The transformative potential of artificial intelligence

Imagine a world where doctors can predict your risk of dementia years in advance, taking proactive measures to mitigate that risk. Envision a reality in which advances in neuroimaging, combined with artificial intelligence (AI), allow physicians to assess brain function and emotional health, including conditions such as depression. What if the futuristic scenes depicted in the movie, Minority Report, became feasible – with access to machines that can interpret your thoughts and experiences?

The convergence of neuroimaging and AI marks a rapidly growing field with immense potential for breakthroughs in computational neuroscience. Techniques such as magnetic resonance imaging (MRI), functional MRI and electroencephalography (EEG) produce vast amounts of complex data that, when analysed through AI, can yield profound insights into the structure and function of the human brain. This synergy is gradually helping medical professionals transform these aspirations into reality.

The strain of dementia

In many developed nations, life expectancy is rising, bringing with it significant health challenges, particularly among the elderly. Deteriorating mental faculties, leading to conditions such as dementia, represent a formidable challenge. The financial and social burdens on families and nations as the prevalence of dementia rises are daunting.

By 2050, the percentage of people aged 65 and older is expected to double, escalating from 9.3 per cent to over 16 per cent of the global population. One of the primary diagnostic tools used to assess brain health and dementia is MRI. Recent advancements in integrating AI with MRI technology hold promise for medical professionals seeking improved outcomes for patients experiencing cognitive decline.

Enhancements in MRI technology

MRI is one of the most sophisticated medical imaging techniques available, utilising strong magnets to generate detailed anatomical and functional images. While it is a safe and non-invasive method that does not employ X-ray radiation, many patients experience anxiety lying still in a narrow tunnel for extended periods. This anxiety can lead to movement, compromising scan quality. A 2015 study published in the Journal of the American College of Radiology estimated that about 20 per cent of MRI scans necessitate repetition – an expensive and time-consuming process for both healthcare providers and patients.

Fortunately, advancements in AI technology are revolutionising the patient experience during MRI scans. Modern MRI machines are now equipped with wider openings and smart, AI-driven algorithms that can automatically adjust for patient positioning, ensuring comfort. For instance, German company Siemens’ “Deep Resolve Boost” integrates advanced deep-learning image reconstruction with acceleration techniques, enhancing MRI scan speeds by up to 72 per cent while maintaining high-resolution images. This innovation allows for a full neurological examination of the brain to be completed in under two minutes, significantly improving clinical efficiency.

A NEWSLETTER FOR YOU

Friday, 2 pm

Lifestyle

Our picks of the latest dining, travel and leisure options to treat yourself.

AI technology also adapts to patients’ unique physiological and anatomical variations during imaging, addressing challenges related to complex anatomies and physiological motion (such as breathing and heartbeat). Siemens’ BioMatrix technology automatically adjusts scans to ensure high-quality images are captured, reducing the need for repeat scans.

Early detection of cognitive impairment

Longitudinal studies reveal that brain atrophy can commence as early as the 40s, with accelerating changes in volume into later decades. A recent study involving over 600 participants demonstrated that the frontal lobe, the region responsible for essential cognitive functions, exhibits a two-phase pattern of volume loss, with significant atrophy occurring before age 50 and from the 70s onwards. Recognising these age-related changes enables doctors to detect mild cognitive impairment earlier and implement preventive strategies to slow cognitive decline.

White matter hyperintensities (WMHs) are another indicator of brain health that can be detected through MRI. Primarily associated with ageing and vascular health, WMHs have been linked to depression, stroke, and cognitive impairment. Although the WMH prevalence is higher in older individuals, studies of healthy younger people have reported that the WMH degree is also correlated with the execution of speed-demanding functions, such as word recall.

AI has revolutionised the way segmented brain volumes and WMHs are quantified, enabling rapid and accurate assessments that can support clinical decisions regarding dementia prevention and treatment.

The quest to prevent dementia

AI-driven software has been developed to identify individuals at high risk for dementia. For example, the US Food and Drug Administration has approved AI-powered tools such as BrainSee and AIRAscore, which assist in earlier detection of Alzheimer’s disease and other dementias through quantitative brain volume analysis. These advanced detection methods can empower clinicians to promptly begin interventions that may delay dementia onset.

Assessing brain function

Functional MRI (fMRI) studies the living brain by measuring changes in blood oxygen levels during specific tasks or emotional states. Notably, fMRI has been instrumental in pharmacological research, particularly for evaluating the effectiveness of treatments for depression – one of the most prevalent psychiatric disorders globally. In cases of treatment-resistant depression (TRD), innovative therapies such as the AI-powered fMRI guided Stanford Accelerated Intelligent Neuromodulation Therapy personalised transcranial magnetic stimulation have shown remarkable success, with about 80 per cent of TRD patients achieving remission after just five days of treatment.

Though they may seem like concepts from science fiction, researchers have successfully trained AI systems to recreate images based on brain activity assessed through fMRI. Teams at the University of Texas have developed brain-computer interfaces that tap into language-producing areas of the brain to interpret imagined speech, analysing blood flow changes to decode neural activity into text. While these technologies are still in their infancy, they hold significant promise in aiding individuals who have lost their ability to communicate due to strokes or brain trauma.

A brighter future

The integration of AI into neuroimaging and neurology represents a paradigm shift in our approach to preventing and managing dementia. Enhanced diagnostic capabilities, earlier interventions and innovative monitoring tools can collectively help mitigate the impact of this escalating global health challenge. As these technologies advance, they have the potential to redefine mental healthcare, offering renewed hope for individuals and families confronting the realities of dementia.

This article is part of a monthly series on health and well-being, produced in collaboration with Royal Healthcare