Reliable detection of fluence anomalies in EPID-based IMRT pretreatment quality assurance using pixel intensity deviations.

PURPOSE: This work uses repeat images of intensity modulated radiation therapy (IMRT) fields to quantify fluence anomalies (i.e., delivery errors) that can be reliably detected in electronic portal images used for IMRT pretreatment quality assurance. METHODS: Repeat images of 11 clinical IMRT fields are acquired on a Varian Trilogy linear accelerator at energies of 6 MV and 18 MV. Acquired images are corrected for output variations and registered to minimize the impact of linear accelerator and electronic portal imaging device (EPID) positioning deviations. Detection studies are performed in which rectangular anomalies of various sizes are inserted into the images. The performance of detection strategies based on pixel intensity deviations (PIDs) and gamma indices is evaluated using receiver operating characteristic analysis. RESULTS: Residual differences between registered images are due to interfraction positional deviations of jaws and multileaf collimator leaves, plus imager noise. Positional deviations produce large intensity differences that degrade anomaly detection. Gradient effects are suppressed in PIDs using gradient scaling. Background noise is suppressed using median filtering. In the majority of images, PID-based detection strategies can reliably detect fluence anomalies of ≥5% in ∼1 mm(2) areas and ≥2% in ∼20 mm(2) areas. CONCLUSIONS: The ability to detect small dose differences (≤2%) depends strongly on the level of background noise. This in turn depends on the accuracy of image registration, the quality of the reference image, and field properties. The longer term aim of this work is to develop accurate and reliable methods of detecting IMRT delivery errors and variations. The ability to resolve small anomalies will allow the accuracy of advanced treatment techniques, such as image guided, adaptive, and arc therapies, to be quantified.

Circulating ghrelin is sensitive to changes in body weight during a diet and exercise program in normal-weight young women.

Ghrelin is directly involved with short-term regulation of energy balance. Although circulating levels of ghrelin are elevated in anorexia nervosa and reduced in obesity, the role of ghrelin in regulating long-term energy balance in healthy women has not been investigated. We examined the effects of a 3-month energy deficit-imposing diet and exercise intervention on circulating ghrelin in normal-weight, healthy women. Body composition, resting metabolic rate, and serum ghrelin were measured at pre-, mid-, and postintervention in controls (n = 7), who performed no exercise, and exercising women who remained weight stable (n = 5) or lost weight (n = 10). Exercise training occurred five times per week, and subjects were fed a specific diet. Ghrelin significantly increased over time (770 +/- 296 to 1322 +/- 664 pmol/liter) in the weight-loss group compared with the controls and the weight-stable group (P < 0.05). Changes in ghrelin were negatively correlated with changes in body weight (r = -0.61; P < 0.05). Body fat, body weight, and resting metabolic rate significantly decreased in the weight-loss group before the increase in ghrelin. These findings suggest that ghrelin responds in a compensatory manner to changes in energy homeostasis in healthy young women, and that ghrelin exhibits particular sensitivity to changes in body weight.

Cardiac rehabilitation without exercise tests for post-myocardial infarction and post-bypass surgery patients.

PURPOSE: To compare the progress of patients who were exercise tested before or during cardiac rehabilitation versus those patients who were not tested. METHODS: Eighty-eight (88) post-myocardial infarction patients and 141 post-bypass surgery patients had a symptom-limited exercise test before or during 12 weeks of cardiac rehabilitation. Another 125 post-myocardial infarction and 146 post-surgery patients were not tested. RESULTS: Caloric expenditure during class increased for the entire group (P < 0.001) from week 1 to week 12. Body weight decreased for the entire group as a result of cardiac rehabilitation (P < 0.001). Tricep skinfolds decreased for the entire group (P < 0.001) while subscapular skinfolds did not change (P = 0.28). The percent change from week 1 to week 12 for both groups was similar for all variables. No problems occurred during cardiac rehabilitation that required emergency medical management. CONCLUSIONS: Patients completing a 12-week cardiac rehabilitation program can be safely progressed in terms of their exercise capacity without an entry exercise test. This is desirable in a managed-care setting for reducing costs while maintaining effective patient care. Such factors as staff training and experience, institutional philosophy, patient referral patterns, and facility location must be considered before adopting a no-test policy.

Homogeneous catalogs of earthquakes.

The usual bias in earthquake catalogs against shocks of small magnitudes can be removed by testing the randomness of the magnitudes of successive shocks. The southern California catalog, 1933-1967, is found to be unbiased in the sense of the test at magnitude 4 or above; the cutoff is improved to M = 3 for the subcatalog 1953-1967.