Searcharxiv⌕ Search

arXiv subjects

L. Bello

Publications and source records attributed to L. Bello.

2 recordsLinked to original sources

Pump quantization changes optimal measurements for quantum sensing

Standard analyses of quantum phase sensing often neglect the overhead of generating squeezed-light by treating the pump as an undepleted classical resource. Realistically, the pump is a quantum object that undergoes significant depletion and entangles with the produced squeezed light, raising the question of how the crossover from a classical to a quantum pump limits the quantum advantage in sensing. Contrary to prior analyses, we show that quantum advantage persists despite pump quantization, but depends critically on the nonlinear $(χ^{(2)})$ source dynamics. In particular, we demonstrate that Heisenberg scaling emerges in the regime of strong pump depletion, rather than maximum signal squeezing. At maximum squeezing, the quantum Fisher information (QFI) for interferometric sensing of an unknown phase scales as $N^{3/2}$, where $N$ is the pump photon number, surpassing the shot-noise limit but establishing that squeezing itself is insufficient to attain Heisenberg scaling of $N^2$. At maximal pump depletion, where the signal flux is maximized, the QFI scales as $N^2$. In this regime, standard photon counting is suboptimal, while a joint nonlinear measurement of the entire pump-signal system is shown to achieve Heisenberg scaling. Our work shows that pump quantization changes both the optimal operating point and the measurement strategies required to preserve a realistic quantum advantage in phase estimation and, more broadly, many-body quantum dynamics.

quant-ph↗

Preoperative brain tumor imaging: models and software for segmentation and standardized reporting

For patients suffering from brain tumor, prognosis estimation and treatment decisions are made by a multidisciplinary team based on a set of preoperative MR scans. Currently, the lack of standardized and automatic methods for tumor detection and generation of clinical reports represents a major hurdle. In this study, we investigate glioblastomas, lower grade gliomas, meningiomas, and metastases, through four cohorts of up to 4000 patients. Tumor segmentation models were trained using the AGU-Net architecture with different preprocessing steps and protocols. Segmentation performances were assessed in-depth using a wide-range of voxel and patient-wise metrics covering volume, distance, and probabilistic aspects. Finally, two software solutions have been developed, enabling an easy use of the trained models and standardized generation of clinical reports: Raidionics and Raidionics-Slicer. Segmentation performances were quite homogeneous across the four different brain tumor types, with an average true positive Dice ranging between 80% and 90%, patient-wise recall between 88% and 98%, and patient-wise precision around 95%. With our Raidionics software, running on a desktop computer with CPU support, tumor segmentation can be performed in 16 to 54 seconds depending on the dimensions of the MRI volume. For the generation of a standardized clinical report, including the tumor segmentation and features computation, 5 to 15 minutes are necessary. All trained models have been made open-access together with the source code for both software solutions and validation metrics computation. In the future, an automatic classification of the brain tumor type would be necessary to replace manual user input. Finally, the inclusion of post-operative segmentation in both software solutions will be key for generating complete post-operative standardized clinical reports.

eess.IV↗