arXiv subjects

Jingjing Peng

Publications and source records attributed to Jingjing Peng.

7 recordsLinked to original sources

arXiv2026

Stabilizing Temporal Inference Dynamics for Online Surgical Phase Recognition

Online Surgical Phase Recognition (SPR) models can reach high frame-wise accuracy, yet their predictions often lack temporal stability, fragmenting workflow understanding and reducing the reliability of downstream assistance. We show that this instability is not random noise but arises from two mechanisms: early misclassifications corrupt temporal feature states and propagate forward to form error cascades, and phase transitions follow evidence-accumulation dynamics whereas most online SPR systems rely on memoryless frame-wise decisions, making them sensitive to transient confidence fluctuations. We propose a unified Train-Inference-Evaluation framework that explicitly stabilizes temporal inference dynamics using model-agnostic, plug-and-play components. For training, the Temporal Error-Cascade (TEC) loss suppresses error onset and mitigates forward error propagation by stabilizing temporal feature evolution. For inference, the Evidence-Gated Transition Predictor (EGTP) enforces evidence-driven state transitions, allowing phase changes only when accumulated evidence exceeds a confidence boundary. For evaluation, we introduce the Temporal Fragmentation Index (TFI), a reliability-aware metric that quantifies instability-induced temporal disagreement beyond conventional frame-wise and token-based measures. Experiments on Cholec80 and AutoLaparo across three representative backbones show that the proposed framework substantially improves temporal stability and reduces prediction fragmentation, while maintaining or modestly improving frame-wise performance.

cs.CV↗

arXiv2026

From Pre- to Intra-operative MRI: Predicting Brain Shift in Temporal Lobe Resection for Epilepsy Surgery

Introduction: In neurosurgery, image-guided Neurosurgery Systems (IGNS) highly rely on preoperative brain magnetic resonance images (MRI) to assist surgeons in locating surgical targets and determining surgical paths. However, brain shift invalidates the preoperative MRI after dural opening. Updated intraoperative brain MRI with brain shift compensation is crucial for enhancing the precision of neuronavigation systems and ensuring the optimal outcome of surgical interventions. Methodology: We propose NeuralShift, a U-Net-based model that predicts brain shift entirely from pre-operative MRI for patients undergoing temporal lobe resection. We evaluated our results using Target Registration Errors (TREs) computed on anatomical landmarks located on the resection side and along the midline, and DICE scores comparing predicted intraoperative masks with masks derived from intraoperative MRI. Results: Our experimental results show that our model can predict the global deformation of the brain (DICE of 0.97) with accurate local displacements (achieve landmark TRE as low as 1.12 mm), compensating for large brain shifts during temporal lobe removal neurosurgery. Conclusion: Our proposed model is capable of predicting the global deformation of the brain during temporal lobe resection using only preoperative images, providing potential opportunities to the surgical team to increase safety and efficiency of neurosurgery and better outcomes to patients. Our contributions will be publicly available after acceptance in https://github.com/SurgicalDataScienceKCL/NeuralShift.

cs.CV↗

arXiv2024

DDSB: An Unsupervised and Training-free Method for Phase Detection in Echocardiography

Accurate identification of End-Diastolic (ED) and End-Systolic (ES) frames is key for cardiac function assessment through echocardiography. However, traditional methods face several limitations: they require extensive amounts of data, extensive annotations by medical experts, significant training resources, and often lack robustness. Addressing these challenges, we proposed an unsupervised and training-free method, our novel approach leverages unsupervised segmentation to enhance fault tolerance against segmentation inaccuracies. By identifying anchor points and analyzing directional deformation, we effectively reduce dependence on the accuracy of initial segmentation images and enhance fault tolerance, all while improving robustness. Tested on Echo-dynamic and CAMUS datasets, our method achieves comparable accuracy to learning-based models without their associated drawbacks. The code is available at https://github.com/MRUIL/DDSB

cs.CV↗

arXiv2017

Photon-gated spin transistor

Spin-polarized field-effect transistor (spin-FET), where a dielectric layer is generally employed for the electrical gating as the traditional FET, stands out as a seminal spintronic device under the miniaturization trend of electronics. It would be fundamentally transformative if optical gating was used for spin-FET. We report a new type of spin-polarized field-effect transistor (spin-FET) with optical gating, which is fabricated by partial exposure of the (La,Sr)MnO3 channel to light-emitting diode (LED) light. The manipulation of the channel conductivity is ascribed to the enhanced scattering of the spin-polarized current by photon-excited antiparallel aligned spins. And the photon-gated spin-FET shows strong light power dependence and reproducible enhancement of resistance under light illumination, indicting well-defined conductivity cycling features. Our finding would enrich the concept of spin-FET and promote the use of optical means in spintronics for low power consumption and ultrafast data processing.

cond-mat.mtrl-sci↗

arXiv2015

Electrical control of magnetism in oxides

This review article aims at illustrating the recent progresses in the electrical control of magnetism in oxides with profound physics and enormous potential applications. In the first part, we provide a comprehensive summary of the electrical control of magnetism in the classic multiferroic heterostructures and clarify their various mechanisms lying behind. The second part focuses on the novel route of electric double layer gating for driving a significantly electronic phase transition in magnetic oxides by a small voltage. The electric field applied on the ordinary dielectric oxide in the third part is used to control the magnetic phenomenon originated from the charge transfer and orbital reconstruction at the interface between dissimilar correlated oxides. At last, we analyze the challenges in electrical control of magnetism in oxides, both on mechanism and practical application, which would inspire more in-depth researches and advance the development in this field.

cond-mat.mtrl-sci↗

arXiv2015

Magnetoelectric coupling induced by interfacial orbital reconstruction

The magnetoelectric coupling effect with profound physics and enormous potential applications has provoked a great number of research activities in materials science. Here, we report that the reversible orbital reconstruction driven by ferroelectric polarization modulates the magnetic performance of ferroelectric ferromagnetic heterostructure. Mn in plane orbital occupancy and related interfacial exotic magnetic state are enhanced and weakened by the negative and positive electric field, respectively. Our findings thus not only present a broad opportunity to fill the missing member, orbital in the mechanism of magnetoelectric coupling, but also make the orbital degree of freedom straight forward to the application in microelectronic device.

cond-mat.str-el↗

arXiv2014

Electrical manipulation of orbital occupancy and magnetic anisotropy in manganites

Electrical manipulation of lattice, charge, and spin has been realized respectively by the piezoelectric effect, field-effect transistor, and electric field control of ferromagnetism, bringing about dramatic promotions both in fundamental research and industrial production. However, it is generally accepted that the orbital of materials are impossible to be altered once they have been made. Here we use electric-field to dynamically tune the electronic phase transition in (La,Sr)MnO3 films with different Mn^4+/(Mn^3+ + Mn^4+) ratios. The orbital occupancy and corresponding magnetic anisotropy of these thin films are manipulated by gate voltage in a reversible and quantitative manner. Positive gate voltage increases the proportion of occupancy of the orbital and magnetic anisotropy that were initially favored by strain (irrespective of tensile and compressive), while negative gate voltage reduces the concomitant preferential orbital occupancy and magnetic anisotropy. Besides its fundamental significance in orbital physics, our findings might advance the process towards practical oxide-electronics based on orbital.

cond-mat.mtrl-sci↗