By Kavya Pillai (Class of 2025)
As aging is a commonality in all our lives, preserving cognitive health is integral to healthy aging. However, detecting cognitive decline is difficult as there is a hard distinction between the decline in mental functionality and the natural cognitive deterioration that is assumed with aging. Around 55 million people worldwide have dementia and with the increasing rate of 10 million cases per year, the cure to this disease is hard to find (4). For this reason, detecting it early on is essential in preventing further damage.
With the recent adoption of artificial intelligence (AI) in healthcare, these logical systems are aimed towards finding an early diagnosis of cognitive decline. The shift from aging to Alzheimer’s disease is known as amnestic mild cognitive impairment (aMCI). While aMCI isn’t as severe as Alzheimer’s disease, it still creates issues on their cognition that affects their daily lives.
A study by Ni Shu PHD (2021) analyzed the relation between cognitive impairment and age using MR images to create a prediction model about the age of a brain (2). The model compared individuals who had aMCI and those who had a normal aging brain (no signs of cognitive deterioration) and found a difference in both models. The predicted age model for aMCI had estimated the brain’s age to be about 3 years older than the individual’s physical age. The model was also able to show the difference amongst a patient who had a more developed aMCI to a patient that had a less progressive aMCI with its increased number of deviations. When people age, natural mental and cognitive deterioration occurs. However, it is important to determine how large this predicted age difference is, as the higher it is the greater the probability towards diseases pertaining to memory loss such as dementia and Alzheimer’s disease (3).
With AI in healthcare, different subsets of techniques are used depending on the functions needed (algorithm derived from labeled data, pattern in unlabeled data, replication to human behavior). A conceptual review by Sarah Graham (2020) analyzed possible predictors of change and decline in cognitive function using methods such as machine learning, supervised learning, unsupervised learning, deep learning, and natural language processing. As the review says, AI would aid in predicting a possible impairment in cognition using these methods but wouldn’t be able to treat it as a clinician would. The study conducted by Ni Shu PHD uses machine learning to predict brain age in individuals with aMCI. After testing the model on patients with aMCI versus patients with healthy controls, the prediction model showed individuals would have aMCI had a higher predicted age difference than those who had healthy controls.
The review by Sarah Graham also considers the sociodemographic data in the progression of cognitive decline (1). It's important to consider social determinants of health, such as access to education, as it gives a general overview of different countries and their dementia rates. Using unsupervised learning, the study made connections with a more inclusive demographic group, which in turn would make findings applicable to a wider population. Dementia is more prevalent in low income/poor resource areas which is represented as a model with a higher PAD. The review also considers clinical and psychometric assessments, neuroimaging data, electronic health record, basic functions, and genomic data which would combat early risk stratification.
As AI technology is developing into playing a role in health care, the prevention of cognitive decline derives strongly from predicting these neurocognitive disorders. With an early diagnosis, experienced clinicians can take action to treat cognitive impairment. While AI isn’t the cure for disorders relating to mental decline, it would greatly help in determining the state of the disease, specifying the prognosis, and coming up with a more thorough treatment plan. With technology constantly advancing, AI is our future into predicting and preventing cognitive decline.
References:
Cover Picture: Keown, A. (2020, October 28). Study suggests COVID-19 infection could age the brain, reduce cognition. BioSpace. Retrieved from https://www.biospace.com/article/covid-19-infection-could-age-the-brain-up-to-10-years-reduce-cognition-study-suggests/.
With the recent adoption of artificial intelligence (AI) in healthcare, these logical systems are aimed towards finding an early diagnosis of cognitive decline. The shift from aging to Alzheimer’s disease is known as amnestic mild cognitive impairment (aMCI). While aMCI isn’t as severe as Alzheimer’s disease, it still creates issues on their cognition that affects their daily lives.
