Guest Article for ARCS Australia.
Due to its many tangible benefits, Artificial Intelligence (AI) is already prevalent throughout the healthcare ecosystem.
The same machine learning algorithms that recognise faces in a crowd and enable self-driving cars can estimate human cognition and analyse complicated medical data. These algorithms can also “learn” patterns in data and create their own “logic”. This means that larger amounts of information to be extracted and analysed with greater efficiency.
AI in the health industry
The most common applications of AI in the healthcare industry are those with repetitive functions, like managing medical data or medication, or analysing tests, x-rays, and CT scans. Radiology and cardiology are two medical disciplines where data analysis is often vast and time-consuming. We are fast moving towards a time where human involvement in these processes is only required for the most complex analyses.
Considering these capabilities, using AI in clinical trials is an excellent way to give the $65B clinical trials market a much-needed makeover.
Artificial intelligence technology has the potential to change every stage of the clinical trials process, from finding a trial, to enrolment to medication adherence.
AI has the potential to change every stage of the clinical trials process, from finding a trial, to enrolment to medication adherence
— Tweet this quote
AI and drug development
The development of pharmaceuticals through clinical trials can be expensive and time-consuming – changing this could change the world. An example of this is the AI-powered software that scanned existing medicines during the height of the Ebola epidemic, between 2013-2016.
The AI technology found two medications that could be redesigned to reduce Ebola infectivity within a day. Analysis of this type generally takes months or even years. Not only could a time saving like this save thousands of lives, the overall approach may be applicable for future pandemics – potentially saving thousands more lives.
AI and medication compliance
The World Health Organisation (WHO) estimate that only 50% of patients take their chronic disease medication, as prescribed. When it comes to clinical studies, non-adherence can not only adversely affect a patient’s health, it can also corrupt data and require that new participants be enrolled.
AI technology which seeks to address this exists in the form of wearable sensors and ingestible sensors, as well as smart pill bottles with patented technology that uses a sensor in the bottle to detect when it has been opened and calculate the number of pills or amount of liquid remaining. The information is wirelessly transmitted to the cloud, from which a patient can be notified by phone, text or email. Patients can also be reminded by a blinking light and sound from the bottle itself.
AI and participant enrolment
From almost every angle, clinical trials are cost and time-intensive for participants – starting at the enrolment phase. Many trials fail due to enrolment issues. Generally, a patient must:
- Complete a preliminary phone screen with a study investigator
- Visit a participating site to see if they are eligible
- Go through evaluations, like laboratory and imaging tests, to make sure they meet all the inclusion and exclusion criteria
Depending on their availability and how far they live from a trial site, some people may be able to complete these procedures in less than a week. But for others juggling jobs, families, or long commutes, the process could take multiple visits.
As a part of confirming eligibility, site investigators collect patients’ medical records from other physicians’ offices. These faxed or emailed copies add an additional layer of complexity in using AI to extract information.
AI Technology could have huge potential for streamlining this clunky process.
Once patient records are digitised and seamlessly integrated with AI technology, it will be possible to improve the clinical trial referral enrolment pathway. An ideal AI solution would be one that extracts relevant information from a patients’ medical records, compares it with ongoing trials, verifies inclusion and exclusion criteria, and suggests appropriate studies – either directly to patients, or via their healthcare practitioner.
Developments like these will ultimately reduce costs, expedite enrolment timelines and help ensure that consumers get access to new therapies in a shorter period.