Alan Kalton, SVP of Global Sales, Aktana05.26.22
Artificial Intelligence (AI) and big data/analytics are transforming the life sciences industry. AI is expected to have a larger commercial uptake in the next few years—up four points to 22 percent.1
The medical device sector has more widely adopted AI in R&D over commercialization due to a historically successful commercial model where in-person complex device presentations to healthcare organizations are often greeted with more interest than in-person rep visits to doctors in the pharmaceutical sector. However, the pandemic and its emphasis on virtual engagement changed commercial expectations, especially as healthcare professionals (HCPs) expect increasingly personalized interactions.
A 2021 analysis showed an unprecedented 45 percent increase in AI-driven omnichannel actions, directly impacting over 700,000 healthcare customers’ engagement experience in the U.S.2 Further, leading device manufacturers are substantially increasing investments in commercial technology enablement, using experimentation to validate scaling AI capability deployment across multiple divisions and geographies.
That’s great news for multinational, enterprise device manufacturers, but emerging and mid-sized firms have been hesitant to adopt intelligence solutions enterprise-wide despite its proven benefits. If they invest at all, smaller companies tend to carefully slow-roll AI deployment—at first, in production mode—which can be self-defeating as AI performance improves in real life with more data inputs. The persistent belief that implementing AI is too heavy a lift or expensive is a misperception. AI is no longer a “Rolls-Royce” technology only attainable by organizations with significant resources. Today, AI is for all.
Modern technology means AI is accessible to companies of all sizes. A 2022 survey of senior technology executives revealed 26 percent of companies across all industries now have AI systems in widespread deployment—over double last year’s survey.3 Modular software and open, cloud-based platforms that seamlessly integrate with internal and external data sources combine to bridge the intelligence divide.
Large or small, device manufacturers can’t delay adopting AI to guide customer interactions, or else they risk losing attention or trust. There are too many channels for commercial teams to adeptly coordinate and HCPs feel overwhelmed by impersonal content pushed out to them from all directions. Only AI can efficiently process millions of terabytes of data and provide accurate, data-driven recommendations on the next best actions to take.
One of the most notable ways AI is impacting the medical device industry is enhancing decision-making with actionable, data-driven insights.
AI can process millions of data points in seconds for deeper understanding of customer profiles, treatment patterns, attitudes, and behaviors. Such in-depth knowledge of stakeholder personas considerably improves segmentation and targeting, enabling fast, easy development of highly customized engagement strategies. Marketing departments can quickly engage physicians based on patient types, geography, and prescribing behavior without focus groups, surveys, and other expensive, time-consuming market research. Content can also be accurately personalized to meet stakeholder needs.
Companies using AI are well-positioned for commercial success, magnified 10x for organizations optimizing an omnichannel engagement strategy.
Roche is considering an enterprise-wide AI technology deployment to drive omnichannel engagement with customers and improve their experience. “A lot of companies talk about omnichannel but they are really still operating as a multichannel commercial organization,” said Jim Lefevere, international business leader for Roche Diabetes Care. “If you want to make the move to full omnichannel, you need to leverage AI to be successful.”
Lefevere continued, “I think the medical device industry will see significant uptick in use of AI for commercial as the pandemic wanes. In my experience in a large organization, most new technology implementations benefit from a ‘pilot, test and learn’ approach rather than an enterprise adoption that you may see in smaller organizations, but things are changing quickly.”
Some manufacturers remain hesitant, committing only to piloting small AI programs, if at all. Fortunately, past hurdles for small to mid-size organizations—costs, tech maturity, and culture—are no longer preventing commercial success.
Investors have poured billions of dollars into AI and machine learning (ML) in recent years. Money funneled into AI-focused life sciences companies increased 30 percent in 2020 compared to 2019.4
As a result, technology providers accelerated development of modern solutions built using scalable models that reduce those solutions’ costs. Now, many more companies followed other consumer industries by wading deeper into AI to boost product success as the technology matures and gets cheaper.
Aktana invested heavily to enable scale by productizing technology assets developed over the last decade to create an off-the-shelf intelligence platform. This contrasts with highly custom solutions from five or 10 years ago with expensive service costs and ongoing maintenance. A modern platform using a “build/buy once and deploy multiple times” approach reflects an economy of scale and lowers total cost of ownership by at least 60 percent (depending on deployment size) while empowering companies to cost-efficiently deploy AI and scale across regions, countries, therapeutic areas, and brands.5
Early AI offerings were hampered by a narrow, singular approach. Most solutions fell neatly into one of two buckets: ML or expert systems (an if-then, rules-based approach). Neither technology alone replicates the way humans think. Without necessary context, ML technologies generate conclusions not always practical in the real world. Simultaneously, it would be impossible to catalog every possible scenario a user might encounter with rules alone.
