and a similarity coefficient; the design of comparative studies of similarity measures; the use of 2D fingerprints for scaffold-hopping searches; and the registration of orphan drugs for rare diseases.

Non-fatal overdose among adult prisoners with a history of injecting drug use in two Australian states

BackgroundRecently released prisoners are at markedly increased risk of death and drug-related causes predominate. Non-fatal overdose (NFOD) is considerably more common than fatal overdose, but has received relatively little research attention and most studies of NFOD in this population have suffered from small samples of unknown representativeness. This study aimed to estimate the prevalence and correlates of lifetime NFOD among prisoners in NSW and Queensland. MethodsCross-sectional surveys of adult prisoners in two Australian states: New South Wales (n=972) and Queensland (n=1316). Use of similar measures and methods in the two states made direct comparison of findings possible. ResultsIn both NSW and Queensland, 23% of participants reported a lifetime history of NFOD and prisoners with a history of injecting drug use were significantly more likely to report lifetime NFOD. The lifetime prevalence of NFOD among prisoners with a history of injecting drug use was significantly higher in NSW than in Queensland (44% vs. 35%; p<0.01). Independent correlates of lifetime NFOD were similar across the two states and included ever attempting suicide, ever injecting heroin, and ever injecting opioids. ConclusionsThe risk of NFOD among prisoners with a history of injecting drug use is high. An understanding of the risk factors for NFOD in this population can inform targeted, evidence-based interventions to reduce this risk.

Trust in Personalized Web Search based on Clustered Query Sessions

Personalized Web Search customizes the search according to the information need of the user on the Web. In Personalized Web search, search is personalized using the search results of those users profile which have the information need similar to that of present/current user. It is realized that similarity measure of the users profile is not the sufficient criterion for identifying the trusted users profile who are good in generating the effective personalized search results. One of the Research done in Personalized Web Search in [10] in which clustered user query sessions are used for generating the personalized web search results for the current user using the similarity measure alone. However the method proved to be effective but the use of similarity measure alone could not help in determining how good the clustered user profiles are in generating the reliable recommendations for effective personalized web search. In this paper an algorithm is proposed in which trust is introduced in personalized web search based on clustered query session for which trust metric is defined for clustered user sessions which measure the goodness of the clusters of query sessions in making the reliable recommendations for effective personalized web search and furthermore trust is not static but updated dynamically depending on the response of the user to the personalized search results generated by the selected trusted cluster. The proposed algorithm uses both the trust metric and the similarity measure for selecting the trusted cluster which is similar to the information need of the current user for the personalization of the web search effectively. Experiment was conducted on the clustered user query session to test the effectiveness of the proposed trust based personalized web search and results confirms the effectiveness of the proposed approach.

Early use of inhaled corticosteroids in the emergency department treatment of acute asthma.

