LONI: Laboratory of Neuro Imaging

Integrated Atlases

We published a broad range of work focused on developing new mathematical approaches for mapping brain development and disease. These expand the probabilistic atlas concept to include large and growing MRI data components from diseased populations.

Methamphetamine and the Brain

We applied computational anatomy techniques to reveal, for the first time, how methamphetamine affects brain structure. Using brain MRI scans from methamphetamine users and healthy control subjects, we created the first 3D maps of the changes in brain structure caused by chronic use of the stimulant. These findings revealed a remarkably selective profile of damage to the brain's limbic system, which is involved in reward and drug craving. Cortical gray matter deficits were greatest in the hippocampus, where deficits were significantly associated with poorer memory performance. Deficits in the cingulate region reached 11%, and their 3D profile was mapped for the first time. Also a diffuse profile of white matter inflammation was detected, which we are now pursuing further, to establish its regional specificity.

Dynamic Atlasing of Pediatric MRI: Time-Lapse Maps of Brain Development

Major progress was made in our longitudinal study of brain development, which charts the trajectory of human brain development using MRI. We created the first dynamic map of brain changes that occur in childhood and adolescence. This project scanned 13 children with brain MRI every 2 years for 10 years. We developed new mathematics based on random effects models and cortical pattern matching to map how cortical gray matter changes over time, resulting in a time-lapse animation of human cortical development.

Brain development was visualized across the age range 4 though 21 in a spatiotemporally-detailed time-lapse sequence. The resulting time-lapse revealed that: 1) higher-order association cortices mature only after lower-order somatosensory and visual cortices whose functions they integrate and 2) phylogenetically older brain areas mature earlier than newer ones. We are now developing the mathematics to compare developmental trajectories from abnormally developing children with the normative database, focusing on childhood onset schizophrenia, and bipolar illness.

Mapping HIV/AIDS Effects on Brain Structure

We began a new collaboration with an established group of AIDS researchers at the University of Pittsburgh. This is the first use of a probabilistic atlas that links brain structure with immune system measures. We created the first detailed maps revealing how HIV/AIDS impacts the human brain.

Using a new brain mapping technique we developed, we identified the brain regions that are most vulnerable to HIV. We found that thinning of the language cortex correlated with immune system deterioration measured via blood levels of CD4+ T-lymphocytes. Specific patterns of tissue loss-with up to 15% loss visualized in the association areas of the brain-correlated with cognitive and motor deficits. Based on this work, it is therefore now feasible to associate T-lymphocyte cell depletion, and cognitive impairment, with

specific brain deficit patterns visualized with MRI. These quantitative MRI-based maps reveal the pattern of damage caused by HIV in the brain.

Mapping Genotype-Phenotype Effects in Williams syndrome

We began to collaborate with research groups at the Salk Institute and Stanford to map the profile of deficits in Williams syndrome. Applying novel brain mapping techniques we developed, and statistically averaging cortical thickness maps from 166 brain hemispheres, we detected and mapped a sharply delimited region of language cortex with increased cortical thickness and gyral complexity. The isolated, thickened cortical region in language areas is remarkable because Williams subjects show remarkable strengths in language function. It is therefore a significant lead in the neuroscience of Williams syndrome, as it reveals the cortical territory affected by the genetic deletion. The affected sector of language cortex is 5-10% thicker and has higher complexity. The findings visualize cortical zones where the WS genetic deletion selectively disrupts gyrification, shedding light on the anatomical scope and timing of the structural deficits in utero, and their behavioral and cognitive sequelae. Discovering the territory affected clarifies which systems develop abnormally in Williams syndrome; importantly, these cortical deficits are linked with cognitive differences and can be mapped in living patients.