Critical analysis of the impact of glottal features in the classification of clinical depression in speech.

The motivation for this work is in an attempt to rectify the current lack of objective tools for clinical analysis of emotional disorders. This study involves the examination of a large breadth of objectively measurable features for use in discriminating depressed speech. Analysis is based on features related to prosodics, the vocal tract, and parameters extracted directly from the glottal waveform. Discrimination of the depressed speech was based on a feature selection strategy utilizing the following combinations of feature domains: prosodic measures alone, prosodic and vocal tract measures, prosodic and glottal measures, and all three domains. The combination of glottal and prosodic features produced better discrimination overall than the combination of prosodic and vocal tract features. Analysis of discriminating feature sets used in the study reflect a clear indication that glottal descriptors are vital components of vocal affect analysis.

Three-dimensional fusion of coronary arteries with myocardial perfusion distributions: clinical validation.

UNLABELLED: Clinical decisions regarding diagnosis and effective treatment of coronary artery disease frequently require integration of information from various imaging modalities, which are acquired, processed, and read at different physical locations and times. We have developed methods to integrate the information in 2 cardiac imaging studies, perfusion SPECT and coronary angiography. Three-dimensional (3D) models of the coronary artery tree created from biplane angiograms were automatically aligned with 3D models of the left ventricular epicardial surface created from perfusion SPECT. Myocardial mass at risk was used as a unique measure to validate the accuracy of the unification. METHODS: Thirty patients were injected with the perfusion agent (99m)Tc-tetrosfosmin during balloon occlusion while undergoing percutaneous transluminal coronary angioplasty for single-vessel coronary artery disease. Thus, a single, severe perfusion defect was induced by a single coronary artery occlusion of known severity and placement. The accuracy of the unification was measured by computing the overlap between physiologic area at risk, determined using SPECT perfusion quantification techniques only, and anatomic area at risk, determined using coronary artery anatomy aligned with the epicardial surface of the left ventricle. RESULTS: The unification resulted in an 80% overlap of areas at risk, and an overlap of 84% of normal areas, for all coronary artery distributions. The mass at risk measured based on the unified anatomic information correlated with the physiologically based mass at risk as y = 0.92x + 10.3 g; r = 0.76, SEE = 10.4 g. CONCLUSION: A unification algorithm for automatically registering 3D models of the epicardial surface from perfusion SPECT and 3D coronary artery trees from coronary angiography has been presented and validated in 30 patient studies.