Digital chest radiography: effect of temporal subtraction images on detection accuracy.

PURPOSE: To improve early detection of disease in chest radiographs, the authors developed a digital processing technique that geometrically warps and subtracts a previous radiograph from a current radiograph to produce a temporal subtraction image. An observer test was performed to evaluate the effects of the temporal subtraction image technique on detection of interval change. MATERIALS AND METHODS: Fifty pairs of chest radiographs, including a baseline examination and a subsequent radiograph, were selected (25 cases in which potentially important new abnormalities had developed, and 25 in which there was no interval change). The baseline examination was chosen from multiple prior radiographs to minimize initial misregistration. By means of receiver operating characteristic (ROC) analysis, the ability of 11 observers to detect pathologic change when viewing the paired digitized baseline and subsequent radiographs was compared with their ability when viewing the same paired radiographs together with temporal subtraction images. Positive cases demonstrated focal new abnormalities that were greater than 1 cm in diameter. RESULTS: The mean area (Az) under the ROC curves increased from 0.89 without to 0.98 with the temporal subtraction images. When the paired digitized previous and current chest radiographs were viewed in conjunction with the temporal subtraction images, a significant improvement in detection of new abnormalities was achieved (P = .00004), whereas the mean interpretation time was reduced by 19.3% (from 52 to 42 seconds, including the time to record the score and to move to the next case) (P = .0019). CONCLUSION: The temporal subtraction technique can significantly improve sensitivity and specificity for detection of interval change in chest radiographs.

Glutamate receptors in the substantia nigra of Parkinson's disease brains.

Glutamic acid and its analogs are excitotoxins that might contribute to the pathogenesis of Parkinson's disease (PD). We measured four subtypes of glutamate binding sites autoradiographically in 20-microns sections from control and PD midbrains. N-Methyl-D-aspartate (NMDA) binding sites (eight control, eight PD) were very low in control (20 +/- 7 [SEM] fmol/mg protein) and were reduced in the PD pars compacta (2.6 +/- 1.1 fmol/mg protein; p less than 0.02). NMDA binding was also reduced in the red nucleus but not in periaqueductal gray (PAG). We measured alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA), metabotropic, and non-NMDA, nonkainate, non-quisqualate (NNKQ) sites in 10 PD and 12 control midbrains. AMPA binding sites were reduced from 175 +/- 20 to 99 +/- 16 (p less than 0.05) fmol/mg protein in PD pars compacta, NNKQ sites from 86 +/- 10 to 50 +/- 12 (p less than 0.05) fmol/mg protein in total nigra, and metabotropic sites (15 +/- 5 fmol/mg protein) were unchanged. AMPA, metabotropic, and NNKQ binding were unchanged in red nucleus and PAG. The very low number of NMDA binding sites suggests that factors other than excitotoxicity mediated via NMDA receptors on nigral cell bodies play roles in the pathogenesis of PD. There may be a generalized loss of NMDA receptors in PD brains. AMPA and NNKQ binding sites appear to be located on dopamine neurons, although the role of NNKQ sites in normal nervous system function and human disease is unknown.