Systemic corticosteroid therapy is central to the management of acute asthma. The use of inhaled corticosteroids (ICS) may also be beneficial in this setting. To determine the benefit of ICS for the treatment of patients with acute asthma managed in the emergency department (ED). We identified controlled clinical trials from the Cochrane Airways Group specialised register of controlled trials. Bibliographies from included studies, known reviews, and texts also were searched. The latest search was September 2012. We included randomised controlled trials (RCTs) and quasi-RCTs. Studies were included if patients presented to the ED or its equivalent with acute asthma, and were treated with ICS or placebo, in addition to standard therapy. Two review authors independently selected potentially relevant articles, and then independently selected articles for inclusion. Methodological quality was independently assessed by two review authors. There were three different types of studies that were included in this review: 1) studies comparing ICS vs. placebo, with no systemic corticosteroids given to either treatment group, 2) studies comparing ICS vs. placebo, with systemic corticosteroids given to both treatment groups, and 3) studies comparing ICS alone versus systemic corticosteroids. For the analysis, the first two types of studies were included as separate subgroups in the primary analysis (ICS vs. placebo), while the third type of study was included in the secondary analysis (ICS vs. systemic corticosteroid). Data were extracted independently by two review authors if the authors were unable to verify the validity of extracted information. Missing data were obtained from the authors or calculated from other data presented in the paper. Where appropriate, individual and pooled dichotomous outcomes were reported as odds ratios (OR) with 95% confidence intervals (CIs). Where appropriate, individual and pooled continuous outcomes were reported as mean differences (MD) or standardized mean differences (SMD) with 95% CIs. The primary analysis employed a fixed-effect model and a random-effects model was used for sensitivity analysis. Heterogeneity is reported using I-squared (I(2)) statistics. Twenty trials were selected for inclusion in the primary analysis (13 paediatric, seven adult), with a total number of 1403 patients. Patients treated with ICS were less likely to be admitted to hospital (OR 0.44; 95% CI 0.31 to 0.62; 12 studies; 960 patients) and heterogeneity (I(2) = 27%) was modest. This represents a reduction from 32 to 17 hospital admissions per 100 patients treated with ICS in comparison with placebo. Subgroup analysis of hospital admissions based on concomitant systemic corticosteroid use revealed that both subgroups indicated benefit from ICS in reducing hospital admissions (ICS and systemic corticosteroid versus systemic corticosteroid: OR 0.54; 95% CI 0.36 to 0.81; 5 studies; N = 433; ICS versus placebo: OR 0.27; 95% CI 0.14 to 0.52; 7 studies; N = 527). However, there was moderate heterogeneity in the subgroup using ICS in addition to systemic steroids (I(2) = 52%). Patients receiving ICS demonstrated small, significant improvements in peak expiratory flow (PEF: MD 7%; 95% CI 3% to 11%) and forced expiratory volume in one second (FEV(1): MD 6%; 95% CI 2% to 10%) at three to four hours post treatment). Only a small number of studies reported these outcomes such that they could be included in the meta-analysis and most of the studies in this comparison did not administer systemic corticosteroids to either treatment group. There was no evidence of significant adverse effects from ICS treatment with regard to tremor or nausea and vomiting. In the secondary analysis of studies comparing ICS alone versus systemic corticosteroid alone, heterogeneity among the studies complicated pooling of data or drawing reliable conclusions. ICS therapy reduces hospital admissions in patients with acute asthma who are not treated with oral or intravenous corticosteroids. They may also reduce admissions when they are used in addition to systemic corticosteroids; however, the most recent evidence is conflicting. There is insufficient evidence that ICS therapy results in clinically important changes in pulmonary function or clinical scores when used in acute asthma in addition to systemic corticosteroids. Also, there is insufficient evidence that ICS therapy can be used in place of systemic corticosteroid therapy when treating acute asthma. Further research is needed to clarify the most appropriate drug dosage and delivery device, and to define which patients are most likely to benefit from ICS therapy. Use of similar measures and reporting methods of lung function, and a common, validated, clinical score would be helpful in future versions of this meta-analysis.

Outline Shape Retrieval Using Textual Descriptors and Geometric Features

Content_based image retrieval is a promising approach because of its automatic indexing, recognition and retrieval. This paper is a contribution in the field of the content Based Image Retrieval (CBIR). Objects are represented by their outlines shapes (silhouettes) and described following the XLWDOS Textual Description (Larabi et al., 2003). Textual Descriptors are sensitive to noise. The authors have already developed an approach to smooth the outlines at different scales (Aouat &amp; Larabi, 2010). The smoothing is performed by applying a convolution using the Gaussian Filter to process noisy shapes in order to match shapes descriptors. The authors have also applied an indexing process after silhouettes smoothing (Aouat &amp; Larabi, 2009). The approaches (Aouat &amp; Larabi, 2010; Aouat &amp; Larabi, 2009) are very interesting for shape matching and indexing, but unfortunately, they are not appropriate to the recognition and retrieval processes because there is no use of similarity measures. In order to perform the retrieval process and select the best model for a query silhouette, the authors use in this paper Geometric features extracted from Textual Description of Outline Shapes.

On the use of similarity metrics for approximate graph matching

Abstract In this paper, we investigate heuristics for approximate graph matching, in particular its formulation as a Maximum Common Edge Subgraph problem. Our experiments suggest that a small percentage of accurate node matches is sufficient to get near optimal solutions using a simple hill-climbing. The real challenge could then be to somehow drag the search in this zone. For this purpose, we discuss the use of similarity measures. We present and assess the performance of two similarity measures. Very good results were obtained on labeled graphs.

