arXiv subjects

Hanyuan Zhang

Publications and source records attributed to Hanyuan Zhang.

At least 19 recordsLinked to original sources

arXiv2026

Bar-induced migration of $ω$ Centauri away from Gaia Sausage-Enceladus

The globular cluster $ω$ Cen has been suggested to have originated in the Gaia Sausage-Enceladus (GSE) merger event, possibly as its nuclear star cluster. However, the present-day orbits of $ω$ Cen and the GSE debris are very different. We investigate the scenario in which $ω$ Cen originated in the GSE and migrated to its current position due to perturbations from the Galactic bar. The [$α$/M] distributions of stars located between the GSE debris and $ω$ Cen in $(L_z,E)$ space tentatively support this scenario, but are not conclusive. We run simulations of the GSE debris and $ω$ Cen in a realistic Milky Way potential with a decelerating bar at various present-day pattern speeds. We find that $ω$ Cen can indeed be traced back to the phase space region occupied by the GSE debris. However, this would likely require a pattern speed of $Ω_\mathrm{b}\lesssim26$ km s$^{-1}$ kpc$^{-1}$, which is much lower than most recent estimates. We conclude that a GSE origin for $ω$ Cen is dynamically and chemically plausible, but only with a re-evaluation of the current consensus on the bar's pattern speed.

astro-ph.GA↗

arXiv2026

A Unified Framework for Joint Detection of Lacunes and Enlarged Perivascular Spaces

Cerebral small vessel disease (CSVD) markers, specifically enlarged perivascular spaces (EPVS) and lacunae, present a unique challenge in medical image analysis due to their radiological mimicry. Standard segmentation networks struggle with feature interference and extreme class imbalance when handling these divergent targets simultaneously. To address these issues, we propose a morphology-decoupled framework where Zero-Initialized Gated Cross-Task Attention exploits dense EPVS context to guide sparse lacune detection. Furthermore, biological and topological consistency are enforced via a mixed-supervision strategy integrating Mutual Exclusion and Centerline Dice losses. Finally, we introduce an Anatomically-Informed Inference Calibration mechanism to dynamically suppress false positives based on tissue semantics. Extensive 5-folds cross-validation on the VALDO 2021 dataset (N=40) demonstrates state-of-the-art performance, notably surpassing task winners in lacunae detection precision (71.1%, p=0.01) and F1-score (62.6%, p=0.03). Furthermore, evaluation on the external EPAD cohort (N=1762) confirms the model's robustness for large-scale population studies. Code will be released upon acceptance.

cs.CV↗

arXiv2026

CT Open: An Open-Access, Uncontaminated, Live Platform for the Open Challenge of Clinical Trial Outcome Prediction

Scientists have long sought to accurately predict outcomes of real-world events before they happen. Can AI systems do so more reliably? We study this question through clinical trial outcome prediction, a high-stakes open challenge even for domain experts. We introduce CT Open, an open-access, live platform that will run four challenge every year. Anyone can submit predictions for each challenge. CT Open evaluates those submissions on trials whose outcomes were not yet public at the time of submission but were made public afterwards. Determining if a trial's outcome is public on the internet before a certain date is surprisingly difficult. Outcomes posted on official registries may lag behind by years, while the first mention may appear in obscure articles. To address this, we propose a novel, fully automated decontamination pipeline that uses iterative LLM-powered web search to identify the earliest mention of trial outcomes. We validate the pipeline's quality and accuracy by human expert's annotations. Since CT Open's pipeline ensures that every evaluated trial had no publicly reported outcome when the prediction was made, it allows participants to use any methodology and any data source. In this paper, we release a training set and two time-stamped test benchmarks, Winter 2025 and Summer 2025. We believe CT Open can serve as a central hub for advancing AI research on forecasting real-world outcomes before they occur, while also informing biomedical research and improving clinical trial design. CT Open Platform is hosted at $\href{https://ct-open.net/}{https://ct-open.net/}$

cs.AI↗

arXiv2026

DeepImagine: Clinical Trial Outcome Prediction via Stepwise Local Counterfactual Imaginations

