The mechanical properties of in vivo soft tissue are generally determined by palpation, ultrasound measurements (US), and magnetic resonance elastography (MRE). While it has been shown that US and MRE are capable of quantitatively measuring soft tissue elasticity, there is still some uncertainty about the reliability of quantitative MRE measurements. Mechanical properties of tissues such as Young's modulus, shear modulus, and bulk modulus are of special interest in tissue characterization. By palpation, the stiffness of the tissue, in particular its resistance to pressure and shear forces, is inspected by the physician's hand. Often, cancerous tissue can be detected since it appears as a hard lesion which is a result of increased stromal density (1). Within the last 10 years various ultrasound (US) and MR methods to quantitatively determine the elasticity of soft tissue have been established. Noninvasive US techniques that measure the elastic properties of soft tissue have been described by Ophir et al. (1) (5) showed that in many cases breast lesions cause changes of the elastic modulus as assessed by US measurements. MR elastography is a more recently proposed technique to measure tissue elastic moduli noninvasively. There are two principally different methods of MRE: static or quasi static (11-16) and dynamic (17-26). Static MRE uses two different static compressional states of the investigated material to determine its corresponding distortions. Dynamic MRE is based on the excitation of mechanical waves in soft tissue. In order to evaluate the quantitative precision of dynamic shear wave MRE, we compared quantitative shear wave MRE results with those from mechanical compression tests. MATERIALS AND METHODS Tissue-Mimicking Phantoms and Compression Test SpecimensFor the MRE measurements and compression tests a series of tissue phantoms and compression test specimens were produced. Ideally, the material of the phantoms and the specimens should mimic human soft tissue. Agar-agar gel was used as a test material because it shows mechanical properties similar to human soft tissues. To produce the phantoms, different amounts of agar-agar powder (Agar Agar Kobe I pulv., Roth, Karlsruhe, Germany) were stirred in distilled water and boiled for about 2 min. Then the liquid agar-agar was poured into cylindrical heat-resistant plastic molds having both a diameter and height of about 16 cm and allowed to cool to room temperature. At about 40°C a chemical cross-linking occurs and the agar-agar changes from a fluid to a solid state. As agar-agar gel is a biological material, water diluted formalin was applied to the surface of the phantoms to inhibit the growth of fungi; this allowed for a lifespan of the phantoms of several months. For the compression tests, cylindrical specimens of 4 cm height and 5 cm diameter were cast. To ensure that the corresponding phantoms and specimens had the same elastic properties, they were made from dilutions with identical agar-agar concentrations.The concentration of the agar-agar powder wa...

The purposes of the present study were to compare the cardiovascular response patterns evoked by three versions of the cold pressor test (either forehead stimulation or hand or foot immersion) and to determine the reproducibility of the responses over a 2-week interval. Blood pressure , heart rate, stroke volume, cardiac output, total peripheral resistance, and systolic time intervals were obtained during rest and during the cold pressor test in 42 young men. Across conditions, the pressor response was supported by peripheral resistance increases wilh COllcomitant stroke volume decreases. Although the response patterns were generally similar across sites, exception s were apparent for heart rate. Forehead stimulation was characterized by no significant change in heart rate, whereas limb (hand or foot) immersion was associated with significant heart rate acceleration . The responses elicited by the three cold pressor test conditions were reliable and showed lillie evidence of attenuation over the test-retest interval.

