arXiv subjects

Changhee Lee

Publications and source records attributed to Changhee Lee.

At least 37 records · Page 2Linked to original sources

arXiv2025

Inversion-asymmetric itinerant antiferromagnets by the space group symmetry

We investigate the appearance of an inversion-asymmetric antiferromagnetism due to an itinerant mechanism in nonsymmorphic systems with magnetic ions at Wyckoff position of multiplicity 2. The key symmetries which underpin the existence of such phases are established, and we derive a Landau free energy from a general microscopic electronic Hamiltonian. Our analysis reveals that the stable antiferromagnetic order is largely determined by the symmetries of Wyckoff position, the nature of the nesting between electronic bands, and the presence of anisotropy or nesting in high-symmetry planes of the Brillouin zone. We illustrate our conclusions with specific microscopic models.

cond-mat.str-el↗

arXiv2024

BioBridge: Unified Bio-Embedding with Bridging Modality in Code-Switched EMR

Pediatric Emergency Department (PED) overcrowding presents a significant global challenge, prompting the need for efficient solutions. This paper introduces the BioBridge framework, a novel approach that applies Natural Language Processing (NLP) to Electronic Medical Records (EMRs) in written free-text form to enhance decision-making in PED. In non-English speaking countries, such as South Korea, EMR data is often written in a Code-Switching (CS) format that mixes the native language with English, with most code-switched English words having clinical significance. The BioBridge framework consists of two core modules: "bridging modality in context" and "unified bio-embedding." The "bridging modality in context" module improves the contextual understanding of bilingual and code-switched EMRs. In the "unified bio-embedding" module, the knowledge of the model trained in the medical domain is injected into the encoder-based model to bridge the gap between the medical and general domains. Experimental results demonstrate that the proposed BioBridge significantly performance traditional machine learning and pre-trained encoder-based models on several metrics, including F1 score, area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), and Brier score. Specifically, BioBridge-XLM achieved enhancements of 0.85% in F1 score, 0.75% in AUROC, and 0.76% in AUPRC, along with a notable 3.04% decrease in the Brier score, demonstrating marked improvements in accuracy, reliability, and prediction calibration over the baseline XLM model. The source code will be made publicly available.

cs.CL↗

arXiv2024

Toward a Well-Calibrated Discrimination via Survival Outcome-Aware Contrastive Learning

Previous deep learning approaches for survival analysis have primarily relied on ranking losses to improve discrimination performance, which often comes at the expense of calibration performance. To address such an issue, we propose a novel contrastive learning approach specifically designed to enhance discrimination \textit{without} sacrificing calibration. Our method employs weighted sampling within a contrastive learning framework, assigning lower penalties to samples with similar survival outcomes. This aligns well with the assumption that patients with similar event times share similar clinical statuses. Consequently, when augmented with the commonly used negative log-likelihood loss, our approach significantly improves discrimination performance without directly manipulating the model outputs, thereby achieving better calibration. Experiments on multiple real-world clinical datasets demonstrate that our method outperforms state-of-the-art deep survival models in both discrimination and calibration. Through comprehensive ablation studies, we further validate the effectiveness of our approach through quantitative and qualitative analyses.

cs.LG↗

arXiv2024

Automatic Quantification of Serial PET/CT Images for Pediatric Hodgkin Lymphoma Patients Using a Longitudinally-Aware Segmentation Network

