Recovering sound sources from embedded repetition.

Cocktail parties and other natural auditory environments present organisms with mixtures of sounds. Segregating individual sound sources is thought to require prior knowledge of source properties, yet these presumably cannot be learned unless the sources are segregated first. Here we show that the auditory system can bootstrap its way around this problem by identifying sound sources as repeating patterns embedded in the acoustic input. Due to the presence of competing sounds, source repetition is not explicit in the input to the ear, but it produces temporal regularities that listeners detect and use for segregation. We used a simple generative model to synthesize novel sounds with naturalistic properties. We found that such sounds could be segregated and identified if they occurred more than once across different mixtures, even when the same sounds were impossible to segregate in single mixtures. Sensitivity to the repetition of sound sources can permit their recovery in the absence of other segregation cues or prior knowledge of sounds, and could help solve the cocktail party problem.

Does fundamental-frequency discrimination measure virtual pitch discrimination?

Studies of pitch perception often involve measuring difference limens for complex tones (DLCs) that differ in fundamental frequency (F0). These measures are thought to reflect F0 discrimination and to provide an indirect measure of subjective pitch strength. However, in many situations discrimination may be based on cues other than the pitch or the F0, such as differences in the frequencies of individual components or timbre (brightness). Here, DLCs were measured for harmonic and inharmonic tones under various conditions, including a randomized or fixed lowest harmonic number, with and without feedback. The inharmonic tones were produced by shifting the frequencies of all harmonics upwards by 6.25%, 12.5%, or 25% of F0. It was hypothesized that, if DLCs reflect residue-pitch discrimination, these frequency-shifted tones, which produced a weaker and more ambiguous pitch than would yield larger DLCs than the harmonic tones. However, if DLCs reflect comparisons of component pitches, or timbre, they should not be systematically influenced by frequency shifting. The results showed larger DLCs and more scattered pitch matches for inharmonic than for harmonic complexes, confirming that the inharmonic tones produced a less consistent pitch than the harmonic tones, and consistent with the idea that DLCs reflect F0 pitch discrimination.

Combined epicardial and endocardial electroanatomic mapping in a porcine model of healed myocardial infarction.

BACKGROUND: Substrate mapping of post-myocardial infarction ventricular tachycardia involves electroanatomic delineation of scarred tissue on the basis of electrogram characteristics during sinus rhythm. A percutaneous transthoracic technique was recently described that allows catheter mapping of the epicardial surface of the heart. This study sought to determine whether the epicardial extent of a myocardial infarct could be defined during sinus rhythm. METHODS AND RESULTS: In a porcine model of healed anterior wall myocardial infarction (n=13 animals), detailed in vivo left ventricular endocardial and ventricular epicardial electroanatomic mapping was performed. Catheter access to the pericardial space was achieved by subxyphoid puncture under fluoroscopic guidance. Bipolar electrogram amplitude and duration characteristics of normal tissue were established on the basis of in vivo epicardial mapping data in 8 additional normal animals. With the use of these criteria, radiofrequency lesions (4 to 11 per animal) were placed along the endocardial and epicardial scar borders as defined by the electroanatomic map. The area of epicardial scar defined by abnormal bipolar voltage correlated well with the dimensions measured on pathological examination. The size and location also correlated well with the scar dimensions defined by electrogram duration criteria. Late potentials were noted in the border zones of both surfaces of the scar. During pathological examination, the radiofrequency lesions were situated at the borders of the epicardial scar. CONCLUSIONS: A 3-dimensional construct of the infarcted myocardium can be rendered by combined epicardial and endocardial electroanatomic mapping. This experimental protocol is propaedeutic to future clinical studies incorporating endocardial and epicardial substrate mapping into catheter ablation strategies to treat post-myocardial infarction ventricular tachycardia.

False positive ST segment elevation during dobutamine stress echocardiography due to left ventricular hypertrophy.

The significance of ST segment elevation in dobutamine stress echocardiography (DSE) remains controversial. In patients with prior Q wave myocardial infarction (MI), it may reflect myocardial ischemia, contractile reserve in the infarct-related area, or dyskinesia of the infarcted areas of myocardium. In the nonpost-MI population, it has been attributed to vasospasm or strongly associated with coronary artery disease and ischemia. We hypothesized that ST segment elevation in the absence of inducible ischemia or prior MI is related to the presence of left ventricular hypertrophy (LVH). During DSE, dobutamine was infused from 5 microg/kg/min up to a maximum of 50 microg/kg/min. Echocardiographic images were obtained at baseline, low dose, peak dose, and recovery. Ischemia was defined as either the development of a new wall-motion abnormality or worsening wall motion at peak dose. We reviewed 682 consecutive DSE tests and found ST elevation in 42 patients (incidence = 6.1%). After excluding two patients for > 10% uninterpretable echocardiographic segments, the study population consisted of 40 patients. In 25 patients with ST elevation and without echocardiographic evidence for dobutamine-induced ischemia, 21 (84%) patients had LVH (P = 0.001). In 15 patients with inducible ischemia, only 4 (27%) patients had LVH. No other significant differences were found except that prior MI was more common in the inducible ischemia group. In the subgroup of 18 patients without prior MI, no inducible ischemia was found in 15 (83%). LVH was present in 14 (93%) of these 15 patients (P < 0.005), and all 14 had a normal baseline left ventricular ejection fraction. None of the three patients in the nonpost-MI subgroup with inducible ischemia had LVH. The 22 patients with prior MI had no significant association with LVH (P = 0.39). We conclude that ST segment elevation during DSE can occur without echocardiographic evidence for ischemia and is associated with LVH in the nonpost-MI population. This ST elevation may be related to transient electrocardiographic repolarization changes in the hypertrophied ventricle in the presence of altered loading conditions and/or altered catecholamine influences rather than true ischemia.

Use of electrogram characteristics during sinus rhythm to delineate the endocardial scar in a porcine model of healed myocardial infarction.

INTRODUCTION: Substrate-based catheter ablation of postmyocardial infarction (post-MI) ventricular tachycardia necessitates electroanatomic definition of the scarred endocardium. We sought to determine whether electrogram criteria during sinus rhythm could identify the location and extent of the myocardial scar by electroanatomic mapping. METHODS AND RESULTS: A porcine model of healed MI was generated by injecting agarose microspheres into the mid left anterior descending coronary artery. At least 4 weeks post-MI, the animals (n = 24) underwent detailed left ventricular endocardial electroanatomic mapping using a 4-mm-tip catheter (BioSense-Webster, Inc.). Based upon mapping data in normal animals, infarcted tissue was defined as bipolar electrogram amplitude < 1.5 mV and electrogram duration > or = 50 msec. Radiofrequency ablation lesions (2-10 per animal) were placed to tag the endocardial borders of the electroanatomic mapping-defined scar. The area of the scar defined by abnormal voltage amplitude was 25.9 +/- 15.4 cm2 (range 6.9-60.5). This area correlated well with that defined as scar by the electrogram duration criteria (26.4 +/- 16 cm2). Of those points remote from the infarct with falsely low voltage amplitude resulting from presumed poor catheter-tissue contact, 94% were correctly identified as normal when using the electrogram duration criteria. Late potentials were observed predominantly along the borders of the infarcted myocardium. The radiofrequency lesions placed to tag the scar borders were located along the scar periphery during gross pathologic examination. CONCLUSION: During normal sinus rhythm, both bipolar electrogram voltage amplitude and electrogram duration criteria are able to help differentiate normal from scarred myocardial tissue. Using these criteria, a detailed reconstruction of the endocardial scar can be rendered by electroanatomic mapping of the heart.