The human brain holds many clues about a person’s long-term health research shows that a person’s brain age is a more useful and accurate predictor of health risks and future disease than their birthdate. A new artificial intelligence model that analyzes MRI brain scans developed by USC researchers could be used to accurately capture cognitive decline linked to neurodegenerative diseases like Alzheimer’s much earlier than previous methods.  

Brain aging is considered a reliable biomarker for neurodegenerative disease risk. Such risk increases when a person’s brain exhibits features that appear “older” than expected for someone of that person’s age. By tapping into the deep learning capability of the team’s novel AI model to analyze the scans, the researchers can detect subtle brain anatomy markers that are otherwise very difficult to detect and that correlate with cognitive decline. Their findings offer an unprecedented glimpse into human cognition.  

“Our study harnesses the power of deep learning to identify areas of the brain that are aging in ways that reflect a cognitive decline that may lead to Alzheimer’s,” said Andrei Irimia, assistant professor of gerontology, biomedical engineering and neuroscience at the USC Leonard Davis School of Gerontology and corresponding author of the study.  

“People age at different rates, and so do tissue types in the body,” Irimia said. “We know this colloquially when we say, ‘So-and-so is 40, but looks 30.’ The same idea applies to the brain. The brain of a 40-year-old may look as ‘young’ as the brain of a 30-year-old, or as ‘old’ as that of a 60-year-old.”  

Brain aging: A more accurate alternative to existing methods  

Researchers collated the brain MRIs of 4,681 cognitively normal participants, some of whom went on to develop cognitive decline or Alzheimer’s disease later in life.  

Using these data, they created an AI model called a neural network to predict participants’ ages from their brain MRIs. First, the researchers trained the network to produce detailed anatomic brain maps that reveal subject-specific patterns of aging. They then compared the perceived (biological) brain ages with the actual (chronological) ages of study participants. The greater the difference between the two, the worse the participants’ cognitive scores, which reflect Alzheimer’s risk. 

The results show that the team’s model can predict the true (chronological) ages of cognitively normal participants with an average absolute error of 2.3 years, which is about one year more accurate than an existing, award-winning model for brain age estimation that used a different neural network architecture.  

“Interpretable AI can become a powerful tool for assessing the risk for Alzheimer’s and other neurocognitive diseases,” said Irimia, who also holds faculty positions with the USC Viterbi School of Engineering and the USC Dornsife College of Letters, Arts and Sciences. 

“The earlier we can identify people at high risk for Alzheimer’s disease, the earlier clinicians can intervene with treatment options, monitoring, and disease management.”

Brain aging: differences according to sex  

The new model also reveals sex-specific differences in how aging varies across brain regions. Certain parts of the brain age faster in males than in females, and vice versa.  

Males, who are at higher risk of motor impairment due to Parkinson’s disease, experience faster aging in the brain’s motor cortex, an area responsible for motor function. Findings also show that, among females, typical aging may be relatively slower in the right hemisphere of the brain.  

An emerging field of study shows promise for personalized medicine  

Applications of this work extend far beyond disease risk assessment. Irimia envisions a world in which the novel deep learning methods developed as part of the study are used to help people understand how fast they are aging in general.  

“One of the most important applications of our work is its potential to pave the way for tailored interventions that address the unique aging patterns of every individual,” Irimia said.  

“Many people would be interested in knowing their true rate of aging. The information could give us hints about different lifestyle changes or interventions that a person could adopt to improve their overall health and well-being. Our methods could be used to design patient-centered treatment plans and personalized maps of brain aging that may be of interest to people with different health needs and goals.” 

The use of artificial intelligence (AI) in the fashion industry has grown significantly in recent years. AI is being used for tasks such as personalizing fashion recommendations for customers, optimizing supply chain management, automating processes, and improving sustainability to reduce waste. However, creative processes in fashion design continue to be human-driven, mostly, and not a lot of research exists in the realm of using AI for designing in fashion. Moreover, studies are generally done with data scientists, who build the AI platforms and are involved with the technological aspect of the process. However, the other side of this equation, i.e., designers themselves, are not roped into research often.

To investigate the practical applicability of AI models to implement creative designs and work with human designers, Assistant Professor Prof. Yoon Kyung Lee from Pusan National University in Korea conducted an in-depth study. Her study was made available online in Thinking Skills and Creativity on September 15, 2022, and subsequently published in Volume 46 of the Journal in December 2022.

“At a time when AI is so deeply ingrained into our lives, this study started instead with considering what a human can do better than AI,” says Prof. Lee, explaining her motivation behind the study. “Could there be an effective collaboration between humans and AI for the purpose of creative design?”

Prof. Lee started by generating new textile designs using deep convolution generative adversarial networks (DC-GANs) and cycle-GANs. The outputs from these models were compared to similar designs produced by design students.

The comparison revealed that though designs produced by both were similar, the biggest difference was the uniqueness and originality seen in the human designs, which came from the person’s experiences. However, the use of AI in repetitive tasks can improve the efficiency of designers and frees up their time to focus on more high-difficulty creative work. AI-generated designs can also be used as a learning tool for people who lack expertise in fashion and want to explore their creativity. These people can create designs with assistance from AI.  Thus, Prof. Lee proposes a human-AI collaborative network that integrates GANs with human creativity to produce designs. The professor also defined and studied the various elements of a complex system that are involved in the human-AI collaborated design. She also went on to establish a human-AI model in which the designer collaborates with AI to create a novel design idea. The model is built in such a way that if the designer shares their creative process and ideas with others, the system can interconnect and evolve, thereby improving its designs.

The fashion industry can leverage this to foresee changes in the fashion industry and offer recommendations and co-creation services. Setting objectives, variables, and limits is part of the designer's job in the Human-AI collaborative design environment. Therefore, their work should go beyond only the visual aspect and instead cover a variety of disciplines.

“In the future, everybody will be able to be a creator or designer with the help of AI models. So far, only professional fashion designers have been able to design and showcase clothes. But in the future, it will be possible for anyone to design the clothes they want and showcase their creativity,” concludes Prof. Lee.

We hope her dreams are very close to realization!