Here, we discuss some examples of how artificial intelligence is contributing to the treatment of age-related diseases and an explanation of the underlying technologies.
Image Credit: Africa Studio/Shutterstock.com
Artificial Intelligence (AI) is playing an increasingly vital role in treating and managing age-related diseases. Recent developments in AI-driven healthcare technologies offer innovative approaches to diagnosis, personalized treatment, drug discovery, and disease prevention.
Early Diagnosis and Disease Detection
Age-related diseases tend to become more severe with time; thus, it is imperative to seek ways to detect these diseases at an early stage to prevent further deterioration. AI has been proven to assist in early diagnosis and disease detection, especially for Alzheimer’s disease, whereby AI adeptly scrutinizes cerebral MRI scans to discern subtle, nascent indicators of the disease through the employment of a deep learning neural network known as Convolutional Neural Networks (CNNs).
CNNs have been able to analyze and uncover information from various forms of data, particularly time series and images. They possess multiple layers, whereby each layer learns to detect various features in an input image.
A filter is applied to the image, making the output more detailed with each layer. Simple features are detected in the early layers, though as you move through the layers, the filters become more complex, recognizing unique characteristics of the input.
The output from each layer becomes the input for the next, and in the final fully connected (FC) layer, the CNN identifies the object. This process repeats for numerous layers, allowing the CNN to recognize the complete object, or image, by piecing together different features.
Hence, regarding Alzheimer's diagnosis, the sophisticated algorithms that underlie CNNs sift through MRI imagery. They then extract intricate features, detect patterns that might elude the human eye, and discern minute structural changes in the brain and indicators for Alzheimer’s disease, such as the presence of beta-amyloid plaques.
AI-Driven Synthetic Imaging
Another method by which AI is leading a charge against age-related diseases is the ability to generate synthetic images of rare diseases, which would provide a large enough dataset to build robust deep learning systems in early identification. The model that underlies this technological leap is known as Generative Adversarial Networks (GANs).
GANs belong to a category of machine learning techniques that involve two neural networks pitted against each other to enhance the accuracy of their predictions, whereby this AI methodology introduces a distinctly human trait, creativity, into the realm of computer technology.
Thus, GANs have been utilised to synthetically generate images of rare diseases, such as age-related macular degeneration (AMD) - a major cause of vision problems worldwide. A research group, Wang et al (2023), demonstrated that GAN models - trained with non-AMD phenotypical dataset - could synthetically generate realistic fundus images of the eye with AMD features.
It is therefore promising that such images generated by GAN models would be implemented to address the lack of dataset of rare diseases, allowing more accurate identification of rare age-related diseases where early detection is vital, such as in AMD.
Image Credit: Gorodenkoff/Shutterstock.com
Drug Discovery and Development
The drug discovery and development process is a lengthy and expensive endeavor. The use of AI within this sphere will have a profound impact, changing the way we conduct the drug discovery and development phase by reducing the time and cost.
With regard to the fight against age-related diseases, AI has been utilized in the discovery of novel senolytics, a class of drugs that target and eliminate senescent cells within the body. Smer-Barreto et al. (2023), were able to implement AI models with complex machine learning algorithms, XGBoost, Random Forest (RF), and Support Vector Machine (SVM) , to predict the molecules that had potential for senolytic properties.
After having trained the AI model with non-senolytic and senolytic molecules, it was then able to analyze molecules that it has never encountered for senolytic potential, whereby after five minutes of analyzing 4830 molecules, the AI model predicted 21 molecules to be senolytics.
Further testing of these 21 molecules identified 3 to (periplocin, oleandrin and ginkgetin) to effectively kill senescent cells while leaving healthy cells unharmed. Thus, implementing AI within the drug discovery and development process will undoubtedly be conducive to reducing the expense and duration of the endeavor, which is vital for age-related diseases as they tend to progress and worsen over time.
Conclusion
In these use cases, AI technologies are at the forefront of healthcare transformation, bringing in a new future of early, precise diagnostics and advanced drug discovery. With the ever-advancing realm of AI, we stand on the precipice of a future where age-related diseases are more easily diagnosed, managed and treated.
References and Further Reading
Heising L, Angelopoulos S. Operationalising fairness in medical AI adoption: detection of early Alzheimer’s disease with 2D CNN. BMJ Health Care Inform. 2022 Apr;29(1):e100485.
Patil V, Madgi M, Kiran A. Early prediction of Alzheimer’s disease using convolutional neural network: a review. The Egyptian Journal of Neurology, Psychiatry and Neurosurgery. 2022 Nov 17;58(1).
Awati R. What is convolutional neural network? - Definition from WhatIs.com [Internet]. SearchEnterpriseAI. 2022. Available from: https://www.techtarget.com/searchenterpriseai/definition/convolutional-neural-network
Vatansever S, Schlessinger A, Wacker D, Kaniskan HÜ, Jin J, Zhou M, et al. Artificial intelligence and machine learning‐aided drug discovery in central nervous system diseases: State‐of‐the‐arts and future directions. Medicinal Research Reviews. 2020 Dec 9;41(3):1427–73.
Wang Z, Lim G, Wei Yan Ng, Tan T, Lim J, Sing Hui Lim, et al. Synthetic artificial intelligence using generative adversarial network for retinal imaging in detection of age-related macular degeneration. Frontiers in Medicine. 2023 Jun 22;10.
Smer-Barreto V, Quintanilla A, Elliott RJR, Dawson JC, Sun J, Campa VM, et al. Discovery of senolytics using machine learning. Nature Communications [Internet]. 2023 Jun 10 [cited 2023 Oct 21];14:3445. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10257182/
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.