Welcome to The Daily Qubit!

Get the latest in top quantum news and research Monday through Friday, summarized for quick reading so you stay informed without missing a qubit.

Have questions, feedback, or ideas? Fill out the survey at the end of the issue or email me directly at [email protected].

And remember—friends don’t let friends miss out on the quantum era. If you enjoy The Daily Qubit, pass it along to others who’d appreciate it too.

Happy reading and onward!

Cierra

Today’s issue includes:

• A hybrid digital-analog quantum convolutional neural network for image classification improves feature extraction and enhance the classification of medical images.

• Researchers applied quantum annealing to optimize robotic assembly line balancing, a manufacturing problem.

• A platform combines quantum sensing and droplet microfluidics. Using nitrogen-vacancy centers in nanodiamonds, this method enables high-precision chemical detection and imaging in flowing, monodisperse microdroplets.

QUANTUM APPLICATION HEADLINES

⚕️ Quantum Model Enhances Medical Image Classification Using Neutral-Atom Processors

APPLICATION: A team from Kipu Quantum and the University of Valencia propose a hybrid digital-analog quantum convolutional neural network for image classification, using neutral-atom quantum processors to improve feature extraction and enhance the classification of medical images.

SIGNIFICANCE: Medical image classification requires models capable of detecting subtle patterns in complex datasets, such as identifying early indicators of diseases like breast cancer or pneumonia. Classical convolutional neural networks are widely used for this purpose, as they apply trainable filters to scan images and extract features like edges, shapes, and textures. However, classical CNNs rely heavily on large, labeled datasets for training, which can be challenging to obtain in medical contexts due to privacy concerns and the high cost of expert annotation. Additionally, these models require significant computational resources as they grow in complexity. Digital-analog quantum convolutional neural networks integrate quantum kernels, which capture correlations in data that classical models struggle to replicate. This may lead to the identification of richer features with fewer parameters.

HOW: The DAQCNN uses quantum kernels derived from the natural Ising interactions in neutral-atom processors. Each image is encoded into quantum states via single-qubit rotations, followed by analog blocks that entangle the qubits to capture complex features. By varying the connectivity graphs of the analog blocks, the network identifies a broader range of image features. These quantum-enhanced features are then processed through classical convolutional layers and dense networks for classification. The approach was tested on datasets for breast cancer and pneumonia, using both single and multiple quantum kernel configurations to assess performance.

BY THE NUMBERS:

• 9 qubits — Largest kernel size used, representing a 3x3 grid of pixels encoded on quantum states.

• 92.6% AUC (Breast Cancer) — The DAQCNN achieved an area under the curve (AUC) of 92.6% for breast cancer detection with four kernel graphs, outperforming classical counterparts with the same architecture.

• 96.2% AUC (Pneumonia) — The DAQCNN reached 96.2% AUC for pneumonia detection, highlighting its applicability to various medical datasets.

• 10x fewer parameters — Compared to large classical models like ResNet-50, DAQCNN uses significantly fewer trainable parameters while achieving competitive results.

🤖 Quantum Annealing for Robotic Assembly Line Optimization in Manufacturing

APPLICATION: Researchers from the Fraunhofer Institute and the University of Stuttgart applied quantum annealing to optimize robotic assembly line balancing (RALB), a manufacturing problem. They used annealing quantum computing to enhance task distribution across workstations and improve manufacturing productivity.

SIGNIFICANCE: Manufacturing systems are becoming increasingly complex due to mass customization, where products are tailored to meet specific customer needs rather than produced uniformly. This shift demands flexible and dynamic production processes, often involving a higher number of tasks, resources, and dependencies that must be managed efficiently. Traditional optimization methods, such as integer programming or heuristic algorithms, struggle to scale to these larger and more intricate problems, as the combinatorial nature of assigning tasks to workstations and equipment quickly gets out of hand. While current quantum hardware is still in its early stages and limited by noise and connectivity constraints, it holds the potential to improve scalability and solution quality for large-scale industrial optimization problems as the technology matures.

HOW: The study reformulated the RALB problem from an integer programming model into a quadratic unconstrained binary optimization problem, which is suitable for quantum computation. This transformation involved turning hard constraints into soft constraints, enabling the use of quantum annealing to identify optimal task distributions. The researchers employed a hybrid approach, where the QUBO problem was partitioned into smaller sub-problems. These sub-problems were then solved on D-Wave’s quantum processor, with additional support from classical heuristics to enhance performance. A Python library was also developed to automate this transformation process for broader adoption of quantum methods for RALB optimization.

BY THE NUMBERS

• 64 qubits — The quantum processor encoded the problem with 64 binary decision variables indicating the ability of current quantum systems to take on moderately complex industrial problems.

• 4.3 seconds (simulated) vs. 10,110 seconds (quantum) — While simulated annealing (a classical technique) completed 1,000 samples significantly faster, quantum annealing shows promise for scalability as quantum hardware improves.

• 34 optimal solutions out of 1,000 samples — The hybrid approach produced 34 configurations that met all constraints and minimized costs.

💧 Quantum Sensing Meets Microfluidics: High-Precision Chemical Detection with Nanodiamond Droplets

APPLICATION: Researchers from the University of California, Berkeley, in collaboration with Lawrence Berkeley National Laboratory, have developed a platform combining quantum sensing and droplet microfluidics. Using nitrogen-vacancy centers in nanodiamonds, this method enables high-precision chemical detection and imaging in flowing, monodisperse microdroplets.

