University of Florida and MD Anderson Radiologists Review Quantum Machine Learning for MRI, Chest X-Ray and CT: What the arXiv Primer of 23 October 2026 Finds
What radiologists at UF Jacksonville and MD Anderson found about quantum machine learning in imaging
Quentir Medicine Monitor
Evidence-based insights for quantum medicine.
Five radiologists from the University of Florida College of Medicine in Jacksonville and MD Anderson Cancer Center, writing with Harshada Gothankar of Johns Hopkins University, have produced a guide to quantum machine learning for practicing radiologists. The 15-page primer appeared on arXiv on 23 October 2026, with Mayur Virarkar of UF Health Jacksonville as corresponding author.
The authors start from a problem every imaging department knows. Deep learning models for volumetric MRI and CT need computing power and memory that grow quickly with image size, and large architectures such as ResNet, DenseNet and vision transformers are hard to deploy where hardware is limited. Quantum computing has been proposed as one answer, and the primer asks how far that answer has come in medical imaging.
Nearly all published systems are hybrid quantum-classical pipelines tested on small, downsampled benchmark datasets or on simulators, and the authors write that clinical deployment is not yet feasible. A radiologist reading the paper gets a map of a young research field, with its strongest claims and its weakest tests set side by side.
The paper is a narrative primer. The authors searched PubMed, arXiv and academic search engines through August 2026, prioritized peer-reviewed original studies, and state that they verified all citations against publisher records. They report no specific funding and no relevant conflicts of interest.
The first half teaches five concepts: qubits, superposition, entanglement, quantum gates and quantum circuits. The authors stress a point that vendor material often skips. An n-qubit register can hold 2n basis states at once, yet a measurement returns a single classical outcome, so any practical speedup depends on algorithm design, data encoding and measurement strategy.
Today's machines belong to the noisy intermediate-scale era, with tens to a few hundred qubits and no full error correction. The primer describes three platforms. Superconducting processors dominate medical imaging research and are reachable through IBM's cloud; trapped-ion systems hold their quantum states longer at smaller scale; photonic chips run at room temperature but remain hard to control and appear rarely in imaging studies.
Dong-A University Opens Its G-LAMP Atomic and Quantum Research Center in Busan on 20 October 2026: KRW 25 Billion for Quantum Computing, Medical Image Analysis and Radioisotope Drugs
What Dong-A University opened in Busan on 20 October 2026
Quentir Medicine Monitor
Evidence-based insights for quantum medicine.
Dong-A University in Busan opened the offices of its new G-LAMP research group on 20 October 2026, with about 100 guests on its Seunghak Campus. The group has roughly KRW 25 billion in national funding to build a research platform on atomic and quantum technology that reaches from basic science into diagnosis and treatment.
The university announced the opening on 21 October, and Korean outlets including Money Today reported it the same day. G-LAMP is a basic science research hub program run by Korea's Ministry of Education and the National Research Foundation of Korea. Dong-A was selected as a new group in 2025, and the October ceremony marks the start of its full research activity. President Lee Hae-woo, project director Jeong Jin-woong, Busan's deputy mayor for administration Kim Kyung-deok and Heo Jeong-eun of the National Research Foundation were among those present.
The research plan, as the university describes it, has a clear medical end. Its listed priorities include quantum computing and information processing applied to medical image analysis, the synthesis of new quantum materials, research using radioisotopes, and the development of diagnostic and therapeutic substances with clinical application. Jeong said the group would build an integrated platform covering everything from basic technology to clinical application, and Lee spoke of combining atomic and quantum science with advanced medical technology to open a new field.
For a reader in a hospital or a pharmaceutical company, the opening is an early signal. It funds people, laboratories and a direction of work in Korea's southeast. The reports of the opening cite no study, no device, no measured outcome and no timetable for clinical work. What it does show is where a Korean university chose to point about KRW 25 billion of public money for quantum research: toward imaging, radiopharmaceuticals and the clinic.
The reports also leave some details uneven. The venue is given as the arts and sports college hall in some accounts and as the natural sciences college lobby in others, and the length of the funding period is reported in two ways. The substance of the plan is the same in every account. The differences matter for anyone who cites the program in a funding comparison.
Kraków's J-PET Group and 21 Co-Authors Publish a Roadmap for Quantum Entanglement PET: What arXiv 2607.12182 and the June 2026 First-in-Human Scan Show
Quentir Medicine Monitor
Evidence-based insights for quantum medicine.
Every PET scan starts with a positron meeting an electron inside the patient. The two annihilate and send out a pair of gamma photons in opposite directions, and the scanner finds the tracer by catching both photons at once. Those photon pairs also carry a quantum property that no clinical scanner reads today: their polarizations are entangled. A group of physicists and imaging engineers now argues that this unused signal could sharpen PET images and add new diagnostic readings.