“Across the med device industry, people have been discussing AI and digital technologies for a while…similar to the buzz around social media in the early 2000s,” added Lefevere. “There’s a lot of talk but not a lot of understanding as to how to put it into practice to accomplish our business goals. However, as customer experience and omnichannel engagement become a greater focus, next-generation AI solutions will become essential to success.”
AI 1.0 was defined by technologies working in isolation; the next generation is about synergy. AI 2.0 blends various analytics technologies and human insights for reliable, human-enhancing levels of intelligence for better decision-making. AI solutions have matured, been proven, and are no longer a risky investment for small to mid-size manufacturers.
Today’s AI solutions are also uniquely designed to incrementally add capabilities as they mature and users become more comfortable. A successful AI program will have a scalable platform at its foundation with modules for new and expanding use cases. An emerging digital therapeutics company can invest in a modern intelligence platform and focus first on field-force enablement, then unlock other modules or capabilities like marketing automation, content selection, and channel orchestration.
“The pandemic slowed some innovation, naturally, but our industry must get better at building one-to-one relationships with customers. We need to understand detailed levels of customer interest, needs, and preferences—and this is hard to do unless you implement an intelligence solution. The solutions affording greater flexibility will make adoption easier and more palatable for the risk averse,” said Lefevere.
Modular AI solutions let companies either progress quickly or take a more measured approach. A company wanting to optimize content use can train the AI solution to follow basic rules. As the company learns more about its customers and content usage, that data informs the AI platform to continuously optimize the type of content and its delivery. Technology has matured to allow a continuous loop of more data in and better insights out.
Driverless cars and doctorless diagnosticians must always work flawlessly, and their users must be confident AI-driven innovations will work as expected. Adopting these technologies and their benefits hinges on trust. It’s a catch-22 and can cause resistance within growing device manufacturers.
As AI evolves into a more advanced, contextual version, its predictions become increasingly pivotal to medical decisions. Customer-facing teams are better equipped to deliver the right information to customers at the right time—for instance, highlighting a new device with fewer side effects to a recently diagnosed patient. Most users have little visibility into how AI generates conclusions and this lack of transparency inevitably casts doubt. Many algorithms for ML—particularly popular deep learning approaches—don’t transparently convey the “how” or “why” behind recommendations.
The lack of rationale makes it difficult to ignore misgivings and accept accuracy of machines whose mechanics we cannot easily see or comprehend. We want computer systems to produce transparent explanations for their decisions—a science known as explainable AI (xAI). The most advanced AI solutions now incorporate xAI with a blend of other analytics technologies and human insight to improve decision-making by building real-life context and plain-language explanations into recommendations.
This is a culture shift; AI democratization isn’t about financial discrepancy between the “haves” or “have nots.” In today’s AI market with plummeting total cost of ownership and skyrocketing technical maturity, user acceptance is all that’s holding device makers back from adoption. xAI is the key.
“Change management should be part of any large project, but I also see most med device companies are going through a digital transformation, so employees are more or less primed for ongoing and fast change,” concluded Lefevere. “Change is going to be part of the corporate world for the next decade. To help lessen the tumult, persistently extol the value of new tech to users while identifying those who are antibodies to the new AI system and delicately manage them.”
As trust between humans and machines continues to build, AI 2.0 will become foundational to deepening engagement in today’s digitally driven commercial model. Nearly half of all global healthcare companies plan to implement AI by 2025, saying it’s crucial for business operations.6 The barriers to adoption are crumbling—medical device firms are recognizing this and embracing behavioral change will see there’s nothing holding them back.
References
Alan Kalton is senior vice president of sales for Aktana and GM of Aktana’s European division. His 27-year career balances global consulting with multiple industries in AI, technology, and analytics with life sciences industry experience working with pharmaceutical and medical technology companies. Kalton currently resides in Barcelona, Spain, Aktana’s headquarters for European service operations. He can be reached at alan.kalton@aktana.com.
The medical device sector has more widely adopted AI in R&D over commercialization due to a historically successful commercial model where in-person complex device presentations to healthcare organizations are often greeted with more interest than in-person rep visits to doctors in the pharmaceutical sector. However, the pandemic and its emphasis on virtual engagement changed commercial expectations, especially as healthcare professionals (HCPs) expect increasingly personalized interactions.
