Capturing AI’s Value in Healthcare
The healthcare industry is witnessing a significant evolution with the advent of Artificial Intelligence (AI), promising transformative changes. AI associated innovation revenues currently make up roughly 5% (USD 245 bn) of the total USD 5.1 tn Healthcare sector revenues in the listed equity universe. When it comes to AI in healthcare, investors are easily lost between hype, promise, and actual potential. Our Singularity Think Tank experts cut through the noise, pointing us towards the needles in the haystack.
Revolutionizing Medicine? AI’s Potential in Transforming Drug Discovery and Development
AI’s burgeoning role in drug discovery can be explained by its ability to swiftly analyze complex biological data, allowing algorithms to identify promising drug candidates at an unprecedented pace. Singularity Think Tank expert Andreas Schneider, PhD, Head of Innovation Delivery Systems at Ypsomed Delivery Systems, notes that “AI is increasingly being applied to the less regulated front-end of the drug discovery process.” By enhancing early-stage research, companies are using AI for more efficient and targeted drug development while circumventing more heavily regulated, lengthy, and cost intensive PhaseII/III clinical stages of the drug development journey.
The pharmaceutical industry has seen examples of AI-designed drug candidates progress to clinical trials in record time, a feat that underscores AI’s potential to shorten the lengthy timelines traditionally associated with drug development. However, as Singularity Think Tank expert for healthcare innovation Carsten Danzer, PhD, notes, the industry is cautious with financial commitments: “Pharma’s AI/ML investments are still minor compared to overall R&D spending, suggesting a lack of trust in AI being the solution for significantly increasing drug discovery success rates.” Morgan Stanley Research projects that AI/ML related Research & Development spending by Biopharma companies will grow from 1.5% in 2023 to 4% in 2030, yet new developments in the fast-moving AI space and successful applications could boost such investments more rapidly.
Discussing AI’s more immediate applications, Danzer notes that “Short term, AI is likely to prove more valuable in the rare disease research domain. Pharma has increasingly allocated resources for discovery and development there to benefit from shorter approval trajectories and as a springboard for other patient populations. But there too, we’re still learning AI’s full potential. Setting up clinical trials as well as monitoring and managing patient compliance in the clinical trial stages is another area where the gains from AI could be immense, especially given the high costs of running such trials.”
Rapid Advancements in Augmented Diagnostics
In one of Danzer’s own ventures, SEQSTANT, he is part of a team developing a software solution for real time holistic analysis and detection of pathogens from clinical samples. The company’s LiveGene product is leveraging existing Next Generation Sequencing technology as well as underused resources in diagnostic laboratories to deliver faster, easy-to-use and comprehensive insights. Among the various use cases, their technology enables faster, on-site detection of bacterial infection strains in hospitals, allowing a more targeted and almost immediate match with suitable antibiotics. While such an application lends itself well for AI, SEQSTANT’s algorithms do not use AI to identify pathogens yet. This is due to the lack of large, high quality datasets, and the difficult approval path under the new European Regulation for In-Vitro Diagnostic Medical Devices (IVDR). Under that regulation, an AI diagnostic device would be categorized among the highest risk classes ©, making market entry both lengthy and costly.
Medical Large Language Models
Beyond drug discovery and development, industry practitioners and researchers in medical imaging are showing great interest in AI’s pattern recognition skills. Algorithms are now able to augment human intelligence by assisting in the detection and diagnosis of diseases, sometimes with greater accuracy than experienced clinicians. Singularity Think Tank expert for healthcare AI, Simone Lionetti, PhD, Senior Researcher in the Algorithmic Business Lab at Hochschule Luzern, observes the profound effect of AI in his domain: “Large Language Models are improving at a pace that has taken most experts by surprise. Medical language models such as MedAlpaca, which is based on Meta’s Platform’s AI’s LLaMA model, and Alphabet’s Med-PaLM 2 are proving very powerful in medical question-answering. Yet the challenges surrounding data reliability — which are paramount to ensure the accuracy and efficacy of AI in clinical settings — are still considerable. Obtaining high quality data in clinical settings, especially in disease areas where data is costly and scarce, is still a major bottleneck,” notes Lionetti.
Tailored Treatments, Global Health, and Patient-Centricity
AI has a big future in healthcare, yet it demands a balanced approach. The practicality of AI’s integration into healthcare systems is slowed down by the need for meticulous testing and validation. Lionetti’s concerns about reliability are a reminder that AI must be vetted rigorously before being trusted in clinical environments.
As the industry moves forward, it is clear that AI will continue to transform healthcare and help medical professionals in areas such as personalized medicine to tailor treatments for individual patients. Another promising application of AI concerns global health management. Projects such as EPIWATCH apply predictive modeling for epidemic monitoring and alerts, helping governments and healthcare organizations to manage and contain disease outbreaks such as the recent COVID-19 pandemic.
The key to unlocking new treatments and understanding diseases more deeply with the help of AI will lie in the responsible development of AI tools. As companies continue to race for new AI applications, only a true patient-centric commitment to the highest standards of care will ensure that AI’s growth in healthcare leads to real benefits. Achieving this challenge will depend on innovative companies’ ability to successfully collaborate with healthcare professionals and patients, navigate regulatory constraints, and tackle difficult ethical considerations such as data diversity and transparency.
The insights from healthcare’s ongoing AI revolution underscore the strategic imperative for TSG’s investment strategy: To identify and invest in applied innovations within the global listed equity universe that not only promise transformative potential but also demonstrate practical, scalable applications. This approach ensures a focus on AI advancements grounded in real-world market viability, poised to deliver tangible value to both the healthcare industry and investors alike.
About The Singularity Group
The Singularity Group (TSG) quantifies applied innovation for investors in listed equities. TSG is the initiator of the Singularity Index™ (Bloomberg ticker: NQ2045), a global, all-sector benchmark and gold standard for applied innovation. The Singularity Strategies include The Singularity Fund (UCITS Lux), Singularity Reshoring(UBS AMC), and the Singularity Small&Mid (UBS AMC). The Swiss investment advisory boutique works closely with the Singularity Think Tank, a network of entrepreneurs and academics with deep insights into innovation value chains. Their input forms the foundation of TSG’s proprietary innovation scoring system that quantifies the engagement of companies within a set of curated Singularity Sectors worldwide across all market capitalizations and industries. The Singularity Innovation Score (SI-Score; see below) defines how much value listed companies are generating through applied innovation.
More: www.singularity-group.com
The Singularity Innovation Score (SI-Score): A company’s SI-Score represents the percentage of its revenues associated with innovation. It reflects a company’s ability to create innovation- versus commoditized -business and -cash flows, and its ability to participate in technological evolution. Changes in the SI-Score are just as important as the absolute value. A company’s SI-Score relative to its overall GICS sector can say a lot about the competitive standing and ability to gain and maintain market share. Regional SI-Scores can be used to evaluate the innovation power of markets as well as to gauge companies’ standing in different regions.
Innovation Revenues: 4% of a 74 trillion USD global market: In 2022, the total revenues generated by the World’s listed equities amounted to USD 74 Trillion. TSG’s unique expert-led innovation screening and scoring methodology allows us to divide that amount into innovation revenues and non-innovation revenues. In 2022, roughly 4% (USD 3 Trillion) of global revenues qualified as innovation revenues.
