ARTIFICIAL INTELLIGENCE
To start with, would you mind giving me a quick overview and definition of what you mean by AI and decentralised trials? A lot of people think machine learning (ML) and artificial intelligence (AI) are the same but they are not. People talk about AI and ML but they don’t understand the basics; artificial intelligence is a broad spectrum and machine learning is a branch of that. Artificial intelligence does something like how a human thinks, it does a job. So for example, if I want to hold a pen and write something, an instruction has been preloaded in my brain – this is similar to how AI works. Machine learning is where a set of instructions have been given and it is being executed to make a solution or a prediction. AI has the ability to think about the situation or the environment that it is facing and then it makes a decision and gives a prediction to the people who are controlling it, so it is the automating of the intelligence. Decentralised clinical trials (DCT) have been around for a while now but historically weren’t very successful as we didn’t have the technologies required. However, things have rapidly improved and the recent pandemic has only accelerated the growth of decentralised trials. A decentralised trial is patient-centric, moving the trial to the patient’s location. This means it can reduce the cost of the drug development and it helps patients to save time and money as well, because there is no longer the cost to the patient of having to take a day off work or half a day off to go to the site of the trial. This type of trial also allows us to enrol patients globally, as long as they have access to the required systems and know how to use the electronic devices, so it is a fantastic tool.
What are the areas in decentralised trials that AI can help with? I would say that it’s from end to end, from the site selection or the protocol development to analysing results. I think a lot of companies are implementing AI into their clinical protocol development in areas such as site selection and identifying the countries where a study can work. Another area is patient selection: through using artificial intelligence and electronic health records, we can identify the patients relevant for the trial and we can select the particular country where the group of patients are based. After patient selection, we can use this AI in data
86 | Outsourcing in Clinical Trials Handbook
“If companies don’t start looking into these technologies, they might start coming more slowly to the development of drugs, leaving them behind in the race.”
mining so that the data collected from patients can then be analysed and used in data analytics. So AI is used throughout; even once the study is finished it can be used to prepare the final report as well, so it is a powerful potential tool. I think AI is a great opportunity, and implementing it is a must on the part of the pharmaceutical industry.
What are the hurdles or barriers to implementing AI? The main one I would say is the cost. For a pharma company to implement an AI project, it is a huge investment, you need resources in terms of manpower – data scientists, data analysts – and then there are the systems themselves. Unfortunately, not all AI programmes are successful because they are not given the proper resources. Another barrier is that sometimes the outcome is not as expected: AI predictions aren’t always as accurate as a human physician could be. I would still say that AI is a useful tool; it is just not human. AI is still not very successful when it comes to certain logical activities, but it is evolving and we are trying to solve the issues.
Are there any issues around safely securing and saving the data gathered?
In terms of safe and secure data collection, this is a very big question and issue because not all patients want to give away their personal data to a company. However, we do follow GDPR guidelines and similar guidelines from other countries. These issues are easier to overcome in places like the US and the UK and other developed nations because they have electronic health records and they have an educated patient population – which helps with ensuring data protection and security. In countries where we don’t have this kind of facility, it is more of a challenge, but there is also the opportunity to develop the guidelines. You do also find that, in many countries that don’t have secure data protection policies, patients will still be interested to enrol in your
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100