SIGNIFICANCE: Quantum sensing uses the unique properties of quantum systems to achieve highly precise measurements, often beyond the limits of classical methods. However, challenges like particle heterogeneity and signal noise have limited its broader use. By encapsulating quantum sensors in picoliter-scale droplets—tiny, controlled liquid volumes—this approach creates stable environments for sensing, significantly improving consistency and reducing noise. This advancement enables high-throughput analysis of analytes with minimal sample volumes. Potential applications include real-time measurements inside living cells, analyzing individual cell metabolites, and developing portable diagnostic devices.

HOW: The platform integrates nanodiamonds with NV centers into droplets formed using water-in-oil emulsions. These droplets act as microchambers, stabilizing the environment for precise quantum sensing. A double lock-in detection method enhances the optical signal-to-noise ratio by averaging fluctuations from individual nanodiamonds within droplets. By flowing through a microfluidic chip, the droplets facilitate efficient mixing and detection of paramagnetic analytes, achieving unprecedented measurement stability over extended durations. The researchers demonstrated the platform’s efficacy by detecting gadolinium ions and radicals like TEMPOL, showcasing its applicability in bioengineering and diagnostics.

BY THE NUMBERS: 

• 0.23% variance — Inter-droplet nanodiamond variability, which shows high consistency in measurement in chemical sensing.

• 2% precision —The double lock-in detection method improved precision in detecting signal changes, allowing for highly accurate analysis of minute chemical differences.

• 500 nM detection limit — The platform can detect gadolinium ions at a concentration as low as 500 nM (equivalent to a few molecules in a billion water molecules), highlighting its sensitivity for detecting trace amounts of substances.

• 1 million droplets/hour — Throughput of the system, making it suitable for high-throughput applications.

RESEARCH HIGHLIGHTS

🐉 A research team from Keio University, RIKEN, Mercari Inc., and others propose "Q-Fly," a scalable optical interconnect architecture for modular quantum computers, inspired by the Dragonfly topology in classical systems. This architecture minimizes photon loss and network hops while supporting high-fidelity entanglement across quantum nodes, necessary for distributed quantum computation. Experimental implementation of a three-node prototype demonstrated two-hop entanglement with fidelities of 0.60–0.64, showing promise for practical large-scale quantum systems.

🔥 The researchers introduced the Disentangling Quantum Neural Network (DEQNN) to unify the estimation of quantum entropies and distance measures, which are critical for understanding quantum systems. By reducing the Hilbert space size while preserving these key physical quantities, DEQNN simplifies the computation of measures such as von Neumann entropy, Rényi entropy, and fidelity.

✅ The researchers developed a hybrid atom tweezer array using dual isotopes of ytterbium (171Yb and 174Yb) for quantum error correction protocols. This system uses nuclear spin qubits for data storage and optical clock qubits for ancilla-based operations, achieving high coherence and low crosstalk during midcircuit measurements. Experimental results demonstrated 99.1% coherence retention and high discrimination fidelity.

NEWS QUICK BYTES

✨ Illinois and IBM have announced the establishment of the National Quantum Algorithm Center in Chicago’s Illinois Quantum and Microelectronics Park. Anchored by IBM’s Quantum System Two, powered by the Heron processor, the center is intended to develop advanced quantum algorithms and hybrid quantum-classical workflows to tackle complex industry challenges.

💰️ D-Wave Quantum Inc. raised $175 million in gross proceeds through two equity offering programs, with the latest $75 million tranche completed at an average share price of $4.8149. The funds will support working capital, capital expenditures, and technical development.

🏫 Heriot-Watt University has secured funding for three projects under the £4.7 million UKRI-funded Photonics and Quantum Accelerator (PQA), designated for advancing Scotland’s photonics and quantum sectors. The projects focus on quantum photonic systems for secure data encryption, waveguides to mitigate atmospheric turbulence in communications, and autonomous semiconductor device assembly for advanced material creation.

📱 Nokia and SK Broadband have deployed a quantum-secure network for Korea Hydro and Nuclear Power (KHNP) to safeguard its critical IT infrastructure against current and future cyber threats, including those posed by quantum computing. The network uses Nokia’s Quantum-Safe MACsec cryptographic technologies and advanced optical solutions, ensuring both high performance and robust security.

❄️ AmpliTech Group has developed and deployed proprietary low-noise cryogenic amplifiers, critical for enabling quantum computers to operate efficiently at ultra-low temperatures of 4 Kelvin (-452°F). These amplifiers, delivered to Fortune 50 quantum computing leaders, universities, and research institutions, minimize noise to ensure precise quantum signal detection, supporting scalable, error-corrected quantum systems.

🤝 Quobly and STMicroelectronics have partnered to develop scalable and cost-effective quantum processor units leveraging ST’s FD-SOI semiconductor process, targeting a 1-million-qubit advance by 2031. The collaboration begins with adapting ST’s 28nm FD-SOI technology for a 100-qubit machine scalable to over 100k qubits, addressing challenges in size, weight, power, and cost (SWaP-C) while enabling long-term scalability through CMOS wafer-scale manufacturing.

🌥️ ZenaTech has launched the Sky Traffic project, integrating quantum computing with AI-driven drones to optimize traffic management and enhance weather forecasting. Using ZenaDrone 1000 drones, the initiative provides real-time traffic data to improve traffic flows, signal management, and public safety while leveraging quantum computing on AWS for rapid data processing. Additionally, the drones will be utilized in weather radar applications, enabling faster and more precise forecasting of extreme events.

QUANTUM MEDIA

LISTEN

Zoran Krunic, Senior Manager of Data Science at Amgen Research and Development, is interviewed by Yuval Boger. Zoran and Yuval talk about leveraging quantum machine learning for clinical trial data, the challenges of implementing new technologies in regulated environments, combining quantum computing with generative AI, strategies for scaling quantum efforts in biopharma, and much more.

THAT’S A WRAP.