Their case is set out in "Roadmap Towards Quantum Entanglement Positron Emission Tomography (QE-PET)" (arXiv 2607.12182), posted on 13 July 2026 by 22 authors led by Paweł Moskal of the Jagiellonian University in Kraków, with teams from the University of California Santa Cruz, IIT Bombay, the University of Tokyo, the University of Sydney, the University of Zagreb, Japan's National Institutes for Quantum Science and Technology and others. The roadmap surveys how quantum entanglement PET could be built, from the plastic-scintillator J-PET scanner in Kraków to crystal, semiconductor and Compton-camera designs. It also describes a new candidate biomarker, the degree of quantum entanglement of the photon pair, which may depend on the tissue in which the positron annihilated.
Two experimental findings from the past three years made the idea worth pursuing. In 2023 an experiment showed that the quantum correlation between the two photons survives when one of them scatters off an electron, which most physicists had expected to destroy it. In 2025 the Kraków group reported in Science Advances that photons from annihilation in matter are less than maximally entangled, and that the degree of entanglement depends on how the annihilation happened. If that dependence tracks the chemistry of tissue, such as oxygen content, a scanner could read it as a new measurement alongside the usual tracer uptake.
A companion preprint makes the idea concrete. On 28 June 2026 the Kraków team and colleagues posted "First-in-human quantum entanglement imaging" (arXiv 2606.29421), reporting that the J-PET scanner produced, in one patient, both an image of a gallium-68 DOTA-TATE tracer and, from the same research acquisition, an image of the degree of entanglement. The roadmap reads that scan as a start. The authors state plainly that whether entanglement-based imaging and pH mapping can reach clinical practice "remains an open question."
Florence and Barcelona Researchers Run a Quantum Diffusion Model's Noise Step on IBM's 133-Qubit Torino for Blood-Cell, Brain MRI and Rib CT Images: What arXiv 2609.31070 Shows
Quentir Medicine Monitor
Evidence-based insights for quantum medicine.
Generative AI can now draw convincing medical images: blood-cell smears, brain scans, slices of a CT volume. Hospitals and research groups want such synthetic images to enlarge small training sets, to share data without exposing patients, and to test diagnostic software on rare cases. The most successful generators today are diffusion models, which learn to turn random noise back into a plausible image. A preprint posted on 25 September 2026 asks whether a real quantum computer can take over one half of that process.
The paper, "Quantum Diffusion Models for Medical Image Analysis" (arXiv 2609.31070), comes from Francesco Aldo Venturelli and Miguel A. González Ballester of Universitat Pompeu Fabra and the Barcelona Supercomputing Center, Stefano Martina, Marco Parigi and Filippo Caruso of the University of Florence, and Alba Cervera-Lierta of the Barcelona Supercomputing Center. They built a hybrid quantum diffusion model in which the step that adds noise to an image runs on IBM's 133-qubit ibm_torino processor, while a classical neural network learns to remove that noise again. They then used it to generate synthetic medical images from three public collections: blood-cell photographs, brain MRI slices and 3D rib CT volumes.
The authors report that their hybrid model is competitive with a classical diffusion model of the same design, and better on some measures, especially for the 3D rib volumes. The comparison is limited: 100 generated samples per dataset, small images, and a classical baseline the team reproduced itself. No radiologist judged the images, and no diagnostic task was run on them.
That leaves a narrow but real result. It is one of the few medical-imaging studies in quantum machine learning that runs part of its pipeline on physical quantum hardware with public medical-image datasets, and it deserves a precise reading of what the hardware did and what the numbers show.
QTomo Says Quantum Computing Can Rebuild CT Scans From 10 Percent of the X-Ray Data: What Kyungtaek Jun's Papers From 2023 to 2026 Tested
Quentir Medicine Monitor
Evidence-based insights for quantum medicine.
A Korean start-up called QTomo says its quantum algorithms can rebuild a CT scan from one tenth of the usual X-ray projections, which would cut a patient's radiation dose by 90 percent. The company's founder, Kyungtaek Jun, made those claims in a newspaper interview published on 24 September 2026, and his own peer-reviewed papers show how far the method has actually been tested.
Jun has published the core of QTomo's approach in Scientific Reports in 2023 and 2025, in two arXiv preprints in April and May 2025 and in EPJ Quantum Technology in July 2026. The experiments in those papers perform quantum CT reconstruction on images of at most 100 by 100 pixels, solve the problem on the D-Wave hybrid solver, which combines a quantum annealer with classical computing, and use test objects such as the Shepp-Logan phantom, a standard computer-drawn head outline, a model tooth or a single existing chest CT slice. The 90 percent figure appears as an estimate in the April 2025 preprint, and none of the papers reports a scan acquired from a patient for the study, a dose measurement or a clinical validation.
That gap matters for radiology departments and hospital buyers, because the interview lists early cancer detection, breast and heart imaging and dental work among the technology's proposed clinical applications. The published papers describe a careful and interesting research program at a much earlier stage.
Set side by side, the interview's figures and the papers' measurements separate into three groups: claims with peer-reviewed support on small test images, one estimate from a preprint, and figures that appear in no paper at all. The first group shows steady progress by a small research team. The second would need a scanner study to confirm it, and the third would need data that QTomo has not yet made public.
Simulated Quantum Kernels Lost All 24 Matched Tests to a Default Classical SVM on Brain MRI and Breast Ultrasound: Journal of Imaging Informatics in Medicine, 21 September 2026
Quentir Medicine Monitor
Evidence-based insights for quantum medicine.