A 2021 analysis showed an unprecedented 45 percent increase in AI-driven omnichannel actions, directly impacting over 700,000 healthcare customers’ engagement experience in the U.S.2 Further, leading device manufacturers are substantially increasing investments in commercial technology enablement, using experimentation to validate scaling AI capability deployment across multiple divisions and geographies.
That’s great news for multinational, enterprise device manufacturers, but emerging and mid-sized firms have been hesitant to adopt intelligence solutions enterprise-wide despite its proven benefits. If they invest at all, smaller companies tend to carefully slow-roll AI deployment—at first, in production mode—which can be self-defeating as AI performance improves in real life with more data inputs. The persistent belief that implementing AI is too heavy a lift or expensive is a misperception. AI is no longer a “Rolls-Royce” technology only attainable by organizations with significant resources. Today, AI is for all.
Modern technology means AI is accessible to companies of all sizes. A 2022 survey of senior technology executives revealed 26 percent of companies across all industries now have AI systems in widespread deployment—over double last year’s survey.3 Modular software and open, cloud-based platforms that seamlessly integrate with internal and external data sources combine to bridge the intelligence divide.
Large or small, device manufacturers can’t delay adopting AI to guide customer interactions, or else they risk losing attention or trust. There are too many channels for commercial teams to adeptly coordinate and HCPs feel overwhelmed by impersonal content pushed out to them from all directions. Only AI can efficiently process millions of terabytes of data and provide accurate, data-driven recommendations on the next best actions to take.
One of the most notable ways AI is impacting the medical device industry is enhancing decision-making with actionable, data-driven insights.
AI can process millions of data points in seconds for deeper understanding of customer profiles, treatment patterns, attitudes, and behaviors. Such in-depth knowledge of stakeholder personas considerably improves segmentation and targeting, enabling fast, easy development of highly customized engagement strategies. Marketing departments can quickly engage physicians based on patient types, geography, and prescribing behavior without focus groups, surveys, and other expensive, time-consuming market research. Content can also be accurately personalized to meet stakeholder needs.
Companies using AI are well-positioned for commercial success, magnified 10x for organizations optimizing an omnichannel engagement strategy.
Roche is considering an enterprise-wide AI technology deployment to drive omnichannel engagement with customers and improve their experience. “A lot of companies talk about omnichannel but they are really still operating as a multichannel commercial organization,” said Jim Lefevere, international business leader for Roche Diabetes Care. “If you want to make the move to full omnichannel, you need to leverage AI to be successful.”
Lefevere continued, “I think the medical device industry will see significant uptick in use of AI for commercial as the pandemic wanes. In my experience in a large organization, most new technology implementations benefit from a ‘pilot, test and learn’ approach rather than an enterprise adoption that you may see in smaller organizations, but things are changing quickly.”
Some manufacturers remain hesitant, committing only to piloting small AI programs, if at all. Fortunately, past hurdles for small to mid-size organizations—costs, tech maturity, and culture—are no longer preventing commercial success.
Investors have poured billions of dollars into AI and machine learning (ML) in recent years. Money funneled into AI-focused life sciences companies increased 30 percent in 2020 compared to 2019.4
As a result, technology providers accelerated development of modern solutions built using scalable models that reduce those solutions’ costs. Now, many more companies followed other consumer industries by wading deeper into AI to boost product success as the technology matures and gets cheaper.
Aktana invested heavily to enable scale by productizing technology assets developed over the last decade to create an off-the-shelf intelligence platform. This contrasts with highly custom solutions from five or 10 years ago with expensive service costs and ongoing maintenance. A modern platform using a “build/buy once and deploy multiple times” approach reflects an economy of scale and lowers total cost of ownership by at least 60 percent (depending on deployment size) while empowering companies to cost-efficiently deploy AI and scale across regions, countries, therapeutic areas, and brands.5
Early AI offerings were hampered by a narrow, singular approach. Most solutions fell neatly into one of two buckets: ML or expert systems (an if-then, rules-based approach). Neither technology alone replicates the way humans think. Without necessary context, ML technologies generate conclusions not always practical in the real world. Simultaneously, it would be impossible to catalog every possible scenario a user might encounter with rules alone.
“Across the med device industry, people have been discussing AI and digital technologies for a while…similar to the buzz around social media in the early 2000s,” added Lefevere. “There’s a lot of talk but not a lot of understanding as to how to put it into practice to accomplish our business goals. However, as customer experience and omnichannel engagement become a greater focus, next-generation AI solutions will become essential to success.”