$\textbf{Purpose}$: Automatic quantification of longitudinal changes in PET scans for lymphoma patients has proven challenging, as residual disease in interim-therapy scans is often subtle and difficult to detect. Our goal was to develop a longitudinally-aware segmentation network (LAS-Net) that can quantify serial PET/CT images for pediatric Hodgkin lymphoma patients. $\textbf{Materials and Methods}$: This retrospective study included baseline (PET1) and interim (PET2) PET/CT images from 297 patients enrolled in two Children's Oncology Group clinical trials (AHOD1331 and AHOD0831). LAS-Net incorporates longitudinal cross-attention, allowing relevant features from PET1 to inform the analysis of PET2. Model performance was evaluated using Dice coefficients for PET1 and detection F1 scores for PET2. Additionally, we extracted and compared quantitative PET metrics, including metabolic tumor volume (MTV) and total lesion glycolysis (TLG) in PET1, as well as qPET and $Δ$SUVmax in PET2, against physician measurements. We quantified their agreement using Spearman's $ρ$ correlations and employed bootstrap resampling for statistical analysis. $\textbf{Results}$: LAS-Net detected residual lymphoma in PET2 with an F1 score of 0.606 (precision/recall: 0.615/0.600), outperforming all comparator methods (P<0.01). For baseline segmentation, LAS-Net achieved a mean Dice score of 0.772. In PET quantification, LAS-Net's measurements of qPET, $Δ$SUVmax, MTV and TLG were strongly correlated with physician measurements, with Spearman's $ρ$ of 0.78, 0.80, 0.93 and 0.96, respectively. The performance remained high, with a slight decrease, in an external testing cohort. $\textbf{Conclusion}$: LAS-Net demonstrated significant improvements in quantifying PET metrics across serial scans, highlighting the value of longitudinal awareness in evaluating multi-time-point imaging datasets.

cs.CV↗

arXiv2024

DSG-KD: Knowledge Distillation from Domain-Specific to General Language Models

The use of pre-trained language models fine-tuned to address specific downstream tasks is a common approach in natural language processing (NLP). However, acquiring domain-specific knowledge via fine-tuning is challenging. Traditional methods involve pretraining language models using vast amounts of domain-specific data before fine-tuning for particular tasks. This study investigates emergency/non-emergency classification tasks based on electronic medical record (EMR) data obtained from pediatric emergency departments (PEDs) in Korea. Our findings reveal that existing domain-specific pre-trained language models underperform compared to general language models in handling N-lingual free-text data characteristics of non-English-speaking regions. To address these limitations, we propose a domain knowledge transfer methodology that leverages knowledge distillation to infuse general language models with domain-specific knowledge via fine-tuning. This study demonstrates the effective transfer of specialized knowledge between models by defining a general language model as the student model and a domain-specific pre-trained model as the teacher model. In particular, we address the complexities of EMR data obtained from PEDs in non-English-speaking regions, such as Korea, and demonstrate that the proposed method enhances classification performance in such contexts. The proposed methodology not only outperforms baseline models on Korean PED EMR data, but also promises broader applicability in various professional and technical domains. In future works, we intend to extend this methodology to include diverse non-English-speaking regions and address additional downstream tasks, with the aim of developing advanced model architectures using state-of-the-art KD techniques. The code is available in https://github.com/JoSangYeon/DSG-KD.

cs.CL↗

arXiv2024

Enhancing Anomaly Detection via Generating Diversified and Hard-to-distinguish Synthetic Anomalies

Unsupervised anomaly detection is a daunting task, as it relies solely on normality patterns from the training data to identify unseen anomalies during testing. Recent approaches have focused on leveraging domain-specific transformations or perturbations to generate synthetic anomalies from normal samples. The objective here is to acquire insights into normality patterns by learning to differentiate between normal samples and these crafted anomalies. However, these approaches often encounter limitations when domain-specific transformations are not well-specified such as in tabular data, or when it becomes trivial to distinguish between them. To address these issues, we introduce a novel domain-agnostic method that employs a set of conditional perturbators and a discriminator. The perturbators are trained to generate input-dependent perturbations, which are subsequently utilized to construct synthetic anomalies, and the discriminator is trained to distinguish normal samples from them. We ensure that the generated anomalies are both diverse and hard to distinguish through two key strategies: i) directing perturbations to be orthogonal to each other and ii) constraining perturbations to remain in proximity to normal samples. Throughout experiments on real-world datasets, we demonstrate the superiority of our method over state-of-the-art benchmarks, which is evident not only in image data but also in tabular data, where domain-specific transformation is not readily accessible. Additionally, we empirically confirm the adaptability of our method to semi-supervised settings, demonstrating its capacity to incorporate supervised signals to enhance anomaly detection performance even further.

