Jason Smith, CTO, AI & Analytics, Within3, describes the possibilities made by AI in supporting the life sciences sector, and how it can allow pharma experts to produce real time results and yield better insights.
Key insights:
- AI innovation in life science can support and enhance what trial teams and other life science teams can achieve.
- It can enable human decision-makers to obtain more tailored, specific and condensed conversation summaries to inform next steps.
- AI-powered natural language processing can let teams source information as quickly as possible to inform strategic decisions.
- Many pharmaceutical and medical device companies still remain reluctant to leverage the innovations of AI, with 37% of people saying they fear technology they don’t understand.
Practical, real-world applications of artificial intelligence (AI) are widely known and used by many industries. For many, the term still carries perceptions influenced by movies and science fiction. If people are new to AI, these stereotypes can be intimidating. But to people more invested in the AI space, the term connotes possibility above all else.
AI applications have created a lot of interest in the life science industry in recent years. The extent of just how far this technology can assist us remains unknown, but what we can see are the places where AI can create even more possibilities for life science teams.
AI in life science can shine a spotlight on the most important information within large sets of technical data. It can do so with accuracy, precision and efficiency, creating value for teams pressed for time. For example, clinical trial teams can more quickly be made aware of potential side effects of a particular drug – information that could reduce protocol amendments or other delays. In this way, AI applications can give trial teams a safety net and provide the means to make more accurate decisions.
This example serves as a good illustration of the most important takeaway for the industry: the purpose of AI is not to replace humans. In actual fact, it cannot. AI innovation in life science supports and enhances what trial teams and other life science teams can achieve.
Time controls all
Life science teams face a number of significant obstacles, and restricted timelines can add additional pressure. Elements of a project that require more observation and management, such as patient recruitment and feedback, may mean that crucial data points are missed or not identified. Unfortunately, missed insights can result in inaccurate trials, extended project timelines and a host of unexpected delays.
So, where does AI come in? AI can create the foundation for more accurate outcomes, all whilst saving time and pushing progress forward. AI tools are streamlining the industry across three particular areas: natural language processing (NLP) to speed up and inform strategic decisions, real-time insights to guide better conversations and yield better insights, and sentiment analysis to understand trends.
Natural language processing trained for life sciences
In interactions and conversations between medical science and healthcare professionals, the goal is always to enable good decisions. However, when discussing specific disease areas or patient experiences, these conversations can take place over long periods of time, with inconsistent data-capturing tools. AI-powered natural language processing can let teams source information as quickly as possible to inform strategic decisions.
NLP can support some of the manual processing tasks that come with collating and interpreting information from trial teams, or when conducting research. It can help answer key questions, such as: What are the key scientific concepts coming out of an interaction? What new information could determine the direction of an upcoming trial? By removing hours of manual analysis, NLP can more efficiently guide teams in the right direction.
AI-powered insights for real time results
In the thick of product development, life science organisations – typically medical affairs teams – conduct many important conversations with experts, patients, and peers. Teams need to ensure these discussions are robust, effective, and revealing of directional information that supports strategy.
AI can help to facilitate these critical conversations. If many fragmented discussions are taking place across multiple time zones, it’s difficult to absorb the details of each and make sense of them as a whole. Human moderators must analyse large amounts of complex information, where inevitably, important scientific insights may be lost. With AI, human decision-makers can obtain more tailored, specific and condensed conversation summaries to inform next steps.
These AI-surfaced insights could lead to life-changing treatments, and this capability highlights AI’s potential to ensure the best content is sourced from a conversation. This does not take away from the capabilities of medical affairs teams – it simply helps to ensure clinical insights are not lost.
Sentiment and trends
Qualitative notes are typically used when medical affairs teams come together to discuss disease states, patient care pathways, or the effectiveness of a trial or treatment. For example: Doctor really loves the latest results from our clinical trial or Doctor has concerns about accessibility.
This information holds value, but the content of the insight only represents a small selection of voices, rather than a larger subset of diverse perspectives. This data does not account for more nuanced observations which are likely to present themselves in real life. AI is changing the way teams gather and dissect notes by uncovering the prevailing sentiment of a conversation.
When notes are captured from multiple medical affairs teams or several one-on-one conversations with key opinion leaders, AI can identify positive or negative sentiment across a larger volume of information. Trends can more easily be extracted from bigger sets of data, which presents teams with important trend lines to inform strategy, influence leadership, or make decisions.
Are there any restrictions?
To go back to the key takeaway, AI supports life science teams to allow them to access and produce better insights. Despite this, many pharmaceutical and medical device companies remain reluctant to leverage these innovations, even as AI is used more frequently in banking, technology, and retail in recent years.
One reason for this is that people who haven’t encountered AI technology may believe it is ineffective, or have a fear of change. Another major issue is simply fear of the unknown, with 37% of people saying they fear technology they don’t understand. If new technology is being used in an industry in which the fundamental purpose is to help people, trust issues are likely to be prominent.
However, the purpose of AI is to ensure important, potentially life-changing insights are not missed. This saves time for life science and healthcare professionals, and adds a degree of confidence to key decisions. For drug and device developers, the reality is that once organisations begin to uncover the benefits of AI, the efficiency of the sector as a whole can only increase.