AI 1.0 was defined by technologies working in isolation; the next generation is about synergy. AI 2.0 blends various analytics technologies and human insights for reliable, human-enhancing levels of intelligence for better decision-making. AI solutions have matured, been proven, and are no longer a risky investment for small to mid-size manufacturers.
Today’s AI solutions are also uniquely designed to incrementally add capabilities as they mature and users become more comfortable. A successful AI program will have a scalable platform at its foundation with modules for new and expanding use cases. An emerging digital therapeutics company can invest in a modern intelligence platform and focus first on field-force enablement, then unlock other modules or capabilities like marketing automation, content selection, and channel orchestration.
“The pandemic slowed some innovation, naturally, but our industry must get better at building one-to-one relationships with customers. We need to understand detailed levels of customer interest, needs, and preferences—and this is hard to do unless you implement an intelligence solution. The solutions affording greater flexibility will make adoption easier and more palatable for the risk averse,” said Lefevere.
Modular AI solutions let companies either progress quickly or take a more measured approach. A company wanting to optimize content use can train the AI solution to follow basic rules. As the company learns more about its customers and content usage, that data informs the AI platform to continuously optimize the type of content and its delivery. Technology has matured to allow a continuous loop of more data in and better insights out.
Driverless cars and doctorless diagnosticians must always work flawlessly, and their users must be confident AI-driven innovations will work as expected. Adopting these technologies and their benefits hinges on trust. It’s a catch-22 and can cause resistance within growing device manufacturers.
As AI evolves into a more advanced, contextual version, its predictions become increasingly pivotal to medical decisions. Customer-facing teams are better equipped to deliver the right information to customers at the right time—for instance, highlighting a new device with fewer side effects to a recently diagnosed patient. Most users have little visibility into how AI generates conclusions and this lack of transparency inevitably casts doubt. Many algorithms for ML—particularly popular deep learning approaches—don’t transparently convey the “how” or “why” behind recommendations.
The lack of rationale makes it difficult to ignore misgivings and accept accuracy of machines whose mechanics we cannot easily see or comprehend. We want computer systems to produce transparent explanations for their decisions—a science known as explainable AI (xAI). The most advanced AI solutions now incorporate xAI with a blend of other analytics technologies and human insight to improve decision-making by building real-life context and plain-language explanations into recommendations.
This is a culture shift; AI democratization isn’t about financial discrepancy between the “haves” or “have nots.” In today’s AI market with plummeting total cost of ownership and skyrocketing technical maturity, user acceptance is all that’s holding device makers back from adoption. xAI is the key.
“Change management should be part of any large project, but I also see most med device companies are going through a digital transformation, so employees are more or less primed for ongoing and fast change,” concluded Lefevere. “Change is going to be part of the corporate world for the next decade. To help lessen the tumult, persistently extol the value of new tech to users while identifying those who are antibodies to the new AI system and delicately manage them.”
As trust between humans and machines continues to build, AI 2.0 will become foundational to deepening engagement in today’s digitally driven commercial model. Nearly half of all global healthcare companies plan to implement AI by 2025, saying it’s crucial for business operations.6 The barriers to adoption are crumbling—medical device firms are recognizing this and embracing behavioral change will see there’s nothing holding them back.
References
- GlobalData, Pharma Intelligence Center, “AI and Big Data Will Continue to Disrupt the Pharmaceutical Sector,” (July 15, 2021).
- PharmaVOICE, “How AI is Helping Pharma Solve Some of its Covid Problems” by Meagan Parrish (January 25, 2022).
- MIT Sloan Management Review, “Companies are Making Serious Money with AI,” by Thomas H. Davenport and Randy Bean (February 17, 2022).
- Deep Pharma Intelligence Report, as seen in Financial Times (February 2022).
- Aktana Data—based on average operating value between 2018 and 2021.
- PharmaNewsIntelligence, “AI in the Pharma Industry: Current Uses, Best Cases, Digital Future,” by Samantha McGrail (April 30, 2021).
Alan Kalton is senior vice president of sales for Aktana and GM of Aktana’s European division. His 27-year career balances global consulting with multiple industries in AI, technology, and analytics with life sciences industry experience working with pharmaceutical and medical technology companies. Kalton currently resides in Barcelona, Spain, Aktana’s headquarters for European service operations. He can be reached at alan.kalton@aktana.com.