Detection of illegal use of mobile communication operators resources

Approach is presented for the rapid and effective analysis of temporal rows, displaying concrete processes in-process provider of mobile communication. The offered solution is based on the method of segmentation of curves and use of sliding window for organization of monitoring of row, and for the analysis of traffic modification of algorithm of voting of estimations is offered with the use of similarity measure

A new approach for detecting scientific specialties from raw cocitation networks

AbstractWe use a technique recently developed by V. Blondel, J.‐L. Guillaume, R. Lambiotte, and E. Lefebvre (2008) to detect scientific specialties from author cocitation networks. This algorithm has distinct advantages over most previous methods used to obtain cocitation “clusters” since it avoids the use of similarity measures, relies entirely on the topology of the weighted network, and can be applied to relatively large networks. Most importantly, it requires no subjective interpretation of the cocitation data or of the communities found. Using two examples, we show that the resulting specialties are the smallest coherent “groups” of researchers (within a hierarchy of cluster sizes) and can thus be identified unambiguously. Furthermore, we confirm that these communities are indeed representative of what we know about the structure of a given scientific discipline and that as specialties, they can be accurately characterized by a few keywords (from the publication titles). We argue that this robust and efficient algorithm is particularly well‐suited to cocitation networks and that the results generated can be of great use to researchers studying various facets of the structure and evolution of science.

Appropriate similarity measures for author co‐citation analysis

AbstractWe provide in this article a number of new insights into the methodological discussion about author co‐citation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors' co‐citation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well‐known cosine. The other two similarity measures have not been used before in the bibliometric literature. We show by means of an example that the choice of an appropriate similarity measure has a high practical relevance. Finally, we discuss the use of similarity measures for statistical inference.

TH-C-I-609-07: Multi-Resolution Rigid Registration of Ultrasound and CT Based On Similarity Measures

Purpose: In radiation therapy, registration of real-time ultrasound (US) images to CT/MR images is valuable in reducing patient setup errors. Any registration process starts with rigid registration. At present practice this often is done manually by visual inspection. This study investigates the use of similarity measures as the basis of US to CT registration. This approach is potentially faster and less prone to errors compared to segmentation or landmarks based approaches. Method and Materials: A special phantom with objects visible in both CT and US was constructed and imaged. 3D-US images were obtained using a 2D probe attached to a mechanical translator. US and CT datasets were co-registered using an iterative registration algorithm that includes: an interpolator, a similarity metric and an optimizer all working at multiple resolution. The fitness of a similarity metric to the registration task was studied by translational and rotational perturbations of rigid transforms to compute functional trends of a similarity score versus deviation. Metrics, if effective, should guide the optimizer back to the true position. A multi-resolution technique was employed to improve speed. Resolution degradation is guided by voxel sizes of both modalities toward isotropic resolution. A general purpose optimizer, Simplex, was used along with the nearest neighbor interpolation method. Results: The Simplex optimizer demonstrated feasibility but not speed. In evaluating similarity metrics, square differences expectedly failed in this multi-modality registration, while mutual information (MI) and cross correlation (CC) were reasonable. Randomly rigid perturbations of up to +/− 10mm translations and +/−5° rotations from the true position converged to a visually accurate position within clinically acceptable limits of 1.0mm shifts and 1.0° rotations. Conclusion: Similarity based registration using CC or MI is feasible and effective even on raw US/CT images. Future improvement will include image filtering, adding a weighting bias, or selecting a better optimizer.

WE‐C‐I‐609‐02: Investigation of Various Methods for Determination of Similarity Measures for Pairs of Clustered Microcalcifications On Mammograms

Purpose: Presentation of images similar to an unknown lesion on mammograms may be useful to radiologists in their image interpretation. The purpose of this study is to investigate different methods for determination of similarity measures that would agree with radiologists' subjective similarity ratings. Method and Materials: We used 881 ROIs that included clustered microcalcifications. We selected 19 “unknown” images each of which was compared with six “known” images to produce 114 pairs. Nine breast radiologists provided similarity ratings based on the overall impression for diagnosis for the 114 pairs. Subjective ratings were marked on a continuous rating scale between 0 and 1, where 0 and 1 correspond to two lesions not similar at all and almost identical, respectively. A number of image features were extracted from the lesions for determination of similarity measures. Two similarity measures were determined by use of the Euclidean distance without and with unique strong features of an unknown image. A psychophysical similarity measure was determined by use of an ANN that can learn the relationship between radiologists' similarity ratings and feature values. The correlation values between the subjective ratings and the similarity measures were determined to evaluate the usefulness of similarity measures. Results: The Euclidean‐distance‐ based measure with unique features provided the improved correlation values for four cases, but degraded correlation values for three cases, compared with that without unique features. When ANN was employed, the correlation coefficients were further improved by use of a leave‐one‐out test method. Conclusion: Simple similarity measures such as those based on the Euclidean distance would not be a reliable measure. The selection of features for a specific unknown case and the use of ANN have the potential to improve the determination of reliable similarity measures. Conflict of Interest: RAS and KD: shareholders, R2 Technology Inc.; KD: shareholder, Deus Technology Inc.