cs.LG↗

arXiv2024

Unified picture of superconductivity and magnetism in CeRh$_2$As$_2$

We study the micorscopic origin of the multiple superconducting and magnetic phases observed in CeRh$_2$As$_2$. We exploit the existence of a van Hove singularity enforced by the nonsymmorphic symmetry to conduct a renormalization group analysis. When Fermi-surface nesting is strong, we find two closely-competing superconducting states with opposite parities, as well as an instability towards specific spin-density wave states, consistent with key features of the phase diagram of CeRh$_2$As$_2$.

cond-mat.supr-con↗

arXiv2023

Domain-adapted large language models for classifying nuclear medicine reports

With the growing use of transformer-based language models in medicine, it is unclear how well these models generalize to nuclear medicine which has domain-specific vocabulary and unique reporting styles. In this study, we evaluated the value of domain adaptation in nuclear medicine by adapting language models for the purpose of 5-point Deauville score prediction based on clinical 18F-fluorodeoxyglucose (FDG) PET/CT reports. We retrospectively retrieved 4542 text reports and 1664 images for FDG PET/CT lymphoma exams from 2008-2018 in our clinical imaging database. Deauville scores were removed from the reports and then the remaining text in the reports was used as the model input. Multiple general-purpose transformer language models were used to classify the reports into Deauville scores 1-5. We then adapted the models to the nuclear medicine domain using masked language modeling and assessed its impact on classification performance. The language models were compared against vision models, a multimodal vision language model, and a nuclear medicine physician with seven-fold Monte Carlo cross validation, reported are the mean and standard deviations. Domain adaption improved all language models. For example, BERT improved from 61.3% five-class accuracy to 65.7% following domain adaptation. The best performing model (domain-adapted RoBERTa) achieved a five-class accuracy of 77.4%, which was better than the physician's performance (66%), the best vision model's performance (48.1), and was similar to the multimodal model's performance (77.2). Domain adaptation improved the performance of large language models in interpreting nuclear medicine text reports.

cs.CL↗

arXiv2023

T-Phenotype: Discovering Phenotypes of Predictive Temporal Patterns in Disease Progression

Clustering time-series data in healthcare is crucial for clinical phenotyping to understand patients' disease progression patterns and to design treatment guidelines tailored to homogeneous patient subgroups. While rich temporal dynamics enable the discovery of potential clusters beyond static correlations, two major challenges remain outstanding: i) discovery of predictive patterns from many potential temporal correlations in the multi-variate time-series data and ii) association of individual temporal patterns to the target label distribution that best characterizes the underlying clinical progression. To address such challenges, we develop a novel temporal clustering method, T-Phenotype, to discover phenotypes of predictive temporal patterns from labeled time-series data. We introduce an efficient representation learning approach in frequency domain that can encode variable-length, irregularly-sampled time-series into a unified representation space, which is then applied to identify various temporal patterns that potentially contribute to the target label using a new notion of path-based similarity. Throughout the experiments on synthetic and real-world datasets, we show that T-Phenotype achieves the best phenotype discovery performance over all the evaluated baselines. We further demonstrate the utility of T-Phenotype by uncovering clinically meaningful patient subgroups characterized by unique temporal patterns.

cs.LG↗

arXiv2022

Linear optical response from the odd parity Bardasis-Schrieffer mode in locally non-centrosymmetric superconductors