Predicting the outcomes of prospective clinical trials remains a major challenge. Clinical trial outcomes result from complex interactions among experimental factors such as drug interventions, participant demographics, and protocols. Here, we introduce DeepImagine, a framework that predicts target trial outcomes through stepwise counterfactual imagination anchored on historical trials with observed results. Starting from a relevant historical trial, DeepImagine sequentially modifies one differing experimental factor at a time. With each step a large language model (LLM) is posed a local counterfactual: how would the current imagined outcome change with this single perturbation? The updated result is carried forward as the input to the next step, until the historical configuration exactly matches the target, yielding the final prediction. Empirically, DeepImagine consistently outperforms direct one-step prediction across several off-the-shelf LLMs, with further gains when multiple imagination pathways, initiated from different historical anchors, are aggregated. We also construct natural counterfactuals augmented with synthetic reasoning traces and train a family of specialized language models, each dedicated to learning one factor's local counterfactual transition. Integrating these learned local operators into DeepImagine yields substantial improvements over general-purpose LLM baselines. Our findings position stepwise counterfactual imagination, distinct from both correlational prediction and explicit structural causal modeling, as a promising direction for clinical trial outcome prediction. All code, training scripts, and evaluation scripts are available at \href{https://github.com/deepimagine-counterfactual/DeepImagine}{https://github.com/deepimagine-counterfactual/DeepImagine}.

cs.CL↗

arXiv2026

The disrupted chemical enrichment history of the Milky Way driven by gas accretion

As the only galaxy enabling temporally-resolved observations from an internal vantage point, the Milky Way as a galaxy offers unique insights into galactic chemical enrichment history, establishment of fundamental scaling relations, and the underlying astrophysical processes. However, this insider perspective also introduces strong selection effects, hindering direct measurement of the Milky Way's global properties and comparison with the broader galaxy population, for the vast majority of which only integrated properties can be measured. {Here we report our measurements of the Milky Way's temporally-resolved galaxy-scale average metallicity using data from the APOGEE survey after correction for the selection function. Our findings unveil a present-day metallicity of the Milky Way close to the Sun, an interrupted integrated age-metallicity relation, and a disturbed evolutionary trajectory in the mass-metallicity diagram, likely caused by dilution and inside-out growth associated with external gas accretion around 7~Gyr ago. Our results highlight the critical role of gas accretion in disrupting the galactic enrichment histories and introducing scatter in mass-metallicity relations.

astro-ph.GA↗

arXiv2026

CenSynCMB: Centre Maps and Physics-Guided Synthesis for Microbleed Detection

Cerebral microbleeds (CMBs) are MRI markers of small vessel disease and the microbleed component of amyloid related imaging abnormalities (ARIA-H), but their small size, sparsity, and similarity to vessels, calcification-like foci, and artefacts make automated detection difficult. We propose CenSynCMB, a centre-guided and mimic-aware framework combining a 3D Attention U-Net, auxiliary centre-map supervision, false-negative-driven reweighting, and fold-wise physics-guided synthesis of positive CMBs and labelled hard negatives. Synthetic data expose the detector to compact lesions and common mimics without validation or test leakage. On VALDO Task 2, CenSynCMB achieved the best local-comparison lesion-level F1 (74.3%, p = 0.020); on external AIBL SWI, it achieved the highest local-comparison recall (88.5%, p = 0.0058) and F1 (65.0%, p = 0.0016). Together, these results support scalable CMB candidate extraction in large, unlabelled MRI cohorts, while highlighting cohort-specific calibration as the next step toward reliable burden estimation.