Back-propagation neural networks were used to classify PET scans as either normal or abnormal, with abnormal subjects defined as subjects who had previously been clinically diagnosed with memory disorders. Numerous neural network experiments were performed in order to achieve optimization with respect to number of hidden unJts and training duration. Optimizations and performance evaluations were based on ROC analysis, in which the area under the ROC curve was the figure of merit The neural network's performance was better than that of dlscrlminant analysis, and comparable to the expen's performance. despite the low resolution image data, which consisted of one value per brain lobe, provided to the network. INTRODUCI10NQuantitatJve approaches to the analysis and/or classification of Positron Emission Tomography (PE1) scans usually involve a reglon-of·interest (ROI) analysis, in which regional metabolic function in the brain is evaluated [I). Pattern recognition stUdies are then performed on these data.Various pattern recognltion technJques, including the back-propagation neural network (2), have been applied to the classification of normal and abnormal PET scans based on ROI data. Neural networks appear to perform better than standard statistical methods like dlscrlminant analysis (3).In the literature describing various recent applications of neural networks, there appears to be relatJvely little standardization In neural-network training. Networks are often trained to satisfy panicular "convergence criteria", which essentially specify how well the hypersurfaces defined by the network are able to separate the different classes comprising the trainJng set A more imponant consIderation in most drcumstances, however, is the abillty of a network to generalize and Identify previOUSly-unseen patterns, an issue which involves the number of traIning patterns, the dimensionality of these patterns, the architecture of the network (e.g., the number of hidden unJts) and number of trainJng Iterations. Complex networks trained on high-dlmenslonaJ patterns for an excessive number of iterations may tend to "memorize" their training sets, and learn criteria that are not generally applicable to populations of given pattern classes. Evaluation of a network's abillty to generalize is accomplished by cross-validation studies, that is, testing trained networks on new and independent data sets. The question then arises: what is the most appropriate figure of merit for performance evaluation?The ROC (Relative-Operating-Characteristic) method of analysis has recently come to be recognized as an Objective and comprehensive way to evaluate diagnostic systems, since It measures a diagnostic system's performance independent of dedsion biases and prior probabillties (4). The ROC curve represents a system's performance at several different settings of the particular dedsion criteria. The area under the curve is the "only performance measure available that is unin1Juenced by dedsion biases and prior probabillties. and it places the per...

No abstract

No abstract

A method of image registration is presented for the case when the deformation between two images can be well approximated with a combination of translation, rotation and global scaling. The method achieves very high accuracy by combining a global optimization in the 4-dimensional discrete parameter space with a local optimization in the 4-dimensional continuous parameter space. The 4-dimensional global optimization is accomplished with two 2-dimensional optimizations. The Fourier magnitude is used to decouple translation from rotation and scaling, and a log-polar mapping of the Fourier magnitude is used to convert rotation and scaling into shifts. Optimal rotation and scaling parameters are determined with a cross-correlation in the log-polar domain. After compensation for rotation and scaling differences, cross-correlation in the spatial domain yields the translation parameters. The four registration parameters are further refined with a local optimization using the correlation coefficient as a similarity measure in the 4-dimensional continuous parameter space. Results are shown from simulations and from registration of retinal images. For simulated images with a signal-to-noise ratio of -5 dB, the accuracy of the registration method is estimated 10 be better than 0.07 degrees, 0.1 %, and 0.3 pixels for rotation, scaling, and translation, respectively. In the case of 512x512 pixel images the computation resource requirements are compatible with high end PCs, i.e., approximately 25 minutes on an Intel 80486(33MHz based IBM/PC compatible.

EKG synchronized ensemble averaging of the impedance cardiogram tends to blur or suppress signal events due to signal jitter or event latency variability. Although ensemble averaging provides some improvement in the stability of the signal and signal to noise ratio under conditions of nonperiodic influences of respiration and motion, coherent averaging techniques were developed to determine whether further enhancement of the impedance cardiogram could be obtained. Physiological signals were obtained from sixteen male and female subjects during resting conditions. while delivering a speech and while undergoing submaximal bicycle exercise. Results indicated that improved resolution of dZ'dt signal events could be obtained using coherent ensemble averaging. Although some improvement in precision of event location was obtained, most enhancement of the impedance cardiogram occurred in measurement of the amplitude of the dZldt maximum (ejection velocity) during speaking and exercise conditions. Validated increases in dZldt maximum exceeding 20% were obtained in some subjects with coherent averaging. suggesting that the diagnostic utility of impedance cardiography can be improved by using this technique. (Supported by NHLBI research grants. HlA 1335 and HL36588.)

In 17 illustrations, a modification of Fujita's technique of UPPP is presented. This operation consists of excising redundant velar tissues from the free margin of the soft palate, tonsillar pillars, and uvula without reducing the muscles of the velum. In addition, the uvula muscle is almost completely preserved. Seventy patients, with habitual snoring or with obstructive sleep apnea syndrome, who underwent this surgery were followed-up from six to 24 months and did not show any palatal insufficiency with air and liquid escape through the nose. Additional remarks are made concerning the difficulties encountered during intubation of 60 percent of patients with chronic rhonchopathy.