A study by Ni Shu PHD (2021) analyzed the relation between cognitive impairment and age using MR images to create a prediction model about the age of a brain (2). The model compared individuals who had aMCI and those who had a normal aging brain (no signs of cognitive deterioration) and found a difference in both models. The predicted age model for aMCI had estimated the brain’s age to be about 3 years older than the individual’s physical age. The model was also able to show the difference amongst a patient who had a more developed aMCI to a patient that had a less progressive aMCI with its increased number of deviations. When people age, natural mental and cognitive deterioration occurs. However, it is important to determine how large this predicted age difference is, as the higher it is the greater the probability towards diseases pertaining to memory loss such as dementia and Alzheimer’s disease (3).
With AI in healthcare, different subsets of techniques are used depending on the functions needed (algorithm derived from labeled data, pattern in unlabeled data, replication to human behavior). A conceptual review by Sarah Graham (2020) analyzed possible predictors of change and decline in cognitive function using methods such as machine learning, supervised learning, unsupervised learning, deep learning, and natural language processing. As the review says, AI would aid in predicting a possible impairment in cognition using these methods but wouldn’t be able to treat it as a clinician would. The study conducted by Ni Shu PHD uses machine learning to predict brain age in individuals with aMCI. After testing the model on patients with aMCI versus patients with healthy controls, the prediction model showed individuals would have aMCI had a higher predicted age difference than those who had healthy controls.
The review by Sarah Graham also considers the sociodemographic data in the progression of cognitive decline (1). It's important to consider social determinants of health, such as access to education, as it gives a general overview of different countries and their dementia rates. Using unsupervised learning, the study made connections with a more inclusive demographic group, which in turn would make findings applicable to a wider population. Dementia is more prevalent in low income/poor resource areas which is represented as a model with a higher PAD. The review also considers clinical and psychometric assessments, neuroimaging data, electronic health record, basic functions, and genomic data which would combat early risk stratification.
As AI technology is developing into playing a role in health care, the prevention of cognitive decline derives strongly from predicting these neurocognitive disorders. With an early diagnosis, experienced clinicians can take action to treat cognitive impairment. While AI isn’t the cure for disorders relating to mental decline, it would greatly help in determining the state of the disease, specifying the prognosis, and coming up with a more thorough treatment plan. With technology constantly advancing, AI is our future into predicting and preventing cognitive decline.
References:
Cover Picture: Keown, A. (2020, October 28). Study suggests COVID-19 infection could age the brain, reduce cognition. BioSpace. Retrieved from https://www.biospace.com/article/covid-19-infection-could-age-the-brain-up-to-10-years-reduce-cognition-study-suggests/.
- Graham, S. A., Lee, E. E., Jeste, D. V., Van Patten, R., Twamley, E. W., Nebeker, C., Yamada, Y., Kim, H.-C., & Depp, C. A. (2020, February). Artificial intelligence approaches to predicting and detecting cognitive decline in older adults: A conceptual review. Psychiatry research. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7081667/.
- Huang, W., Author AffiliationsFrom the State Key Laboratory of Cognitive Neuroscience and Learning, RC, P., BC, D., F, S., Et Al, L, deT.-M., JH, C., K, F., I, B., HG, S., C, G., F, L., LC, L., J, W., M, L., K, N., X, L., Jr, J. C. R., … YY, L. (2021, June 23). @Radiology_AI. Radiology: Artificial Intelligence. Retrieved from https://pubs.rsna.org/doi/10.1148/ryai.2021200171.
- Morley, J. E., Morris, J. C., Berg-Weger, M., Borson, S., Carpenter, B. D., Del Campo, N., Dubois, B., Fargo, K., Fitten, L. J., Flaherty, J. H., Ganguli, M., Grossberg, G. T., Malmstrom, T. K., Petersen, R. D., Rodriguez, C., Saykin, A. J., Scheltens, P., Tangalos, E. G., Verghese, J., … Vellas, B. (2015, September 1). Brain health: The importance of recognizing cognitive impairment: An IAGG consensus conference. Journal of the American Medical Directors Association. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4822500/.
- World Health Organization. (n.d.). Dementia. World Health Organization. Retrieved from https://www.who.int/health-topics/dementia.
Proudly powered by Weebly