Template Matching Scheme Research Articles

Computer-assisted detection of mammographic masses: a template matching scheme based on mutual information.

The purpose of this study was to develop a knowledge-based scheme for the detection of masses on digitized screening mammograms. The computer-assisted detection (CAD) scheme utilizes a knowledge databank of mammographic regions of interest (ROIs) with known ground truth. Each ROI in the databank serves as a template. The CAD system follows a template matching approach with mutual information as the similarity metric to determine if a query mammographic ROI depicts a true mass. Based on their information content, all similar ROIs in the databank are retrieved and rank-ordered. Then, a decision index is calculated based on the query's best matches. The decision index effectively combines the similarity indices and ground truth of the best-matched templates into a prediction regarding the presence of a mass in the query mammographic ROI. The system was developed and evaluated using a database of 1465 ROIs extracted from the Digital Database for Screening Mammography. There were 809 ROIs with confirmed masses (455 malignant and 354 benign) and 656 normal ROIs. CAD performance was assessed using a leave-one-out sampling scheme and Receiver Operating Characteristics analysis. Depending on the formulation of the decision index, CAD performance as high as A(zeta) = 0.87 +/- 0.01 was achieved. The CAD detection rate was consistent for both malignant and benign masses. In addition, the impact of certain implementation parameters on the detection accuracy and speed of the proposed CAD scheme was studied in more detail.

Confidence-based automated tracking of small gold seeds in single frame amorphous silicon megavoltage images

Implant tracking in radiotherapy is dominated by classical implementations of the less than robust correlation template matching scheme, frequently using large gold seeds under idealised imaging conditions. Consequently, objective quantification of seed detection has attracted little attention. For dynamic, automated seed tracking to be adopted clinically, the reliability of small seed detection must be known for the most testing of imaging conditions. In this study, the authors report a robust, confidence-based development of correlation matching using a novel hybrid template, which is intended for use with small, linear gold seeds, similar to those used in prostate brachytherapy. For subsecond, single frame, 8 MV amorphous silicon imaging of randomly arranged test seeds, the hybrid approach produces separation between true and false seed distributions in both the anterior–posterior (AP) and lateral (LR) imaging directions through the bony pelvis. The AP detection rate is better than predicted at 85%, with effectively no false identifications. The LR rate of 29% is poor, as expected, with a high false identification rate. However, the LR results are three times better than predicted, even without a priori information. This suggests that minimal relative positional information for seeds has the potential to significantly boost performance.

Normalized Idioventricular QT Interval Variability in Patients with Q- and Non-Q Wave Myocardial Infarction

Background and Objectives:The aim of the study was to evaluate the difference of temporal lability in my- ocardial repolarization between acute non-Q (NQMI and Q-wave myocardial infarction (QMI, and to discern whether the locations of myocardial infarction influence such temporal lability. Subjects and Methods:Tw- elve patients with NQMI and 28 with QMI, including 16 anterior (AMI and 12 inferior MI (IMI patients were enrolled. Twenty four-hour ambulatory ECG recordings of each patient were analyzed, and the digitized data was partitioned into 30-min sections. The QT intervals were measured using a template matching strategy. We then calculated the low (LF:0.03-0.15 Hz and high frequency (HF:0.15-0.4 Hz power of the QT interval variability using an algorithm capable of removing the influence of the RR-interval on QT interval va- riability (Normalized Idioventricular QT variability Index:IV-QT. Results:For patients with QMI, the low frequency IV QT (LF IV-QT was higher than that of NQMI (1.941±0.101 and 1.556±0.114 respectively, p<0.05. No difference was seen in the high frequency IV QT (HF IV-QT of the two groups. For QMI pat- ients, both the LF IV-QT and HF IV-QT were higher in day time (6AM-6PM than in night time (6PM-6AM. Comparing the differences of these indices by the location of QMI, both the LF IV-QT and HF IV-QT of AMI were higher than those of IMI patients (2.231±0.135 vs 1.355±0.131 and 2.341±0.161 vs 1.346±0.145 respectively, p<0.0005. Both indices of each group also demonstrated a circadian change. Conclusion:In cases of QMI, the temporal lability in myocardial repolarization is larger than that seen in NQMI. Moreover, it was worse in AMI than IMI. Finally, such temporal repolarization lability tends to have a circadian pattern in QMI. (Korean Circulation J 2001;31(12 :1281-1289

QT and RR interval variability and spectral characteristics in response to physiologic autonomic stimulation

