AI-powered brain microstructure imaging

This fellowship develops innovative computational methods and magnetic resonance (MR) techniques to reveal new non-invasive markers of brain microstructure. The ultimate goal is to provide non-invasive tools for improved diagnostic information as powerful as invasive techniques.

Nature has built one of the most extraordinary machines we could ever conceive: the brain, using just basic cellular components of neurons and glia. Understanding how these individual components are designed (cell morphology) and assembled together (tissue microstructure) is the key to understanding both brain's structure and function, and more importantly its degeneration/dysregulation in diseases. However, it is currently impossible to quantify tissue microstructure in a non-invasive way. In fact, standard methods like histology can reveal microscopic characteristics of tissue architecture but at the cost of invasive interventions like biopsies and limited coverage of the investigated tissue, undermining diagnostic power. In contrast, sensitizing the MR imaging (MRI) contrast to water diffusion process, state-of-the-art technologies based on diffusion MRI (dMRI) provides an indirect but non-invasive probe of the tissue microstructure at the micrometer scale. This makes the acquired dMRI signal sensitive to tissue features like cellular size/shape, density, etc. Unfortunately, the tissue microstructure is highly complex while the dMRI signal is quite simple, so the mapping from signal to microstructure (inverse problem) is ill-posed. This represents a major obstacle for dMRI based techniques to replace histology as harmless diagnostic tools.

To overcome these limitations, the current paradigm of microstructure imaging uses mathematical models, which relate the dMRI signal to underlying tissue properties, to estimate and map those properties by fitting the models voxel-by-voxel to dMRI data. However, there are three main limitations hampering its diagnostic power: poor sensitivity to complex tissue features that cannot be described by mathematical models; a lack of specificity to different cell types, and ambiguity in the inverse problem's solution.

In this fellowship I propose a three-component shift of paradigm to address these key limitations: (i) employing detailed simulation of the tissue architecture to encode the forward problem (from tissue microstructure to MR signal), (ii) modern AI to solve the inverse problem and (iii) estimate of uncertainty to quantify ambiguity and significance of the results. I will demonstrate this in the brain by simulating normal tissue environments and those representing pathologies of common neurological diseases, like Multiple Sclerosis (MS) and Alzheimer's disease (AD). This will provide higher sensitivity to novel and important microstructural features like cell soma size/density and neurites size/complexity, leading to a new generation of quantitative imaging techniques based on water diffusion. To gain unprecedented specificity to different cell types in the brain and develop a new set of imaging markers for neurological conditions, I will combine the new paradigm with metabolites' diffusion measurements using dMR spectroscopy (dMRS). Indeed, metabolites are more cell-specific molecules than water: some are found mostly in neurons, others mostly in glia. I will prototype and validate the new technologies in controlled animal models of MS and AD and eventually provide proof-of-concept application in human patients. These innovative techniques offer great promise in the decades to come for the realisation of 'virtual histology' across a wide range of medical applications.

Although the fellowship focuses on neurological diseases, it also aims to initiate follow-on projects to explore other applications, like body cancer. Alternative contrast methods will extend the methods to other MR modalities beyond diffusion for complementary and additional information on healthy and diseased tissues.

Neurological disorders like Multiple Sclerosis (MS) and Alzheimer's disease (AD) carry a significant burden to the individual, their families and carers, the National Health System (NHS), and to society as a whole. People with these conditions have the lowest health-related quality of life of any long-term condition. It has been estimated that 71% of patients experience severe pain or discomfort and 70% are restricted in their activities, with tremendous impact on their and their carers lifes (Falling Short 2016). The Hospital Activity Compendium has estimated over 1.5M hospital admissions and about 1M emergency admissions due to neurological disorders in 2016/17 (>20% increase over the five years to 2016/17). Mainly due to aeging population, the total number of neurological cases in England has now reached 14.7 million (Neuro Numbers 2019), and the number of deaths relating to neurological disorders rose by 39% over 13 years (Public Health England's 2018 Neurology Mortality). Resulting national costs are consistent: NHS spent £3.3 billion in 2012-13 (3.5% of NHS spend) and 14% of the social care budget on people living with neurological conditions (Neuro Numbers 2019). Main reason behind these numbers is that it is not always possible for patients who are experiencing neurological symptoms to get a diagnosis. Some people may have to wait a long time before the cause of their symptoms is identified, while others may not be able to get a diagnosis, and the prognosis as well as treatment options often follow a diagnosis. Thus, providing more effective tools for an earlier and more accurate diagnosis of neurological diseases is thus crucial to make an impact on patients life, their families and carers, and on society as a whole.

This project develops non-intrusive magnetic resonance imaging (MRI) methods to provide improved diagnostic information as powerful as invasive techniques. Primarily, this will ameliorate patient's quality of life, while assisting better clinical detection, staging and diagnosis by earlier and enhanced knowledge of what is happening at the tissue level. In particular, this project will ultimately give doctors powerful tools to diagnose disease and to test how well a particular therapy may be working. Improved characterization of the underpinning tissue microstructure can reduce the number of undiagnosed/misdiagnosed cases and support more accurate treatment planning. Better diagnosis can also decrease the cost of unnecessary or wrong treatments and contribute to the sustainability of the healthcare system by drastically reducing the number of emergency admissions.

The new microstructure MRI methods powered with artificial intelligence (AI) and information on uncertainty will provide new insight into the biology of neurodegenerative diseases, e.g. concerning the mechanisms of demyelination and glial scar in MS, neuronal atrophy, gliosis and protein entanglement in AD, and subtle or difficult to detect effects such as distinguishing changes in neuron and glial fine microstructure (soma size/density, neurites complexity), encouraging exciting new research prospects. Further exploitation of such computational methods can lead to identification of disease subgroups that when correlated with genetics could lead to the design of individualised therapies, achieving patient stratification for precision medicine.

The development of new AI-powered MRI techniques can impact the commercial and industrial sector, such as MRI manufacturers and software companies, offering opportunities that can drive future development of medical hardware/software. Siemens Healthineers and AInostics have already expressed their strong support and interest in the project. Finally, AI-powered microstructure imaging can also inform drug development and trials, revealing the effects of treatment earlier, moving trials more quickly to the final stages and accelerating the availability of new promising cancer treatments to patients.