by Amara Graps

Health became a number one global issue during the height of the pandemic, with covid-19 becoming the most studied disease in human history (438,953 publications as of August 28, 2024). Human health was on everyone’s minds in the quantum technology field too, with quantum chemistry providing a valuable application focus for the quantum computing field. For the first-generation, quantum computing annealing architecture: D-Wave, optimization problems (see GQI’s Quantum Solvers Focus Report: Optimization) were, and are, the ‘low-hanging fruit’.  The quantum algorithms company: 1QBit,  after broadening its algorithms to work on gate-based computers, spun out in 2022, the company: Good Chemistry. The Good Chemistry company, in January 2024 was acquired by SandboxAQ, which has a Life Sciences division.

Today, if one combines ‘Life Sciences’ with ‘Health’ and ‘Chemicals and Materials’ in GQI’s Tracker: Quantum Computing Use Cases, (See Figure), quantum chemistry is the largest category. Its ‘toy demonstrations’ label shouldn’t bely its serious activity. This is a lucrative sub-field with a high social impact.

Quantum Computing Use cases presented by Industry and Implementation Status in GQI’s Use Cases Tracker(*)

Early in the pandemic in June 2020, in an online presentation for the BC Quantum User Group, Arman Zaribafiyan presented what I call the most compelling Use Case of quantum chemistry: minimizing the molecular ground state energy, to help pharmaceutical companies. Zaribafiyan is the former first employee of 1Qbit, who became the Founder and CEO of Good Chemistry, before the company’s acquisition by SandboxAQ. The molecular ground-state energy was determined using the hybrid VQE algorithm with a gate-based quantum computer. He described his algorithm approach beginning with Kandala et al., 2017’s Hardware-efficient variational quantum eigensolver for small molecules and quantum magnets. See QCR’s The Many Faces of Hybrid Classical-Quantum Computing: Part 3 which described the VQE in more detail. (Previous parts to The Many Faces of Hybrid Classical-Quantum Computing are here: Part 1, and here: Part 2)

Zaribafiyan described in his June 2020 talk, the expensive journey, which pharmaceutical companies undergo in their development of pharmaceutical drugs. They must invest $1-4B and 10-15 years on average. If we compare pharmaceutical development to the space industry, SpaceX’s Falcon 9 took five years and roughly $400M to complete. Such an expensive journey gives rise to orphan drugs, where less than 10% reach the market to treat patients. If one simulates the molecular energies before the expensive laboratory work, then drug development expenses and time are reduced. In his 2020 presentation, Zaribafiyan showed a study on hydrocarbons, where 60–85% fewer qubits were required to accurately mimic their total energies. Fewer qubits fit our NISQ era.  You can see Zaribafiyan’s own quantum computing research, which drove some of these business endeavors, in his PhD.  A recent summary of quantum computers in drug development by Santagati et al., 2024: Drug design on quantum computers, provides a state-of-the-art view, and an older and more detailed summary can be seen in Blunt et al., 2022.

Multidisciplinary Health and Medicine Quantum Computing Use Cases

One of my favorite quantum computing survey papers is Floether’s The state of quantum computing applications in health and medicine. It’s only 10-pages, yet, readable in its conciseness, and presents a useful Table that lists 40 Health and Medical Use Cases.

Floether’s paper groups the health and medicine research into three main areas:

  1. Genomics and clinical research
  2. Diagnostics
  3. Treatments and interventions.

His Figure 1 links those three areas to three key quantum computing use cases:

  1. Nature simulation, encompassing physics, chemistry, and materials science.
  2. Handling complexly structured data, such as factoring, ranking, and artificial intelligence/machine learning (AI/ML).
  3. Search and optimization, which includes risk assessment, sampling, and pricing

The wider the connecting line, in the figure, the more applicable the category.

Quantum AI/ML Terminology

Notice the widely-used theme in that figure: ‘Processing data with complex structure’, i.e., Handling complexly structured data, such as factoring, ranking, and artificial intelligence/machine learning (AI/ML).  AI/ML in quantum computing is a broad and deep topic.

To help clarify the terminology in the research literature, Floether writes that in the context of quantum AI/ML, variational quantum circuits are sometimes regarded as the fundamental components of quantum neural networks (QNNs). QNNs are neural networks in which parameterized quantum circuits (PQCs) are added to the hidden layers. In other cases, a PQC, quantum circuit learning (QCL), and a variational quantum circuit (VQC) are considered synonymous with a QNN.

In future articles, I’ll describe the importance of AI/ML to augment quantum computing, provide more explanations on the AI/ML broad and deep topic, and highlight some of the most active Health Use Cases.

(*) For those eager to explore the interactive quantum computing use case dashboard, which offers the ability to filter results by specific companies, geographies, technology approaches, or hardware compatibilities, please don’t hesitate to contact info@global-qi.com.

September 2, 2024