The use of deep learning with neuroimaging for diagnosing and monitoring disease

Despite clear clinical necessity, many neurodegenerative diseases lack comprehensive, generalizable tools for reliable diagnosis, progression monitoring, and personalized approaches to disease management. Quantitative medical imaging holds rich, patient-specific disease information which may address these clinical deficits, but it is rarely used to its full potential due to challenges with inherent uncertainties and time-consuming data curation. To address these challenges, the field is moving beyond qualitative, manual image interpretation towards a quantitative data science approach to patient management which considers both imaging and clinical data. Advanced computational techniques, particularly in deep learning, can detect hidden quantitative imaging features, efficiently maximize information output from neurological images, and synthesize this information with clinical data to detect, predict, and model disease. This appreciation for quantitative science is opening opportunities for medical physics to advance the understanding of neurodegenerative disease and enhance treatment decision-making in the clinic.

In this talk, I will discuss how deep learning models are being employed using neuroimaging inputs, such as MRI, ¹⁸F-FDG PET, and Aβ-PET, to help manage neurodegenerative disease. I will review current areas of ongoing research on the use of DL in the neuroimaging space and will show examples from our work using a combination of convolutional and recurrent neural networks to capture both spatial and temporal disease patterns on ¹⁸F-FDG PET and T1-weighted MRI in the diagnosis of Alzheimer's disease and the detection of chemotherapy-related cognitive impairment, each in comparison to normal brain aging patterns. I will also place the great potential of deep learning techniques in context, discussing its limitations and what is still needed for responsible clinical translation.