cs.CV↗

arXiv2026

Depth Augmented and FE Free 3D/2D Liver Registration for Laparoscopic Liver AR

Augmented reality (AR) guidance in laparoscopic liver surgery requires accurate registration of preoperative 3D models to intraoperative 2D video, but remains challenging due to partial visibility, specularities, and tissue deformation. Existing methods often rely on contour-based rigid initialization and finite-element (FE) models for deformable registration, increasing modeling and engineering complexity. We present a depth-augmented, FE-free 3D--2D registration pipeline that combines robust rigid initialization with patient-specific non-rigid refinement. For rigid alignment, we adapt the RefineNet module of FoundationPose to laparoscopic liver scenes by using multi-class contour maps and monocular depth for relative pose refinement. For deformable alignment, we construct a patient-specific statistical deformation model from non-rigid ICP (NICP) correspondences and optimize pose and shape parameters using a coarse-to-fine L-BFGS-B strategy. On a public clinical laparoscopic liver dataset, the proposed method achieves a mean target registration error (TRE) of 14.73\,mm under a controlled manual-contour setting designed to isolate registration performance. Ablation studies show that monocular depth improves rigid initialization over contour-only inputs, while tumor-mapping analysis indicates that good surface alignment does not necessarily translate into lower target localization error. On an external dataset without ground truth, the method produces visually plausible overlays for qualitative assessment. These results suggest that depth-augmented pose refinement and FE-free statistical deformation modeling provide a promising alternative to FE-based pipelines for controlled 3D--2D liver registration in surgical AR.

cs.CV↗

arXiv2026

Warm-Started Reinforcement Learning for Iterative 3D/2D Liver Registration

Registration between preoperative CT and intraoperative laparoscopic video plays a crucial role in augmented reality (AR) guidance for minimally invasive surgery. Learning-based methods have recently achieved registration errors comparable to optimization-based approaches while offering faster inference. However, many supervised methods produce coarse alignments that rely on additional optimization-based refinement, thereby increasing inference time. We present a discrete-action reinforcement learning (RL) framework that formulates CT-to-video registration as a sequential decision-making process. A shared feature encoder, warm-started from a supervised pose estimation network to provide stable geometric features and faster convergence, extracts representations from CT renderings and laparoscopic frames, while an RL policy head learns to choose rigid transformations along six degrees of freedom and to decide when to stop the iteration. Experiments on a public laparoscopic dataset demonstrated that our method achieved an average target registration error (TRE) of 15.70 mm, comparable to supervised approaches with optimization, while achieving faster convergence. The proposed RL-based formulation enables automated, efficient iterative registration without manually tuned step sizes or stopping criteria. This discrete framework provides a practical foundation for future continuous-action and deformable registration models in surgical AR applications.

cs.CV↗

arXiv2026

Maximizing T2-Only Prostate Cancer Localization from Expected Diffusion Weighted Imaging

Multiparametric MRI is increasingly recommended as a first-line noninvasive approach to detect and localize prostate cancer, requiring at minimum diffusion-weighted (DWI) and T2-weighted (T2w) MR sequences. Early machine learning attempts using only T2w images have shown promising diagnostic performance in segmenting radiologist-annotated lesions. Such uni-modal T2-only approaches deliver substantial clinical benefits by reducing costs and expertise required to acquire other sequences. This work investigates an arguably more challenging application using only T2w at inference, but to localize individual cancers based on independent histopathology labels. We formulate DWI images as a latent modality (readily available during training) to classify cancer presence at local Barzell zones, given only T2w images as input. In the resulting expectation-maximization algorithm, a latent modality generator (implemented using a flow matching-based generative model) approximates the latent DWI image posterior distribution in the E-steps, while in M-steps a cancer localizer is simultaneously optimized with the generative model to maximize the expected likelihood of cancer presence. The proposed approach provides a novel theoretical framework for learning from a privileged DWI modality, yielding superior cancer localization performance compared to approaches that lack training DWI images or existing frameworks for privileged learning and incomplete modalities. The proposed T2-only methods perform competitively or better than baseline methods using multiple input sequences (e.g., improving the patient-level F1 score by 14.4\% and zone-level QWK by 5.3\% over the T2w+DWI baseline). We present quantitative evaluations using internal and external datasets from 4,133 prostate cancer patients with histopathology-verified labels.

cs.CV↗

arXiv2025

Subsampled Randomized Fourier GaLore for Adapting Foundation Models in Depth-Driven Liver Landmark Segmentation

