High fidelity tonotopic mapping using swept source functional magnetic resonance imaging
Highlights
► We developed an efficient swept source imaging (SSI) fMRI for tonotopic mapping. ► SSI integrates continuous frequency sweeping, bSSFP sequence and Fourier analysis. ► High resolution tonotopic mapping using SSI is demonstrated in inferior colliculus. ► Tonotopic injury following developmental noise exposure is studied using SSI. ► Varying sound pressure level allows SSI to probe local auditory neuronal response.
Introduction
Humans and many animal species rely on effective hearing to survive and prosper in a competitive world. The sense of hearing and the auditory system that processes acoustic information are critical for performing core functions such as communicating, finding food and avoiding danger. The auditory system is very adept at distinguishing among frequency components in a sound, which allows us to hear the fine differences in words and allows animals to distinguish the sounds of predators and prey. For example, humans can distinguish two pure tone sounds that differ by only 2 to 6 Hz even though the sounds are in the frequency range of 1 to 4 kHz (Longstaff, 2005, Malmierca et al., 2009). To enhance our abilities to perform these vital functions or to treat auditory disorders such as tinnitus (Muhlnickel et al., 1998) that reduce quality of life, we need to advance our understanding of frequency encoding in the auditory system.
Much of our knowledge of auditory function and frequency encoding comes from invasive studies (Ehret and Fischer, 1991). Sound pressure waves enter the ear and vibrate the ear drum. These vibrations are transmitted to the basilar membrane, whose motions produce periodic depolarization and hyperpolarization of hair cells (Longstaff, 2005), sending neuronal signals to the cochlear nucleus of the central auditory pathway. Axons carrying neuronal signals run from the cochlear nucleus to the superior olivary complex, then the lateral lemniscus, inferior colliculus (IC), medial geniculate body and the auditory cortex (Malmierca, 2003). At each structure of the auditory pathway, electrophysiology studies have observed that the majority of neurons have sharp frequency tuning curves, meaning these neurons are most sensitive to a narrow spectrum of sounds centered about a characteristic frequency (CF).
The findings of invasive electrophysiology and immunohistochemistry studies also suggest that neurons in a structure with similar CFs are positioned close together (Ehret and Fischer, 1991). This indicates the presence of tonotopic organization, a topographic encoding of frequency. The ideal technique for studying tonotopy would be sensitive to neuronal activity and be capable of mapping a large field of view (FOV) with high spatial and frequency resolution. Unfortunately, the traditional techniques are not ideal. Electrophysiological recordings cannot achieve the continuous spatial coverage and large FOV needed to thoroughly study tonotopy. Immunohistochemistry techniques would require an infeasible number of animals to cover a broad frequency range and are difficult to use, if not inapplicable, in longitudinal investigations.
Functional magnetic resonance imaging (fMRI) has potential for tonotopic mapping. fMRI using the blood oxygenation level-dependent (BOLD) signal for endogenous contrast is widely used in brain mapping (Bilecen et al., 1998b, Cheung et al., 2012, Ogawa et al., 1990). BOLD fMRI is noninvasive and applicable to longitudinal human and animal studies. It is typically implemented with echo planar imaging (EPI) sequences, which provide adequately high spatial resolution (~ 1 mm in clinical scanners). However, EPI emits sporadic acoustic noise which adversely affects auditory fMRI (Seifritz et al., 2006). Auditory studies typically employ sparse temporal sampling paradigms, which use EPI sequences with long repetition time (on the order of 10 s), to reduce the adverse effects of scanner noise (Hall et al., 1999). This technique has proven useful, but it is highly time inefficient and like conventional EPI, images suffer from distortion and susceptibility induced signal loss (Jezzard and Balaban, 1995), which hamper studies in many fine structures in the auditory system. These limitations become more apparent at high magnetic fields and will restrain the growth of high resolution auditory fMRI.
Stimulation in fMRI tonotopy studies is typically presented in block-design paradigms (Baumann et al., 2011, Bilecen et al., 1998a). Block-design involves presenting a pure tone sound to the subject in an on–off pattern and using statistical analysis to identify brain regions where the BOLD signal correlates with the stimulus on–off timing. Recently, we applied a block-design paradigm to map tonotopic organization in the rat IC and map the ascending auditory pathway (Cheung et al., 2012). Block-design provides high statistical power and sensitivity. However, it cannot map tonotopic organization with high frequency resolution as only a limited number of pure tones can be presented in a study session. All together, the conventional fMRI techniques for tonotopic mapping using EPI and block-design suffer from image distortion, signal loss and low frequency resolution.
We develop a novel fMRI technique named magnetic resonance swept source imaging (SSI) that maps the tonotopic organization of auditory structures with high spectral and spatial resolution. Instead of EPI, SSI uses balanced steady state free precession (bSSFP), a fast MRI acquisition sequence that provides T2/T1 contrast without sparse temporal sampling (Lee et al., 2008, Zhou et al., in press), image distortion, susceptibility-induced signal loss and sporadic noise. Therefore, bSSFP avoids the time inefficiencies and image artifacts of EPI and is ideally suited for auditory fMRI studies. To improve upon block-design, SSI uses a frequency sweeping stimulation paradigm along with Fourier transformation analysis that maps tonotopic organization over a continuous frequency spectrum.
In this study, we first describe SSI and demonstrate its ability to map tonotopy in the rat IC. In normal animals, the new tonotopic maps show significantly higher frequency resolution and spatial fidelity compared with conventional fMRI maps. We subsequently apply SSI to study the IC of animals injured by early post-natal noise exposure (NE) and find that the tonotopic organization is significantly disrupted. We also observe with SSI the subtle effects of sound pressure level (SPL) on tonotopic maps, reflecting the neuronal responses associated with asymmetric tuning curves.
