Machine Learning within the Clinical Trial Process

Applied Clinical Trials

Machine learning technology is finding it’s way into today’s marketplace, specifically within business processes, by driving insurmountable value for corporations within the life science industry, writes Adam Ross Miller.

Many are familiar with the reputation of the clinical trial process, and the ethical issues that surround them. Whether you blame the medical risk, political agendas, or monetary incentives we all hear about unfortunate experiences during these trials.  

The clinical trial process is the most dangerous part of medical discover, and there is no exact science to medical discovery which is represented very clearly in the past.1  The top reason why people submit themselves through this highly-scrutinized process, based on a poll executed for Thomson CenterWatch, are for motives of finding “better medical treatment.”2 The second reason is to help push forward medical research.3

The focus here is that nothing is certain nor predictable when going through a clinical trial. Physical technology is pushing us toward efficiencies in caring for patients, but there is a bigger need in process technology within the early medical discovery stage. Process technology, like machine learning, allows for efficiencies in a tightly regulated industry such as the life sciences.

As data capturing technologies continue to improve within the clinical trial process the opportunity to leverage more qualified data into a machine learning technology will diminish patient risk and enhance pharma’s time and quality. The idea that a technology can identify data points to coach professionals to act in a risk averse manner for patients while being more productive in time will result in faster and better quality medicine improvements to society.   

 

Diminishing Patient Risk

Driving data consistency across patient interaction will allow for a machine learning technology to coach and suggest higher quality methods. The medical risk that goes with enrolling into a clinical trial is extremely high. The risk will likely never conclude but has the potential to improve. Everything surrounds data and how we analyze that data to be more productive.

Today, pharma companies are spending more time and money analyzing data-finding them in a situation where they are too late to apply the correct methods on a patient within a clinical trial-than applying the safest and most productive methods toward the patient. As you can see, this time gap of data analysis impacts a professional’s ability to focus on patient safety along with driving innovation through time and quality productivity.

 

Enhancing Pharmaceutical Company’s Quality & Results

According to Forte Research Systems, benchmark data suggest that only 7 of 100 people interested in enrolling in a clinical trial completes the trial process from start to finish.4 You can imagine the opportunity and need for more efficient processes that increase enrollment numbers to produce viable medical discoveries in a shorter period. Forte also suggests that of the individuals who have dropped out, 24% felt like expectations of the trial fell short. I would argue that this is a cause of manual and slow data analysis; something that machine learning can overcome in a trial.

With the increase of pharma corporation’s medical results because of machine learning, we will see a positive shift in the clinical trial reputation, which in turn will increase the supply of clinical trial enrollment, allowing the industry to focus less on political regulation/monetary self-interest, and more time on innovating toward human’s medical needs.

This push of this technology will not eliminate the need for human interaction, in fact I believe it will increase the need. Just like the past, we go through innovation periods and find that human interaction is highly necessary but maybe in other aspects within the industry. Innovation breeds more innovation and human interaction is the foundation of new ideas.

In conclusion, with access to large data pools within the life sciences industry, there are opportunities being missed daily by not efficiently analyzing that data and translating it into coaching professionals on most efficient and safe results. Building a technology that will do this on it’s own will allow for more time to focus on the broader infrastructure of innovating medical discoveries.

 

 

Adam Ross Miller is Account Executive at Octiv. He can be reached by email at milleradamross@gmail.com or on twitter @_adammiller_

 

References

1O'Meara, Alex. Chasing Medical Miracles: the Promise and Perils of Clinical Trials. Walker & Co., 2010. P 21.
2 “The psychology of clinical trials:understanding physician motivation and patient perception,” Thompson CenterWatch, Survey of 749 study volunteers, 2006, http://www.centerwatch.com/professional/ cw_commentary_psychology.html.
3O'Meara, Alex. Chasing Medical Miracles: the Promise and Perils of Clinical Trials. Walker & Co., 2010.P 173.
4“Lopienski, Kristina. Retention in Clinical Trials: Keeping Patients on Protocols. Forte. forteresearch.com/news/infographic/infographic-retention-in-clinical-trials-keeping-patients-on-protocols/