Accurate detection and delineation of anatomical structures in medical imaging are critical for computer-assisted interventions, particularly in laparoscopic liver surgery where 2D video streams limit depth perception and complicate landmark localization. While recent works have leveraged monocular depth cues for enhanced landmark detection, challenges remain in fusing RGB and depth features and in efficiently adapting large-scale vision models to surgical domains. We propose a depth-guided liver landmark segmentation framework integrating semantic and geometric cues via vision foundation encoders. We employ Segment Anything Model V2 (SAM2) encoder to extract RGB features and Depth Anything V2 (DA2) encoder to extract depth-aware features. To efficiently adapt SAM2, we introduce SRFT-GaLore, a novel low-rank gradient projection method that replaces the computationally expensive SVD with a Subsampled Randomized Fourier Transform (SRFT). This enables efficient fine-tuning of high-dimensional attention layers without sacrificing representational power. A cross-attention fusion module further integrates RGB and depth cues. To assess cross-dataset generalization, we also construct a new Laparoscopic Liver Surgical Dataset (LLSD) as an external validation benchmark. On the public L3D dataset, our method achieves a 4.85% improvement in Dice Similarity Coefficient and a 11.78-point reduction in Average Symmetric Surface Distance compared to the D2GPLand. To further assess generalization capability, we evaluate our model on LLSD dataset. Our model maintains competitive performance and significantly outperforms SAM-based baselines, demonstrating strong cross-dataset robustness and adaptability to unseen surgical environments. These results demonstrate that our SRFT-GaLore-enhanced dual-encoder framework enables scalable and precise segmentation under real-time, depth-constrained surgical settings.

cs.CV↗

arXiv2025

Dynamical streams in the local stellar halo

Co-moving groups of stars (streams) are well known in the velocity space of the disc near the Sun. Many are thought to arise from resonances with the Galactic bar or spiral arms. In this work, we search for similar moving groups in the velocity space of the halo, at low angular momentum. From the asymmetry of the radial velocity distribution $v_R$, we identify two inward-moving streams with $v_R<0$ and small $|v_ϕ|$. These are projections of the `chevrons' previously discovered in radial phase space $(R,v_R)$. A test particle simulation in a realistic Milky Way potential with a decelerating bar naturally produces analogues of these features, and they are observed across a wide range of metallicity. They are therefore very likely to be dynamical streams created by trapping in the bar's resonances. Specifically, they occupy regions of phase space where orbits are trapped in the corotation and outer Lindblad resonances respectively. By tracing these streams across a range of radii in $(R,v_R)$ space, we fit resonant orbits to their tracks in a flexible potential with variable bar pattern speed. This allows us to simultaneously constrain the mass profile of the Milky Way for $r\lesssim20$ kpc and the pattern speed $Ω_\mathrm{b}$. We estimate the mass enclosed within $r=20$ kpc to be $M_{20}=(2.17\pm0.21)\times10^{11}M_\odot$, and the pattern speed to be $Ω_\mathrm{b}=31.9_{-1.9}^{+1.8}$ km/s/kpc. Our fitted potential is in excellent agreement with previous results, while we favour a slightly slower pattern speed than most recent estimates.

astro-ph.GA↗

arXiv2025

The ones that got away: chemical tagging of globular cluster-origin stars with Gaia BP/RP spectra

Globular clusters (GCs) are sites of extremely efficient star formation, and recent studies suggest they significantly contributed to the early Milky Way's stellar mass build-up. Although their role has since diminished, GCs' impact on the Galaxy's initial evolution can be traced today by identifying their most chemically unique stars--those with anomalous nitrogen and aluminum overabundances and oxygen depletion. While they are a perfect tracer of clusters, be it intact or fully dissolved, these high-[N/O], high-[Al/Fe] GC-origin stars are extremely rare within the current Galaxy. To address the scarcity of these unusual, precious former GC members, we train a neural network (NN) to identify high-[N/O], high-[Al/Fe] stars using low-resolution Gaia BP/RP spectra. Our NN achieves a classification accuracy of approximately $\approx99\%$ and a false positive rate of around $\approx7\%$, identifying 878 new candidates in the Galactic field. We validate our results with several physically-motivated sanity checks, showing, for example, that the incidence of selected stars in Galactic GCs is significantly higher than in the field. Moreover, we find that most of our GC-origin candidates reside in the inner Galaxy, having likely formed in the proto-Milky Way, consistent with previous research. The fraction of GC candidates in the field drops at a metallicity of [Fe/H]$\approx-1$, approximately coinciding with the completion of spin-up, i.e. the formation of the Galactic stellar disk.