On the recent report of a magnetic field induced first order transition between an even-parity superconductivity and an odd-parity superconductivity in $\mathrm{CeRh_{2}As_{2}}$, the microscopic physics is still under investigation. However, if, in the vicinity of this transition, the coupling strengths of the even and odd pairing channels are comparable, a particle-particle excitonic collective mode referred to as the Bardasis-Schrieffer (BS) mode should generically exist below the pair-breaking continuum. This BS mode can couple to the light and thus affect the optical response of the superconductor, as it arises from a pairing channel with the parity opposite to that of the ground state pairs. Here, by using a generic bilayer model Hamiltonian for the electronic degree of freedom, which is globally centrosymmetric despite each layer being locally non-centrosymmetric, we study the change of the excitation gap of the BS mode with respect to the out-of-plane magnetic fields and demonstrate that its coupling to the light is possible even in the linear response regime. The linear coupling is attributed to the presence of multiple electronic bands, which is a generic feature of a bilayer system. Our result shows the microwave absorption as the signature of the BS mode, and hence a smoking gun signature of the parity-switching at the transition between two superconducting phases.

cond-mat.supr-con↗

arXiv2022

SurvITE: Learning Heterogeneous Treatment Effects from Time-to-Event Data

We study the problem of inferring heterogeneous treatment effects from time-to-event data. While both the related problems of (i) estimating treatment effects for binary or continuous outcomes and (ii) predicting survival outcomes have been well studied in the recent machine learning literature, their combination -- albeit of high practical relevance -- has received considerably less attention. With the ultimate goal of reliably estimating the effects of treatments on instantaneous risk and survival probabilities, we focus on the problem of learning (discrete-time) treatment-specific conditional hazard functions. We find that unique challenges arise in this context due to a variety of covariate shift issues that go beyond a mere combination of well-studied confounding and censoring biases. We theoretically analyse their effects by adapting recent generalization bounds from domain adaptation and treatment effect estimation to our setting and discuss implications for model design. We use the resulting insights to propose a novel deep learning method for treatment-specific hazard estimation based on balancing representations. We investigate performance across a range of experimental settings and empirically confirm that our method outperforms baselines by addressing covariate shifts from various sources.

cs.LG↗

arXiv2022

Topological multiband s-wave superconductivity in coupled multifold fermions

We study three-dimensional time-reversal-invariant topological superconductivity in noncentrosymmetric materials such as RhSi, CoSi, and AlPt which host coupled multifold nodes energetically split by the spin-orbit coupling at the same time-reversal-invariant momentum (TRIM). The topological superconductivity arises from the $s_{+} \oplus s_{-}$ gap function, which is $\boldsymbol{k}$ independent, but with opposite signs for the two nodes split at the same TRIM. We consider various electron-electron interactions in the tight-binding model for RhSi and find that the topological superconducting phase supporting a surface Majorana cone and topological nodal rings is favored in a wide range of interaction parameters.

cond-mat.supr-con↗

arXiv2021

A Variational Information Bottleneck Approach to Multi-Omics Data Integration

Integration of data from multiple omics techniques is becoming increasingly important in biomedical research. Due to non-uniformity and technical limitations in omics platforms, such integrative analyses on multiple omics, which we refer to as views, involve learning from incomplete observations with various view-missing patterns. This is challenging because i) complex interactions within and across observed views need to be properly addressed for optimal predictive power and ii) observations with various view-missing patterns need to be flexibly integrated. To address such challenges, we propose a deep variational information bottleneck (IB) approach for incomplete multi-view observations. Our method applies the IB framework on marginal and joint representations of the observed views to focus on intra-view and inter-view interactions that are relevant for the target. Most importantly, by modeling the joint representations as a product of marginal representations, we can efficiently learn from observed views with various view-missing patterns. Experiments on real-world datasets show that our method consistently achieves gain from data integration and outperforms state-of-the-art benchmarks.

