France: AI & The Quest for Life Sciences 4.0


With Minister of the Economy and Finance, Bruno Le Maire, emphasizing the importance of “winning the battle on data” and eccentric mathematician Cédric Villani, well in the process of drawing up a national strategy on artificial intelligence (AI) in his capacity as Chargé de Mission, there is little doubting the new administration’s credentials when it comes to embracing digital disruption and upending traditional practices of drug discovery and healthcare provision.

“Deep learning is an absolutely extraordinary tool that can help us to realize considerable leaps forward in prevention, prediction, personalized medicine and precision treatment so it is absolutely imperative that we seize this opportunity,” exclaims Villani.


De-Risking Drug Discovery

On the supply side, AI offers a real prospect of de-risking the drug discovery process and doing away with many traditional time-consuming approaches at a juncture when drug development has been becoming unsustainable: costing an average of $2.6 billion and 12 years to bring a new therapy to market. “AI can be successfully harnessed both to accelerate the drug discovery process and to augment human skills in the execution of clinical trial procedures… It is our sincere belief that the technology will ultimately completely reshuffle how we treat diseases, think about patients, and design clinical studies,” posits Thomas Clozel, co-founder and CEO of data science outfit, OWKIN, which counts ex-Novartis Oncology CEO Bruno Strigini as its chairman.

“There is already a strong precedence in drug discovery, but the best way for AI to be deployed actually relates to enhanced usage of real-world data,” he muses. “Augmentation is trying to better understand the diseases and predict which variations will be resistant to new therapies or how new treatments may react with the patient and outcome research such as response to targeted therapies are frankly difficult to measure without this tool.”

Moreover, AI opens up new doors to the practice of “drug repurposing” whereby existing molecules are combined in ways that furnish them with therapeutic powers that each lacks in isolation. Much still depends, however, on from where data is sourced. “Any algorithm in AI needs to be trained on sound data and the challenge is that many of the first companies active in AI-backed drug discovery are built on public data much of which is imperfect…we expect this to change with time, though, as exciting new companies come on stream that have been creating their own datasets,” notes Clozel.


Towards “Industrial Medicine”

On the demand side (i.e. healthcare provision), AI augurs nothing less than a switch over to “industrial medicine” and the extensive optimization of care pathways. This is because the technology can be mobilized to enable providers to diagnose illnesses earlier with greater accuracy and consequently to ultimately manage disease it more effectively.

“The dream is to pivot away from the current ‘break & fix’ model of medicine towards a posture of ‘predict & prevent.’ This model is naturally considerably less costly for society. Through AI, we can predict the risk of experiencing cardiovascular events, for instance, increasing the precision of these forecasts from 50 percent to 75 percent,” explains Amgen’s former general manager, Jean Monin.

It is in the medtech segment, however, where this shifting of paradigm is likely to be most keenly felt first. “AI might be taking a bit of time to upheave the pharma sector, but actors like us are already deeply invested in these technologies… We are increasingly integrating deep learning in our solutions to help doctors diagnose more rapidly, confidently and precisely. Our biggest priority is to make sure we can connect the dots and build a network of data,” relates Christophe Lala, GE Healthcare’s general manager for Western Europe.


Federated Learning

OWKIN, meanwhile, has developed a technology platform called federated learning. This overcomes the data sharing problem, building collective intelligence from distributed data at scale while preserving data privacy and security. “We build the algorithms for our program within the firewalls of hospitals, ensuring that the data never leaves the hospital. Once the algorithm is trained, we are able to extract the program and apply it to other uses. This unique approach has helped us to build trust with many hospitals in both the US and France, leading to the creation of many fruitful partnerships,” reveals Clozel.

Still there are many hurdles yet to overcome. “In France, there are many initiatives to encourage data sharing, but it is hard to determine who should hold the rights to the data. There is still a need for a more clearly defined legal framework in this context. Moreover, benchmarks of anonymization are not perfectly set up, which creates something of a problem as we enter the genomic era where it is possible to cross-reference data with public genetic public databases thus rendering it possible to find specific patients or family members,” he warns.


France In Pole Position

Many believe, however, that France is one of the nations best placed to grasp the initiative and fully exploit ‘big data’ technologies. “First, France possesses the scale, centralized structures and the competencies o become a frontrunner of digitalization in life sciences. Second, we benefit from a large pool of talented scientists, a strong educational system and a patient-centric approach to healthcare. Thirdly there is the political courage and willingness in place,” argues Lala.

Moreover, the establishment of a nationwide “Health Data Hub” looks set to endow France with a considerable advantage in relation to its peers. The Hub’s director, Jean-Marc Aubert, explains that “The goal is to gather all the data spread across hospitals, clinics, research institutes, and so forth, together in order to analyse the information with the most modern sophisticated AI technology available.”

François Vorms, managing director of Canon Medical Systems, adds, “Having homogeneous data and large amounts of it is absolutely critical to the success of AI. The major barrier to progress thus far has been that companies have been forced to work in silos mostly limited to their own internal data and that of their partners, so having the Hub in place promises to be a complete game changer.”

B. Braun president, Marc-Alexander Burmeister, very much agrees. “We have to ensure that we are fully prepared to be able to support this interface with whatever we do, either as a product supplier or a service provider of healthcare… we will be looking to fit right into this new environment and are intent on making sure that we are not part of the problem where data is trapped because of a lack of connectivity,” he affirms.


Integrator Services

Luckily help will also be on hand from a new breed of life science actors: namely those with the capabilities to play the role of integrator. “Connectivity and interoperability lie at the very core of our value proposition as a trusted partner in the health data journey,” exclaims Elie Lobel, CEO of Orange Healthcare. “In the past, healthcare organizations like hospitals worked in virtual isolation, not really interacting with other parts of the care ecosystem. With today’s care pathways, however the need for regulated and public health data infrastructure is rising. These channels are coordinated between different actors across the healthcare spectrum and the need to exchange and share data between different organizations, heterogeneous applications, and medical devices has never been more urgent,” he clarifies.

Even the factory floor is vulnerable to disruption. Medtech icon, B. Braun, for instance is working on incorporating collaborative robots or ‘co-bots’ into shared workplaces from where they can physically interact with the company’s human capital. “With the technology of AI, we are finding we can now have human workers assisted by robots managing the production of our devices. The co-bot actually looks very similar to a human being, with a torso and arms and can function to help workers assemble delicate pieces and learn as well. Our objective is to integrate cobots in way that affords employees better flexibility and minimizes the risk of repetitive strain injuries,” explains company president, Marc-Alexander Burmeister.

“This does, of course, imply that the profile of workers needed in our manufacturing sites will need to evolve, and the redefinition of job profiles is clearly something we need to think hard about and anticipate,” he adds.

It is precisely this need for integration that is fueling Orange Healthcare’s organizational engine as it strives to provide integrated platforms that render it possible for big hospital groups and territories to work together collaboratively and seamlessly. “In Europe, the experience of a patient at a hospital compared to that of, say, a frequent flyer with an airline is currently very different and we seek to change all that by remodelling how a patient interacts with the healthcare ecosystem he or she is passing through,” argues Lobel.

Related Content

Latest Report