While artificial intelligence continues to become more of a factor in business and consumer fields, one sector in particular that has yet to fully utilize this tool is the global healthcare system. In the past, the time-tested wet lab studies were used to develop drugs. This process involves intricate work with an array of protein and nucleic acid samples in order to identify signatures of current pathogens. This process is labor intensive, expensive, and quite long.
Now, however, scientists are beginning to use artificial intelligence to accelerate drug development. AI accomplishes this by identifying and predicting the development of superbugs—antibiotic-resistant bacteria—along with determining what protein signatures and drug compounds currently on market would be effective launching points for fighting the new pathogen [2]. This eliminates the need for time consuming preliminary research which served a vital role in catching up with the COVID-19 pandemic. One example of this is the implementation of a viral protein prediction software called AlphaFold. This software can determine what types of proteins are plausible candidates for a virus to possess based on its lineage and pathogenic consequences [5]. Such software assists in cutting down the preliminary testing needed to determine the genetic makeup of viruses. This is key to accelerating the production of a vaccine — the efficacy of which is dependent on mimicking an immune response effective against infection by the actual virus [3].
Along with this isolated software usage of AI, teams across the globe are working together to create a network of data, so the benefits of artificial intelligence can be felt worldwide. For instance, iLearn is a platform that analyzes a variety of data, such as DNA, RNA, and protein sequences, using a plethora of different analytical techniques under the umbrella of machine learning to more accurately identify a variety of new diseases that stem from both viruses and bacteria [1]. An example of this is the open-source Python toolkit, iFeature, which can conduct precise feature analysis on previous pathogens and help predict how novel disease will interact with individuals — such as predictions on how mutations may arise [1].
Furthermore, artificial intelligence shows signs of being able to accurately predict combatants for pathogens that have never been researched before and are largely unrelated to discovered pathogens of the past. By applying machine learning to ⍺-helical host defense peptides, the underlying blueprint of these antimicrobial peptides (AMPs) can be derived; this enables scientists to design more effective and permutable defenses to yet undiscovered bacterial pathogens [4]. This indicates that machine learning can find patterns within the natural defense mechanisms that past biological studies have already isolated. By leveraging the understanding of these patterns, scientists can design defense mechanisms suited to overcome the pathogens that affect humans, and novel bacterial strains can more quickly be rendered harmless.
Additionally, artificial intelligence is currently being applied to research regarding chronic diseases. Alerting AKI is a software that accurately warns of critical illness due to acute kidney injury (AKI). It does so through the use of techniques including a random forest, support vector machine, and a neural network classifier [7]. Thus, artificial intelligence is serving as a resourceful asset for combating not only transient infections brought about by passing disease vectors but also prolonged diseases that quietly impact the lives of millions.
A final expansion of the use of AI software in healthcare reaches slightly beyond human care and into the field of plant health and wellness. In environments with large farmland crops, artificial intelligence is being used to accurately survey acres of land to efficiently identify fungal diseases in apple crops. This includes scab, which is caused by Venturia inaequalis, as well as rust, which is caused by Gymnosporangium juniperi-virginianae [6]. Consequently, AI is not restricted to healthcare research in the traditional sense of assisting in disease treatment; it also contributes more broadly to the global food supply chain and ecosystem that impacts our lives every day.
Ultimately, these are only a few of the leading implementations of artificial intelligence software in healthcare research and ecosystem health research. To the acute innovator, there are many potential permutations of such systems to diversify the application of artificial intelligence across the entire breadth, depth, and complexity of healthcare.
References:
Chen, Z., Zhao, P., Li, F., Marquez-Lago, T. T., Leier, A., Revote, J., Zhu, Y., Powell, D. R., Akutsu, T., Webb, G. I., Chou, K.-C., Smith, A. I., Daly, R. J., Li, J., & Song, J. (2019). ILEARN: An integrated platform and meta-learner for feature engineering, machine-learning analysis and modeling of DNA, RNA and protein sequence data. Briefings in Bioinformatics, 21(3), 1047–1057. https://doi.org/10.1093/bib/bbz041
David, L., Brata, A. M., Mogosan, C., Pop, C., Czako, Z., Muresan, L., Ismaiel, A., Dumitrașcu, D. I., Leucuta, D. C., Stanculete, M. F., Iaru, I., & Popa, S. L. (2021). Artificial Intelligence and antibiotic discovery. Antibiotics, 10(11), 1376. https://doi.org/10.3390/antibiotics10111376
Kang, S.-M., & Compans, R. W. (2009). Host responses from innate to adaptive immunity after vaccination: Molecular and cellular events. Molecules and Cells, 27(1), 5–14. https://doi.org/10.1007/s10059-009-0015-1
Lee, E. Y., Fulan, B. M., Wong, G. C. L., & Ferguson, A. L. (2016). Mapping membrane activity in undiscovered peptide sequence space using machine learning. Proceedings of the National Academy of Sciences, 113(48), 13588–13593. https://doi.org/10.1073/pnas.1609893113
Park, Y., Casey, D., Joshi, I., Zhu, J., & Cheng, F. (2020). Emergence of new disease: How can artificial intelligence help? Trends in Molecular Medicine, 26(7), 627–629. https://doi.org/10.1016/j.molmed.2020.04.007
Roy, A. M., & Bhaduri, J. (2021). A deep learning enabled multi-class plant disease detection model based on computer vision. AI, 2(3), 413–428. https://doi.org/10.3390/ai2030026
Yuan, Q., Zhang, H., Deng, T., Tang, S., Yuan, X., Tang, W., Xie, Y., Ge, H., Wang, X., Zhou, Q., & Xiao, X. (2020). Role of Artificial Intelligence in Kidney Disease. International journal of medical sciences, 17(7), 970–984. https://doi.org/10.7150/ijms.42078
Comments