Using advanced AI to improve Alzheimer’s disease classification

Warning signs for Alzheimer’s disease (AD) can begin in the brain years before the first symptoms appear. Seeing these glimpses could lead to lifestyle changes that could delay the destruction of the brain.

Improving the diagnostic accuracy of Alzheimer’s disease is an important clinical goal. If we can increase the diagnostic accuracy of the models in ways that enable the acceleration of existing data such as MRI scans, that can be very beneficial. “

Vijaya B. Kolachalama, PhD, Corresponding Author, Associate Professor of Medicine, Boston University School of Medicine (BUSM)

Using an advanced AI (artificial intelligence) framework based on game theory (known as genetic host network or GAN), Kolachalama and his team processed brain images (some of low and high quality) to generate a disease-capable model. Classify Alzheimer’s with development. accuracy.

The quality of an MRI scan depends on the scanner instrument being used. For example, a 1.5 Tesla magnet scanner has a slightly lower quality image than an image taken from a Tesla 3 magnet scanner. Magnetic strength is a key parameter associated with a particular scanner. The researchers obtained brain MR images from both 1.5 Tesla and the 3 Tesla scanners of the same subjects taken simultaneously, and developed a GAN model that learned from those two images.

Because the model was “learning” from the 1.5 Tesla and 3 Tesla images, it generated images that were of better quality than the 1.5 Tesla scanner, and these created images also predicted quality. Alzheimer’s disease on these people would be more achievable using models based on 1.5 Tesla images alone. “Our model is basically capable of capturing 1.5 images with a Tesla scanner and generating better quality images and we can also use the resulting images to make better predictions. Alzheimer ‘s disease than what we could do using just 1.5 Tesla – based images alone, “he said.

Globally, the population aged 65 and over is growing faster than all other age groups. By 2050, one in six people in the world will be over 655. Although the total health care costs for AD treatment) in 2020 were estimated at $ 305 billion and was expected to rise to more than 65%. $ 1 trillion as the population ages. The great burden on patients and their carers, in particular, the family carers of AD patients suffer great hardship and distress that represents a great but often hidden burden.

According to the researchers it may be possible to generate high-quality images of disease groups that have previously used the 1.5T scanners, and in those centers that still rely on 1.5T scanners. “This would allow us to reconstruct the earliest stages of AD, and build a more accurate model of predicting the status of Alzheimer’s disease than would be possible using data from 1.5T scanners. only, “said Kolachalama.

He hopes that such advanced AI techniques can be put to good use so that the medical imaging community can make the most of the advances in AI. Such frameworks, he believes, can be used to harmonize image data across multiple studies so that models can be developed and compared across different numbers. This can lead to the development of better approaches for AD detection.