Purpose:The purposes of this study were to compare the magnitude and phase between the RR interval and QT interval variability in the frequency domain. Methods:Twenty four, 12-13 year old healthy males were randomly selected. At resting state and for 5 minutes, ECGs were obtained, and they were digitized to 1000Hz. After measurement of RR interval, QT interval variability was measured using template matching strategy. After normalization of the RR and QT interval time series, power spectral and cross spectral analysis were performed. From each of the time series, low(0.04-0.15 hertz) and high(0.15-0.4 hertz) frequency power were measured. From the phase spectrum, the phases and time lags between the two time series at each of the two frequency range were calculated. Results:The average of RR interval and QT interval was 616.0±71.0, 364.0±47.0 msec, respectively. Their normalized lowand highfrequency power was 4.4±7.9 NU (normalized unit), 0.1±0.1 NU (p<0.005), and 11.0±30.0 NU, 0.3±0.3 (NU, p<0.005), respectively. The phase differences and resulting time lags between the two interval were -0.5±0.4 πradian (-0.9 seconds) and -0.2 ±0.3 πradian (-0.4 seconds) in the lowand high-frequency range, respectively. Conclusion:During resting state, when compared to RR interval, QT interval oscillates in significantly lower amplitude in both lowand highfrequency ranges. However, the oscillations precede those of the RR interval 0.9 seconds and 0.4 seconds, respectively. (Korean Circulation J 2000;30(12):1507-1514)

A Micropower Adaptive Linear Transform Vector Quantiser

Micropower signal classification and compression are becoming important requirements for implantable cardioverter defibrillators although they are currently limited by power and computational constraints. This paper describes an integrated circuit that facilitates Intracardiac Electrogram (ICEG) classification and compression of 30 dimensional analogue vectors while consuming a maximum of 2.5 μ W power for a heart rate of 60 beats per minute (1 vector per second) from a 3.3 V supply. This represents a significant advance on previous work which achieved ultra low power supervised morphology classification [6] since the templated matching scheme used in this chip enables unsupervised blind classification of abnormal rhythms and the computational support for low bit rate data compression. The adaptive template matching scheme used is tolerant to amplitude variations, and inter- and intra-sample time shifts. Micropower performance is achieved using CMOS analogue circuits biased in weak inversion in order to minimise energy per computation. Results from the fabricated chip demonstrate the impact of the amplitude and shift tolerance on ICEG data and performance for blind classification of an abnormal rhythm in five heart patients. For four out of the five patients, no false negative classifications and a worst case of 11\% false positive classifications were made.

Template Matching as a Strategy for Assessment of and Intervention for Preschool Students with Disabilities

A template-matching strategy was used to develop an intervention for facilitating the transition between preschool and kindergarten for preschoolers with disabilities. Three classes of preschool students who would be entering kindergarten the following year participated: an Intervention group, an Assessment Only group, and a Control group. The study was conducted in two phases: In Phase 1, templates were developed to evaluate the differences between instructional environments in the sending preschool classes and the receiving kindergarten classes. The design of these templates was based on direct observations of the preschool and kindergarten classes using the Eco-Behavioral System for Complex Assessments of Preschool Environments (ESCAPE) and the Assessment Code/Checklist for Evaluating Survival Skills (ACCESS). Ecobehavioral variables were then graphed for each class. In Phase 2, an intervention focusing on some of the major differences found between the environments was designed in collaboration with the preschool staff and implemented for 8 weeks. This intervention was effective in more closely aligning the preschool environments, and teacher and child behaviors, with those reflected in the kindergarten data. At follow-up, preschool students involved in the intervention who were now in kindergarten exhibited fewer competing behaviors and were the recipients of fewer individual prompts during independent work tasks.

Orientation estimation of 3D surface patches

Orientation estimation of a surface patch is an essential step in an automatic inspection job or a 3D template matching scheme for object recognition. Two main problems of the estimation are non-uniform surface sampling and self-occlusion. To solve these problems, we propose to use a 3D spherical window to extract the correct patch and to use closed contours on the surface patch for matching. 3D resampling on a contour is simple and uniform for all viewing directions. Orientation estimation based on the resampled contours is reasonably accurate and can be computed at high speed. However, better accuracy is obtained if more surface patches are used to compute an object orientation first, and the patch orientation is then deduced from the model of the object. The method is designed mainly for irregular-shaped objects.

Segmentation of a seismic section using image processing and artificial intelligence techniques

Methods for segmenting stacked seismic data into zones of common signal character based on texture analysis are described. Their performance is demonstrated on a line of seismic data from the Gulf of Mexico that had been manually segmented. Two segmentation methods are described. The first is a template matching scheme that matches previously selected data templates with a block of pixels. The second uses statistics determined by examining the run-length of seismic reflection events. The run-length method is extended, through a decision process called the RESOLVER, to incorporate heuristic rules to influence the segmentation. A comparison is made between the automatic segmentations of the section and a manual interpretation.

Individual differences in the detection of embedded figures

The individual differences in processing obtained by Hock (1973) were extended to two experiments in which the embedded figures test was recast into the form of a yes-no reaction time task. Ss emphasizing analytic processes detected the embedded figures more rapidly than Ss emphasizing structural processes, supporting the hypothesis that attention to the parts of the embedded figures was the strategy most appropriate for the task. Based on the interaction between familiarity (normal vs rotated orientation) and embeddedness (intact vs embedded figures), it was concluded in both experiments that familiarity facilitated the detection of the embedded figures for analytic Ss. For structural Ss, familiarity was similarly found to facilitate the detection of the embedded figures, but only when a template-matching strategy was possible (when the target figure and the embedded figure were physically identical).