On 21 September 2026 the Journal of Imaging Informatics in Medicine published a study by four computer scientists in Lahore, Abu Dhabi and Manchester that put two quantum machine-learning classifiers against ordinary statistical tools on brain MRI scans and breast ultrasound images. In all 24 matched comparisons the top-scoring quantum model scored lower than the top-scoring classical model, and simulating the quantum models on ordinary computers took roughly 50 to more than 220 times as long as running the classical one.
The paper, by M. Usman Hashmi, M. Adnan Hashmi, Muazzam Ali and Raheem Sarwar, asks whether quantum kernels improve medical image classification once every model gets the same features, the same training budget and the same protection against data leakage. Its answer, for the settings it tested, is no: a standard RBF-SVM with its default settings outperformed both quantum kernels in every condition. The authors describe their design as a leakage-controlled benchmark, and that design is the reason the study deserves a close reading.
Many earlier papers on quantum classifiers for radiology and pathology reported encouraging accuracy figures. Several compared a tuned quantum model with a weak classical one, or tested on a single split of the data, where a small overlap between training and test images can inflate a score. This study sets out to give both sides the same conditions and then measure what is left.
For a radiology department or a hospital buyer, the result gives a practical reference point. When a vendor or research partner proposes a quantum classifier for imaging, the first question is whether it was compared with a strong, well-configured classical baseline under the same conditions. This paper shows how large the difference can be when that comparison is made carefully. It also lists the controls such a comparison needs: identical inputs, preprocessing fitted on training images only, several random seeds and reported calibration.
What the Word Quantum Means on a Hospital CT Scanner
The label describes a detector
Siemens Healthineers has introduced a second generation of its photon-counting CT family in Vietnam. The NAEOTOM Alpha Class is a commercial imaging platform for hospitals, with versions aimed at routine imaging, cardiology, and tertiary or specialist centers. The word quantum points to the detector's handling of discrete X-ray photons. No qubits run a computation inside the gantry. Tiền Phong reports that all three models offer 0.2 millimeter ultra-high-resolution slices and multi-energy data. Alpha.Pro and Alpha.Peak reach temporal resolution down to 66 milliseconds, while Alpha.Peak reaches a reported scan speed of up to 737 millimeters per second.
One platform now serves three clinical settings
The product segmentation turns a physics story into a hospital procurement question. A routine radiology department, a cardiac service, and a tertiary referral center carry different case mixes, motion problems, staffing needs, and service expectations. The same detector principle can therefore produce different value across institutions. Configuration-specific comparisons remain essential: image quality for the intended examinations, radiation dose, contrast use, reconstruction performance, throughput, training, uptime, and the clinical decisions influenced by spectral information. The launch also included SOMATOM On.site, a separate mobile head-and-neck CT system. Its presence makes the distinction useful. One device changes the detector architecture; the other changes where imaging reaches a critically ill patient.
The technology is already in clinical use
Siemens says the first NAEOTOM Alpha entered clinical use in 2021. The company describes its detector as directly converting X-rays into electrical signals while measuring each photon's energy, making spectral information available with every scan. That places the underlying sensing technology at TRL 9 on the shared readiness ladder. Commercial maturity still leaves a local buying question. Hospitals need to establish which model and protocols fit their patients, clinicians, facilities, and budgets. Precise language helps: photon-counting CT is an advanced diagnostic-imaging technology grounded in quantum physics, and its performance belongs in ordinary clinical and operational comparisons.
Does a Quantum Layer Change What a Medical AI Sees?
One layer makes the comparison unusually clean
A July 2026 preprint compares two medical-image classifiers that share the same convolutional backbone and a comparable number of trainable parameters. They differ in one intermediate layer. One branch uses a dense classical layer; the other uses a four-qubit circuit emulated on a conventional computer. This parameter-matched comparison helps isolate the contribution of the circuit-shaped representation without changing the rest of the network.
The relative gain appeared between small and large datasets
The models classified retinal optical coherence tomography images, with training sets ranging from 200 to 30,000 examples. The hybrid model performed best relative to the classical comparator around 800 and 2,000 training images, an intermediate-data regime. At the largest sizes, the classical CNN moved ahead and reached the study's highest retinal test accuracy, 93.7 percent. A robustness check using two-dimensional slices from an OASIS-1 dementia MRI dataset followed the same broad pattern: the hybrid advantage narrowed with more data, then reversed slightly.
The attention maps changed with sample size
The researchers also compared SHAP attention maps, which estimate where image pixels influence a model's output. With 1,000 retinal images, the hybrid maps appeared more concentrated around retinal layers and disease-related structures, while the classical maps were more fragmented in the examples shown. At 30,000 images, the models' highlighted regions overlapped much more. These maps do not establish clinically validated biomarkers. The study includes no physical quantum-hardware run, prospective clinical test, external hospital validation, clinician reader study, or patient outcome. Its contribution is narrower and useful: a controlled account of when a small classically simulated circuit changed model performance and apparent visual focus, and when additional data favored the matched classical layer.