
Machine Learning – 3 ways it’s a game-changer for Pharma

With the tech world enjoying its most lucrative AI and Machine Learning spring yet, it is no surprise that AI’s impact has spread to one of the largest industries on the planet – pharmaceuticals. Indeed, AI’s potential influence is evident throughout the drug development process: from the initial discovery phase, through the pre-clinical and clinical research stages, to the monitoring of drug performance after general release. Here, we will explore some of the innovative ways that this new technology is revolutionising the pharma world.
Regarding drug discovery, AI’s chief attribute is its ability to accelerate the process by analysing a vast amount of data.
Examples of AI drug discovery start-ups
Meta
For example, Meta – a pharma start-up from Toronto – designed a machine learning model that organises biomedical research to provide researchers with personalised feeds informing on important, novel research relevant to their current drug targets. This maximises the efficiency of the initial research aspect of drug development by providing concise and actionable information.
Mozi
Similarly, Mozi – a Chinese start-up based in Shenzhen – uses Machine Learning to identify patterns in biomedical data and suggest feasible hypotheses for further investigation. This permits researchers to upload relevant datasets and gain valuable insights and suggestions pertaining to diagnostic and treatment strategies. Again, this will save a significant amount of time while ensuring an accurate and effective development process.
Helix AI
Another notable start up using Machine Learning to aid and hasten drug discovery is HelixAI. Similar to the smart speakers that have found their way into our living rooms and kitchens, their software uses AI to respond to verbal questions and requests in a laboratory setting. This keeps researchers continually informed on relevant studies, allows them to manage their inventory with great efficiency and maximises productivity within the lab. Evidently, therefore, AI can be used in a host of imaginative ways to ensure the research and development stage of drug discovery is as robust and speedy as possible.
Machine Learning aided clinical trials
Next, AI is becoming increasingly evident in clinical trials. First, it is aiding with participant recruitment – a notorious challenge as identifying appropriate patients who are willing to partake is a time-consuming exercise. An AI solution would address this issue by extracting relevant information from a patient’s medical records, comparing the outcome with data from ongoing trials, and suggesting clinical studies that would best fit individual patients. Next, for these trials to be effective, participating patients must adhere to trial rules. If they take medication at the wrong times, forget to take it, or take the wrong medication, this can negatively impact their health and jeopardise the accuracy of outcomes.
AICure, a New-York based tech company, uses computer vision technology to track adherence. Patients film themselves taking a pill, and AiCure’s software confirms that the right person is taking the correct medication at the appropriate time. This can be taken further by introducing an AI assistant. It would be voice controlled to maximise the likelihood of daily interaction and allow patients to set reminders, ask for advice if a dose is inadvertently skipped and help them understand how the treatment is impacting their health. Thus, AI technology can guide entire clinical trials to help guarantee accuracy, safety and efficiency.
Optimise drug market return
Finally, Machine Learning can play a crucial role in ensuring that, once on the market, drugs are being used in the most effective way possible. For example, it can facilitate dynamic dosing whereby drug dosage is optimised at the individual level. An AI system would analyse data that outlines a patient’s response to a drug (e.g. vital signs and symptom severity) and monitor changes to offer suggestions as to how much of a given drug a patient should take on any one day. This personalised approach would maximise the efficiency of treatment while maintaining safety and minimising waste.
Evidently, as is the case with the vast majority of industries, AI seems to be playing a leading role in driving the growth of the pharmaceutical sector. As these technologies improve and innovative ideas continue to be proposed, there is a good chance that AI will render the pharma world unrecognisable in the years to come.
References
https://www.investopedia.com/articles/investing/072913/8-stages-new-drug-development.asp
https://blog.benchsci.com/startups-using-artificial-intelligence-in-drug-discovery
https://medicalfuturist.com/top-companies-using-a-i-in-drug-discovery-and-development/
https://pharma.bayer.com/ai-drug-discovery-accelerating-development-new-treatments
https://www.cbinsights.com/research/clinical-trials-ai-tech-disruption/