cs.LG↗

arXiv2021

Non-slice 3-stranded pretzel knots

Greene-Jabuka and Lecuona confirmed the slice-ribbon conjecture for 3-stranded pretzel knots except for an infinite family $P(a,-a-2,-\frac{(a+1)^2}{2})$ where $a$ is an odd integer greater than $1$. Lecuona and Miller showed that $P(a,-a-2,-\frac{(a+1)^2}{2})$ are not slice unless $a\equiv 1, 11, 37, 47, 59 \pmod{60}$. In this note, we show that four-fifths of the remaining knots in the family are not slice.

math.GT↗

arXiv2020

Temporal Phenotyping using Deep Predictive Clustering of Disease Progression

Due to the wider availability of modern electronic health records, patient care data is often being stored in the form of time-series. Clustering such time-series data is crucial for patient phenotyping, anticipating patients' prognoses by identifying "similar" patients, and designing treatment guidelines that are tailored to homogeneous patient subgroups. In this paper, we develop a deep learning approach for clustering time-series data, where each cluster comprises patients who share similar future outcomes of interest (e.g., adverse events, the onset of comorbidities). To encourage each cluster to have homogeneous future outcomes, the clustering is carried out by learning discrete representations that best describe the future outcome distribution based on novel loss functions. Experiments on two real-world datasets show that our model achieves superior clustering performance over state-of-the-art benchmarks and identifies meaningful clusters that can be translated into actionable information for clinical decision-making.

physics.med-ph↗

arXiv2019

Emergent Anisotropic Non-Fermi Liquid at a Topological Phase Transition in Three Dimensions

Understanding correlation effects in topological phases and their transitions is a cutting-edge area of research in recent condensed matter physics. We study topological quantum phase transitions (TQPTs) between double-Weyl semimetals (DWSMs) and insulators, and argue that a novel class of quantum criticality appears at the TQPT characterized by emergent anisotropic non-Fermi liquid behaviors, in which the interplay between the Coulomb interaction and electronic critical modes induces not only anisotropic renormalization of the Coulomb interaction but also strongly correlated electronic excitation in three spatial dimensions. Using the standard renormalization group methods, large $N_f$ theory and the $ε= 4-d$ method with fermion flavor number $N_f$ and spatial dimension $d$, we obtain the anomalous dimensions of electrons ($η_f=0.366/N_f $) in large $N_f$ theory and the associated anisotropic scaling relations of various physical observables. Our results may be observed in candidate materials for DWSMs such as HgCr$_2$Se$_4$ or SrSi$_2$ when the system undergoes a TQPT.

cond-mat.str-el↗

arXiv2018

Estimation of Individual Treatment Effect in Latent Confounder Models via Adversarial Learning

Estimating the individual treatment effect (ITE) from observational data is essential in medicine. A central challenge in estimating the ITE is handling confounders, which are factors that affect both an intervention and its outcome. Most previous work relies on the unconfoundedness assumption, which posits that all the confounders are measured in the observational data. However, if there are unmeasurable (latent) confounders, then confounding bias is introduced. Fortunately, noisy proxies for the latent confounders are often available and can be used to make an unbiased estimate of the ITE. In this paper, we develop a novel adversarial learning framework to make unbiased estimates of the ITE using noisy proxies.

cs.LG↗

arXiv2017

Phase transitions of polariton condensate in 2D Dirac materials

For the quantum well in an optical microcavity, the interplay of the Coulomb interaction and the electron-photon (e-ph) coupling can lead to the hybridizations of the exciton and the cavity photon known as polaritons, which can form the Bose-Einstein condensate above a threshold density. Additional physics due to the nontrivial Berry phase comes into play when the quantum well consists of the gapped two-dimensional (2D) Dirac material such as the transition metal dichalcogenide (TMDC) MoS$_2$ or WSe$_2$. Specifically, in forming the polariton, the e-ph coupling from the optical selection rule due to the Berry phase can compete against the Coulomb electron-electron (e-e) interaction. We find that this competition gives rise to a rich phase diagram for the polariton condensate involving both topological and symmetry breaking phase transitions, with the former giving rise to the quantum anomalous Hall and the quantum spin Hall phases.

cond-mat.str-el↗