Dr. Naman Jain

Functional magnetic resonance imaging (fMRI) has emerged as a popular tool for investigating brain function owing to its non-invasive nature. Hemodynamic changes occur in response to neural stimulation, resulting in alterations in cerebral blood flow, blood volume, and oxygenation. Such hemodynamic changes can indirectly measure neuronal activity using the Blood Oxygen Level Dependency (BOLD) signal, a very sensitive method to map brain activity. However, it suffers from poor spatial specificity due to the involvement of large pial vessels, making it challenging to pinpoint neuronal activity accurately. Alternative approaches utilise changes in cerebral blood volume (CBV) are considered more spatially specific, but these often use contrast agents such as superparamagnetic iron oxide nanoparticles (MION) that cannot easily be used in healthy human volunteers. Alternatively, Vascular Space Occupancy (VASO) fMRI - a method that uses blood as an endogenous contrast, is CBV-weighted but has low sensitivity.

In my PhD project, I set out to investigate MION and VASO CBV fMRI in mice to better understand changes in the blood volume in fMRI and investigate their characteristics with regard to spatial specificity. As a model, I propose to investigate CBV fMRI in mice with the challenge that animal fMRI studies need to employ anaesthesia to reduce motion while scanning, which impacts brain haemodynamics, i.e. vasodilatory or vasoconstrictive effects complicate the qualitative and quantitative analysis of blood volume changes. In my thesis, I developed a set of experiments to (a) demonstrate a T1-based positive CBV-fMRI contrast that allows mapping the contributions of the whole-brain vasculature in a mouse, (b) demonstrate the feasibility of VASO in the mouse brain, and (c) work towards awake animal imaging without the need of anaesthesia that alters CBV. My results show that by scanning at short echo times, we could see responses within all vascular compartments, pial and parenchymal, whereas at longer TE, we observed that the responses were predominantly within the parenchyma. I demonstrated that using a MION contrast agent combined with an ultra-short echo time (UTE) readout, no significant T2* dephasing is observed; thus, the T1 effects become relevant, resulting in T1-based CBV contrast showing fMRI signal changes in all vascular compartments. I show the feasibility of performing CBV using endogenous contrast, i.e., Vascular Space Occupancy (VASO) in mice. Upon stimulation, a negative signal change is observed with increasing Cerebral Blood Volume (CBV). Also, I worked towards functional imaging of mice in an awake condition, which poses several challenges due to the high acoustic noise caused by the scanner and the physical restraint of the animal. In the present study, I sought to evaluate multiple stress metrics associated with the use of restraint devices and exposure to elevated levels of MRI sound during a state of wakefulness, with concurrent monitoring of subject movement using the camera. I further developed the design and setup for habituating mice in a mock MRI scanner to closely replicate scanner conditions and support the habituation of the mice, as using a real scanner is impractical.

In conclusion, the results of this study suggest that (a) UTE imaging may be a valuable tool in detecting changes in both surface and deep cortical regions of the brain. However, further research is needed to better understand the mechanisms behind the observed results and to optimise the use of UTE imaging in clinical settings. (b) CBV-weighted VASO fMRI is possible in mice, but the sequences and parameters need further optimisation, such as implementing a dynamic division approach, which has been implemented by previous studies to reduce the effects of BOLD contamination. (c) Development of awake animal imaging, especially mice, has been challenging, better habituation protocol is needed, and the possible use of quieter MRI sequences can reduce stress.