astro-ph.GA↗

arXiv2024

Kinematics and dynamics of the Galactic bar revealed by Gaia long-period variables

We use low-amplitude, long period variable (LA-LPV) candidates in \textit{Gaia} DR3 to trace the kinematics and dynamics of the Milky Way bar. LA-LPVs, like other LPVs, are intrinsically bright and follow a tight period-luminosity relation, but unlike e.g. Mira variables, their radial velocity measurements are reliable due to their smaller pulsation amplitudes. We supplement the \textit{Gaia} astrometric and radial velocity measurements with distance moduli assigned using a period-luminosity relation to acquire full 6D phase space information. The assigned distances are validated by comparing to geometric distances and StarHorse distances, which shows biases less than $\sim5\%$. Our sample provides an unprecedented panoramic picture of the inner Galaxy with minimal selection effects. We map the kinematics of the inner Milky Way and find a significant kinematic signature corresponding to the Galactic bar. We measure the pattern speed of the Galactic bar using the continuity equation and find $Ω_{\rm b}=34.1\pm2.4$ km s$^{-1}$ kpc$^{-1}$. We develop a simple, robust and potential-independent method to measure the dynamical length of the bar using only kinematics and find $R_{\rm b}\sim4.0$ kpc. We validate both measurements using N-body simulations. Assuming knowledge of the gravitational potential of the inner Milky Way, we analyse the orbital structure of the Galactic bar using orbital frequency ratios. The $x_1$ orbits are the dominant bar-supporting orbital family in our sample. Amongst the selected bar stars, the $x_1 v_1$ or "banana" orbits constitute a larger fraction ($\sim 15\%$) than other orbital families in the bar, implying that they are the dominant family contributing to the Galactic X-shape, although contributions from other orbital families are also present.

astro-ph.GA↗

arXiv2024

Deciphering the Milky Way disc formation time encrypted in the bar chrono-kinematics

We present a novel method to constrain the formation time of the Milky Way disc using the chrono-kinematic signatures of the inner Galaxy. We construct an O-rich Mira variable sample from the Gaia Long-period Variable catalogue to study the kinematic behaviour of stars with different ages in the inner Galaxy. From the Auriga suite of cosmological zoom-in simulations, we find that the age of the oldest stellar population with imprints of the bar in density and kinematics matches the disc spin-up epoch. This is because stars born before the spin-up show insufficient rotation and are not kinematically cold enough to be efficiently trapped by the bar. We find that the bar kinematic signature disappears for Mira variables with a period shorter than 190 days. Using the period-age relation of Mira variables, we constrain the spin-up epoch of the Milky Way to be younger than $\sim11-12$~Gyr (redshift $\sim3$). We also discuss and compare our method and result to other evidence of the Milky Way spin-up epoch under the context of a realistic age uncertainty. Age uncertainty leads to an overestimation of the disc formation time when performing backward modelling. Our constrain of the spin-up epoch is independent from previous studies because it relies on the kinematics of the inner Galaxy instead of the solar vicinity.

astro-ph.GA↗

arXiv2024

The Pristine survey -- XXIII. Data Release 1 and an all-sky metallicity catalogue based on Gaia DR3 BP/RP spectro-photometry

