RPI and Kipu Quantum Ran Autism Biomarker Selection on IBM's 120-Qubit Nighthawk Processors: What the 24 September 2026 arXiv Preprint Found
Quentir Medicine Monitor
Evidence-based insights for quantum medicine.
A blood sample from a young child can yield hundreds of metabolite measurements, while a typical autism biomarker study enrolls a few dozen children. Choosing which handful of those measurements a diagnostic model should rely on is a hard combinatorial search. A team at Rensselaer Polytechnic Institute in Troy, New York, working with the Berlin company Kipu Quantum, posted a paper on arXiv on 24 September 2026 that runs this search on IBM quantum processors.
The paper, "Quantum Feature Selection for Biomedical Data Analysis" by Hongbin Liu, Robert Lahmann, Benjamin Campbell, Zhemin Zhang, Zhiding Liang, Siona Bapat and Juergen Hahn, poses metabolomic feature selection as a binary optimization problem and solves it on two IBM Nighthawk processors, ibm_rensselaer and ibm_miami, each with 120 qubits. The test data come from three published studies, two of blood plasma and one of the gut microbiome, in children with autism spectrum disorder and typically developing peers. The quantum solver picked five-feature subsets that classified about as well as several standard classical methods, and the authors state plainly that their results do not claim a quantum advantage.
For clinicians and hospital laboratories the interest lies in the diagnostic problem underneath. Blood-based autism screening has been studied at least since 2012, when the oldest of the three datasets used here was published, and every such test depends on a short, stable list of markers. This preprint tests whether current quantum hardware can help build that list. It does not test a diagnostic, and it enrolled no new patients.
The work is a first hardware feasibility study on retrospective data. It has not yet been peer reviewed, and its datasets are small: the largest group has 83 children with an autism diagnosis and 76 without.
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.
AOI Biosciences' Roughly 500 Million Yen Financing and Share Transaction of 14 September 2026 Lists Its Neo-Self Antibody Test Ahead of the Toshiba SQBM+ Quantum-Inspired Drug Platform
Quentir Medicine Monitor
Evidence-based insights for quantum medicine.
On 14 September 2026 AOI Biosciences, a diagnostics and drug-discovery company in Ibaraki, Osaka Prefecture, announced a financing and secondary-share transaction of roughly 500 million yen with existing and new investors; the company says the new-share portion takes its cumulative funding to about 1 billion yen, and it does not state how much of the 500 million is new money. The release describes two core businesses. One is a blood test for a "neo-self" autoantibody that is already sold to more than 300 Japanese medical institutions. The other is an allosteric drug-discovery method built with Toshiba on the quantum-inspired SQBM+ solver, the method whose patent Toshiba announced on 3 September and this Monitor examined on 6 September.
The release lists three planned uses for the proceeds, and that list is the useful fact for a hospital buyer or a pharma partner, with one caution: the order says nothing about how much money goes where or which business the company values more. The neo-self antibody test fills the first two entries, with domestic roll-out, Asian expansion and new autoimmune indications, and further research investment in the quantum-inspired drug discovery platform is the third. No amounts are attached to any entry. The clinical figures the release cites for the test, a pregnancy rate more than doubled and a live-birth rate 1.7 times higher, come from two Frontiers in Immunology cohort studies whose treated and untreated groups can be counted, and this note counts them.
Nanodiamond Charge Drift Reads Macrophage Inflammation: Chicago and Iowa in Advanced Materials, 4 February 2026
Nanodiamond quantum sensors have been read as thermometers inside living cells for more than a decade. One of them, a 70-nanometer crystal sitting inside a mouse macrophage, drifted downward by 0.27 megahertz over 200 seconds of measurement, which on the usual arithmetic means the cell warmed by 3.62 degrees Celsius. A team at the University of Chicago and the University of Iowa has now put a different quantity behind that same number.
Their paper, Probing cellular activity via charge-sensitive quantum nanoprobes, appeared in Advanced Materials on 4 February 2026, with Uri Zvi as first author and Denis R. Candido at Iowa, Aaron Esser-Kahn and Peter C. Maurer at Chicago as corresponding authors. It was received on 15 March 2025 and accepted on 15 January 2026. The University of Iowa publicized it on 26 August 2026, six months after the article went online, which is why it reaches this Monitor now.
What the zero-field splitting actually tracks, and the dipole term that was being dropped
The sensor is a nitrogen-vacancy center, an atomic defect in diamond whose spin can be initialized with green light, driven with microwaves and read out optically. Each 70-nanometer crystal used here carries roughly 100 of them. The measured quantity is the zero-field splitting, the frequency gap between the spin's ground sublevels, which sits near 2.87 gigahertz and moves by about 74 kilohertz for every kelvin. That temperature coefficient is what turned these particles into intracellular thermometers in the first place.