Imagine a patient with breast cancer, who's been prescribed a standard chemotherapy treatment. The patient completes the treatment only to find out the cancer is still spreading. Upon further genetic testing, it is discovered that a patient has some rare mutation that made them resistant to the standard treatment. Precision medicine, tailored to her specific genetic profile, is then introduced. A targeted therapy is developed that successfully attacks cancer cells without harming healthy tissues. Wouldn't we all want this if we were battling cancer? That's the power of personalized medicine or precision medicine as some call it.

For far too long, the medical practitioners have been prescribing treatments and medications which are best suited for an average person. It doesn't usually address the individual nuances which could be at play helping enhance or hinder the treatment. Personalized medicine (PM) aims to solve for this by ensuring the right treatment is administered for the right individual at the right time. While there are a lot of advancements in the reactive care for patients, PM also opens up a great deal of opportunities in preventive care for patients. But even with all its promises, PM has not yet become mainstream and there are a lot of barriers to entry.

1. The main basis of PM is genomic testing, which is data intensive and historically has been very difficult to conduct at scale
2. The need for re-training of medical practitioners to be more patient centric than disease centric
3. Updating healthcare systems to accommodate for additional drugs + diseases + treatment combinations is a major regulatory and operational overhaul

However, with technology advancements specifically in the field of AI, there is a clear line of sight in making it mainstream with larger medical institutions already adopting these practices in pockets as well as newer health-tech startups coming up with innovative and disruptive solutions to make PM more mainstream. Here are some of the ways AI is helping accelerate the adoption of PM:

1. Analyzing vast datasets: Processing and analyzing massive amounts of patient data, including genetic information, medical history, and clinical trial results, to identify patterns and trends that may not be apparent to humans and sequence them into consumable datasets which can be further analyzed by medical professionals. Without AI and latest computing devices, this is an almost impossible task to do at scale and is a huge assist to further the PM cause.

2. Predicting disease risk: Use machine learning algorithms to predict a patient's risk of developing certain diseases based on their individual characteristics, lifestyle factors, and genetic makeup is becoming a major game changer for preventive care. Doctors can create personalized plans for high risk individuals and recommend lifestyle, diet changes, etc. Integrating the data sources from edge patient monitoring devices such as heart rate monitors, diabetes monitors, CPAP machines, etc can also allow doctors to get valuable data centrally and make real time changes to any preventive recommendations.

3. Developing targeted treatment: By analyzing patient data, AI is able to help researchers develop new drugs and therapies that are more effective and have fewer side effects for specific patient populations. Particularly in the area of drug discovery, AI-designed drugs are becoming more mainstream, helping to significantly improve the probability of success of clinical trials, which have historically been very low due to the high cost and time required to get good representative trial data.

4. Optimizing treatment plans: AI can assist healthcare providers in selecting the most appropriate treatment options for individual patients based on their unique characteristics and the latest medical research.

5. AI-assisted recovery: Using personalized assistants, most patients can have a recovery journey which is more closely supervised and adequate personnel interventions are made at the appropriate time through triggered alerts, etc.

6. Monitoring patient progress: AI-powered systems can track patient progress over time, identify potential complications early on and adjust treatment plans as needed.

A big factor in using AI for PM will also be the governance and privacy controls which are put in place for the scaled adoption. Current regulations may need to be revisited along with making data sharable and accessible more globally. The cost of genomic testing and other compute intensive resources will also need to be subsidized initially so that this doesn't become only the medicine of the rich, but is able to go beyond the economic barriers as well. Appropriate data policies along with maintaining an ethical implementation of the ML models will also be critical for the mass success.

PM isn't science fiction … it's reality

What is perhaps most exciting about PM is that it isn't too good to be true and it is driving real innovation today. For instance, biotechnology startups are already harnessing AI to revolutionize protein analysis and cut reduction in prediction times.

In addition, teams are using AI to revolutionize the way we detect cancerous cells. For example, currently, the detection process for cancerous cells is incredibly time intensive and requires qualified medical teams to scan each sample under a microscope, creating delays in testing and patient treatment as a result. However, through AI, it is becoming possible to automate the determination of cell normality – using similarity matching to quickly compare normal cells and abnormalities to a database of known lesions – dramatically reducing the time it takes to detect potentially cancerous cells.

These are just a couple of examples of how powerful PM can be. And if the right resources are invested into the space and the proper guardrails are established, the potential for PM to redefine patient care is nothing short of jaw dropping.

Originally published on Medcity news.