By Assoc. Prof. Ir. Ts. Dr. Asnida Abdul Wahab
There is a transition from a ‘one-size-fits-all’ approach in medicine to a more personalized one, in which your doctor can tailor treatments and preventative measures to your unique biology and lifestyle. This exciting field utilizes the potential of artificial intelligence (AI) to revolutionize how we approach health. It can manage your health proactively by anticipating your individual needs, not a generic plan. Therefore, such future medicine could benefit people’s well-being, hence contributing to the country’s economy.
Electronic health records, genetic information, lifestyle choices, and, in some cases, data from wearable devices can all contribute to a person’s health data bank. AI, which mimics human cognitive processes, analyzes these data to determine patterns in one’s health. Leading to the next level of healthcare, these new insights can result in predicting disease outbreaks, say, for example, how the COVID-19 pandemic started and will reach its zenith, as covered in our recent publication (DOI: 10.47836/mjmhs.18.s6.14), planning treatments according to an individual’s genetic makeup and tracking their health pattern.
To clarify the previous point, using an example in the personalized medicine scenario, AI algorithms might discover that a particular genetic variant interacts poorly with a popular diabetes treatment. By analyzing large amounts of patient data, an algorithm can be developed that looks into how different individuals respond to various treatments and are impacted by disease risks. This information could be used to identify individuals who are at high risk of adverse effects and modify their treatment plans to account for that.
Why Type 2 Diabetes is Ideal for AI-driven Personalized Medicine?
Diabetes is a chronic condition affecting millions globally. There are two main types: type 1 diabetes (T1D) is an autoimmune disease, whereas type 2 diabetes (T2D) occurs due to the rejection of insulin by the body or its inability to produce enough amounts of insulin. Despite the numerous and complex treatment options available, proper treatment of T2D also involves constant monitoring and adjustment. Thus, personalized medicine can readily be applied to T2D.
Regrettably, in many countries, particularly Malaysia, managing these patients is currently driven by an algorithmic approach. Most people with T2D receive the same empirically based treatment plans. These protocols do not take into account not only the individual causes of disease development but also the characteristics of each patient. Moreover, medications are selected primarily based on the disease progression, while ignoring possible reactions and side effects.
T2D’s variability suggests the disease could have several subtypes, with research now focusing on tailoring treatments to these subtypes based on clinical data, as discussed in our recent review article with DOI: 10.1007/s10462-022-10202-8, thus enhancing treatment efficacy and personalization.
Where AI Makes a Difference?
Unlike T1D, T2D is strongly influenced by a lot of factors, including genetics, lifestyle choices, and individual responses to medications. This is where AI-driven personalized treatment comes in. AI excels at analyzing the complex interplay of variables that contribute to T2D.
Here’s how AI can revolutionize T2D treatment! AI enables one to exploit large data of patients with T2D, which consists of information like an individual’s genetic makeup, blood sugar trends, dietary habits, activity levels, and even medication responses. By considering all these factors, AI algorithms can create a comprehensive picture of an individual’s T2D.
This empowers doctors to develop personalized treatment plans that are more effective in controlling blood sugar, minimizing complications, and ultimately, improving the patient’s quality of life. AI has the potential to move us away from a one-size-fits-all approach and towards a future of truly individualized T2D management in Malaysia and beyond.
While the potential of personalized medicine powered by AI is vast, there are challenges to overcome. The interpretability of AI algorithms for healthcare practitioners remains a paramount concern. While AI can identify patterns in data, explaining the “why” behind those patterns is crucial for doctors to trust and utilize the recommendations effectively.
Beyond Diabetes: A Broader Impact
Despite these hurdles, there is no question that the future of healthcare is personalized. Through the use of AI, we are finally approaching an era in which treatments can be made feasible, thus ensuring a healthier and more informed society. The possibilities of AI-driven personalized medicine do not end with diabetes. Variations of this approach can be utilized to predict the likelihood of a person developing specific types of cancer by considering genetic and lifestyle-based predispositions. Hence, they would be provided with early detection and time-saving intervention.
Additionally, AI can assist in optimizing drug therapies by predicting how a patient will respond to a specific medication, enabling doctors to personalize treatment plans and minimize the risk of side effects. Furthermore, by identifying individuals with a predisposition to certain diseases, we can develop personalized preventative measures to keep them healthy. This showcases the vast potential of AI in transforming healthcare.
Co-Author: Nor Nisha Nadhira Nazirun
Edited By: Zuhaili Idham