We used the spectro-photometric information of ~219 million stars from Gaia's DR3 to calculate synthetic, narrow-band, metallicity-sensitive CaHK magnitudes that mimic the observations of the Pristine survey, a survey of photometric metallicities of Milky Way (MW) stars that has been mapping >6,500 deg^2 of the northern sky with CFHT since 2015. These synthetic magnitudes were used for an absolute recalibration of the deeper Pristine photometry and, combined with broadband Gaia information, synthetic and Pristine CaHK magnitudes were used to estimate photometric metallicities over the whole sky. The resulting metallicity catalogue is accurate down to [Fe/H]~-3.5 and is particularly suited for the exploration of the metal-poor MW ([Fe/H]<-1.0). We make available here the catalogue of synthetic CaHK_syn magnitudes for all stars with BP/RP information in Gaia DR3, as well as an associated catalogue of more than ~30 million photometric metallicities for high S/N FGK stars. This paper further provides the first public data release of the Pristine catalogue in the form of higher quality recalibrated Pristine CaHK magnitudes and photometric metallicities for all stars in common with the BP/RP spectro-photometric information in Gaia DR3. When available, the much deeper Pristine data greatly enhance the quality of the derived metallicities, in particular at the faint end of the catalogue (G_BP > 16). Combined, both photometric metallicity catalogues include more than two million metal-poor star candidates ([Fe/H]_phot<-1.0) as well as more than 200,000 and ~8,000 very and extremely metal-poor candidates ([Fe/H]_phot<-2.0 and <-3.0, respectively). Finally, we show that these metallicity catalogues can be used efficiently, among other applications to hunt for the most metal-poor stars, and to study how the structure of the MW varies with metallicity.

astro-ph.GA↗

arXiv2024

On the existence of a very metal-poor disc in the Milky Way

There has been a discussion for many years on whether the disc in the Milky Way extends down to low metallicity. We aim to address the question by employing a large sample of giant stars with radial velocities and homogeneous metallicities based on the Gaia DR3 XP spectra. We study the 3D velocity distribution of stars in various metallicity ranges, including the very-metal poor regime (VMP, [M/H] $<-2.0$). We find that a clear disc population starts to emerge only around [M/H] $\sim -1.3$, and is not visible for [M/H] $<-1.6$. Using Gaussian Mixture Modeling (GMM), we show that there are two halo populations in the VMP regime: one stationary and one with a net prograde rotation of $\sim80\,\mathrm{km/s}$. In this low-metallicity range, we are able to place constraints on the contribution of a rotation-supported disc sub-population to a maximum of $\sim 3$\%. We compare our results to previous claims of discy VMP stars in both observations and simulations and find that having a prograde halo component could explain most of these.

astro-ph.GA↗

arXiv2024

YNetr: Dual-Encoder architecture on Plain Scan Liver Tumors (PSLT)

Background: Liver tumors are abnormal growths in the liver that can be either benign or malignant, with liver cancer being a significant health concern worldwide. However, there is no dataset for plain scan segmentation of liver tumors, nor any related algorithms. To fill this gap, we propose Plain Scan Liver Tumors(PSLT) and YNetr. Methods: A collection of 40 liver tumor plain scan segmentation datasets was assembled and annotated. Concurrently, we utilized Dice coefficient as the metric for assessing the segmentation outcomes produced by YNetr, having advantage of capturing different frequency information. Results: The YNetr model achieved a Dice coefficient of 62.63% on the PSLT dataset, surpassing the other publicly available model by an accuracy margin of 1.22%. Comparative evaluations were conducted against a range of models including UNet 3+, XNet, UNetr, Swin UNetr, Trans-BTS, COTr, nnUNetv2 (2D), nnUNetv2 (3D fullres), MedNext (2D) and MedNext(3D fullres). Conclusions: We not only proposed a dataset named PSLT(Plain Scan Liver Tumors), but also explored a structure called YNetr that utilizes wavelet transform to extract different frequency information, which having the SOTA in PSLT by experiments.

eess.IV↗

arXiv2024

Can We Trust LLMs? Mitigate Overconfidence Bias in LLMs through Knowledge Transfer

The study explores mitigating overconfidence bias in LLMs to improve their reliability. We introduce a knowledge transfer (KT) method utilizing chain of thoughts, where "big" LLMs impart knowledge to "small" LLMs via detailed, sequential reasoning paths. This method uses advanced reasoning of larger models to fine-tune smaller models, enabling them to produce more accurate predictions with calibrated confidence. Experimental evaluation using multiple-choice questions and sentiment analysis across diverse datasets demonstrated the KT method's superiority over the vanilla and question-answer pair (QA) fine-tuning methods. The most significant improvement in three key metrics, where the KT method outperformed the vanilla and QA methods by an average of 55.3% and 43.1%, respectively. These findings underscore the KT method's potential in enhancing model trustworthiness and accuracy, offering precise outputs with well-matched confidence levels across various contexts.

cs.CL↗