Section snippets
Methods
All animal experiments were approved by the local animal research ethics committee. Four different fMRI experiments were performed in this study: (1) mapping inferior colliculus tonotopy in normal animals with magnetic resonance swept source imaging; (2) studying the effect of increasing stimulus sound pressure level on the BOLD signal amplitude; (3) mapping tonotopic changes caused by early post-natal noise exposure; and (4) studying the effect of SPL on observed tonotopic maps. This section
Results
The findings of this study are organized according to the four experiments described in the Methods section.
Discussion
Swept source imaging mapped tonotopy with high fidelity at approximately 2 kHz resolution and 40 kHz bandwidth in 30 min. The resulting tonotopic maps from each animal yielded significantly higher frequency resolution and spatial specificity compared with maps acquired using conventional fMRI and were in excellent agreement with previous invasive findings (Huang and Fex, 1986, Webster et al., 1984). Using SSI, we observed that early post-natal noise exposure significantly disrupted the tonotopic
Acknowledgments
This work was supported by Hong Kong Research Grants Council (General research grants HKU7826/10M and HKU7837/11M to E.X.W.).
References (54)
- et al.
Tonotopic organization of the human auditory cortex as detected by BOLD-FMRI
Hear. Res.
(1998) - et al.
Tonotopic organization of the human auditory cortex as detected by BOLD-FMRI
Hear. Res.
(1998) - et al.
Functional MRI of postnatal visual development in normal and hypoxic–ischemic-injured superior colliculi
Neuroimage
(2010) - et al.
Balanced steady-state free precession with parallel imaging gives distortion-free fMRI with high temporal resolution
Magn. Reson. Imaging
(2011) - et al.
BOLD fMRI investigation of the rat auditory pathway and tonotopic organization
NeuroImage
(2012) - et al.
Reduction of gradient acoustic noise in MRI using SENSE-EPI
Neuroimage
(2002) - et al.
Neuronal activity and tonotopy in the auditory system visualized by c-fos gene expression
Brain Res.
(1991) - et al.
The inferior colliculus of the rat: a quantitative analysis of monaural frequency response areas
Neuroscience
(2005) - et al.
Non-plastic reorganization of frequency coding in the inferior colliculus of the rat following noise-induced hearing loss
Neuroscience
(2008) - et al.
BOLD responses in the superior colliculus and lateral geniculate nucleus of the rat viewing an apparent motion stimulus
Neuroimage
(2011)
The cytoarchitecture of the inferior colliculus revisited: a common organization of the lateral cortex in rat and cat
Neuroscience
The structure and physiology of the rat auditory system: an overview
Int. Rev. Neurobiol.
Computer-assisted 3-D reconstructions of Golgi-impregnated neurons in the cortical regions of the inferior colliculus of rat
Hear. Res.
Persistent effects of early augmented acoustic environment on the auditory brainstem
Neuroscience
Development of frequency-selective domains in inferior colliculus of normal and neonatally noise-exposed rats
Brain Res.
Modification of tonotopic representation in the auditory system during development
Prog. Neurobiol.
Enhancing BOLD response in the auditory system by neurophysiologically tuned fMRI sequence
Neuroimage
Increased blood oxygen level-dependent (BOLD) sensitivity in the mouse somatosensory cortex during electrical forepaw stimulation using a cryogenic radiofrequency probe
NMR Biomed.
Orthogonal representation of sound dimensions in the primate midbrain
Nat. Neurosci.
Noise exposure during early development impairs the processing of sound intensity in adult rats
Eur. J. Neurosci.
Changing tune in auditory cortex
Nat. Neurosci.
Critical period window for spectral tuning defined in the primary auditory cortex (A1) in the rat
J. Neurosci.
Spectral and intensity coding in the auditory midbrain
Spatial map of frequency tuning-curve shapes in the mouse inferior colliculus
Neuroreport
Retinotopic organization in human visual cortex and the spatial precision of functional MRI
Cereb. Cortex
Anatomy of the inferior colliculus in rat
Anat. Embryol. (Berl)
Development of afferent patterns in the inferior colliculus of the rat: projection from the dorsal nucleus of the lateral lemniscus
J. Comp. Neurol.
Cited by (26)
BOLD-fMRI in the mouse auditory pathway
2018, NeuroImageThe multi-level impact of chronic intermittent hypoxia on central auditory processing
2017, NeuroImageCitation Excerpt :Higher up, the medial geniculate body of the thalamus and the auditory cortex are also important auditory centers. Our group has pioneered novel fMRI techniques for investigating the rat central auditory system (Cheung et al., 2012a, 2012b; Gao et al., 2015a, 2014, 2015b; Lau et al., 2015a, 2013, 2015b; Zhang et al., 2013). The rat auditory system is similar to that in humans (Malmierca, 2003).
Chronic exposure to broadband noise at moderate sound pressure levels spatially shifts tone-evoked responses in the rat auditory midbrain
2015, NeuroImageCitation Excerpt :The 7 kHz stimulus was within the 30 kHz bandwidth of the exposure while the 40 kHz stimulus was outside of the bandwidth. Both stimulation frequencies have been employed in our earlier studies and lead to robust responses with high contrast-to-noise in the auditory system (Cheung et al., 2012a,b). Each tone was pulsed at 4 Hz with 80% duty cycle (each pulse was 200 ms in duration).
Subcortical processing in auditory communication
2015, Hearing Research
- 1
Authors Matthew M. Cheung and Condon Lau contributed equally to this work.