arXiv subjects

Jin Sung Kim

Publications and source records attributed to Jin Sung Kim.

At least 19 recordsLinked to original sources

arXiv2026

Mixture of Multicenter Experts in Multimodal AI for Debiased Radiotherapy Target Delineation

Clinical decision-making reflects diverse strategies shaped by regional patient populations and institutional protocols. However, most existing medical artificial intelligence (AI) models are trained on highly prevalent data patterns, which reinforces biases and fails to capture the breadth of clinical expertise. Inspired by the recent advances in Mixture of Experts (MoE), we propose a Mixture of Multicenter Experts (MoME) framework to address AI bias in the medical domain without requiring data sharing across institutions. MoME integrates specialized expertise from diverse clinical strategies to enhance model generalizability and adaptability across medical centers. We validate this framework using a multimodal target volume delineation model for prostate cancer radiotherapy. With few-shot training that combines imaging and clinical notes from each center, the model outperformed baselines, particularly in settings with high inter-center variability or limited data availability. Furthermore, MoME enables model customization to local clinical preferences without cross-institutional data exchange, making it especially suitable for resource-constrained settings while promoting broadly generalizable medical AI.

eess.IV↗

arXiv2026

A Guideline-Aware AI Agent for Zero-Shot Target Volume Auto-Delineation

Delineating the clinical target volume (CTV) in radiotherapy involves complex margins constrained by tumor location and anatomical barriers. While deep learning models automate this process, their rigid reliance on expert-annotated data requires costly retraining whenever clinical guidelines update. To overcome this limitation, we introduce OncoAgent, a novel guideline-aware AI agent framework that seamlessly converts textual clinical guidelines into three-dimensional target contours in a training-free manner. Evaluated on esophageal cancer cases, the agent achieves a zero-shot Dice similarity coefficient of 0.842 for the CTV and 0.880 for the planning target volume, demonstrating performance highly comparable to a fully supervised nnU-Net baseline. Notably, in a blinded clinical evaluation, physicians strongly preferred OncoAgent over the supervised baseline, rating it higher in guideline compliance, modification effort, and clinical acceptability. Furthermore, the framework generalizes zero-shot to alternative esophageal guidelines and other anatomical sites (e.g., prostate) without any retraining. Beyond mere volumetric overlap, our agent-based paradigm offers near-instantaneous adaptability to alternative guidelines, providing a scalable and transparent pathway toward interpretability in radiotherapy treatment planning.

cs.CV↗

arXiv2025

Parallel-plate chambers as radiation-hard detectors for time-based beam diagnostics in carbon-ion radiotherapy

Accurate range verification of carbon ion beams is critical for the precision and safety of charged particle radiotherapy. In this study, we evaluated the feasibility of using a parallel-plate ionization chamber for real-time, time-based diagnostic monitoring of carbon ion beams. The chamber featured a 0.4 mm gas gap defined by metallic electrodes and was filled with carbon dioxide (CO$_2$), a non-polymerizing gas suitable for high-rate applications. Timing precision was assessed via self-correlation analysis, yielding a precision approaching one picosecond for one-second acquisitions under clinically relevant beam conditions. This level of timing accuracy translates to a water-equivalent range uncertainty of approximately 1 mm, which meets the recommended clinical tolerance for carbon ion therapy. Furthermore, the kinetic energy of the beam at the synchrotron extraction point was determined from the measured orbital period, with results consistently within 1 MeV/nucleon of the nominal energy. These findings demonstrate the potential of parallel-plate chambers for precise, real-time energy and range verification in clinical carbon ion beam quality assurance.

physics.med-ph↗

arXiv2025

End-to-End Breast Cancer Radiotherapy Planning via LMMs with Consistency Embedding

Recent advances in AI foundation models have significant potential for lightening the clinical workload by mimicking the comprehensive and multi-faceted approaches used by medical professionals. In the field of radiation oncology, the integration of multiple modalities holds great importance, so the opportunity of foundational model is abundant. Inspired by this, here we present RO-LMM, a multi-purpose, comprehensive large multimodal model (LMM) tailored for the field of radiation oncology. This model effectively manages a series of tasks within the clinical workflow, including clinical context summarization, radiation treatment plan suggestion, and plan-guided target volume segmentation by leveraging the capabilities of LMM. In particular, to perform consecutive clinical tasks without error accumulation, we present a novel Consistency Embedding Fine-Tuning (CEFTune) technique, which boosts LMM's robustness to noisy inputs while preserving the consistency of handling clean inputs. We further extend this concept to LMM-driven segmentation framework, leading to a novel Consistency Embedding Segmentation (CESEG) techniques. Experimental results including multi-centre validation confirm that our RO-LMM with CEFTune and CESEG results in promising performance for multiple clinical tasks with generalization capabilities.

cs.CV↗

arXiv2025

Distribution-aware Fairness Learning in Medical Image Segmentation From A Control-Theoretic Perspective

Ensuring fairness in medical image segmentation is critical due to biases in imbalanced clinical data acquisition caused by demographic attributes (e.g., age, sex, race) and clinical factors (e.g., disease severity). To address these challenges, we introduce Distribution-aware Mixture of Experts (dMoE), inspired by optimal control theory. We provide a comprehensive analysis of its underlying mechanisms and clarify dMoE's role in adapting to heterogeneous distributions in medical image segmentation. Furthermore, we integrate dMoE into multiple network architectures, demonstrating its broad applicability across diverse medical image analysis tasks. By incorporating demographic and clinical factors, dMoE achieves state-of-the-art performance on two 2D benchmark datasets and a 3D in-house dataset. Our results highlight the effectiveness of dMoE in mitigating biases from imbalanced distributions, offering a promising approach to bridging control theory and medical image segmentation within fairness learning paradigms. The source code will be made available. The source code is available at https://github.com/tvseg/dMoE.

eess.IV↗

arXiv2025

Susceptibility of Large Language Models to User-Driven Factors in Medical Queries

Large language models (LLMs) are increasingly used in healthcare, but their reliability is heavily influenced by user-driven factors such as question phrasing and the completeness of clinical information. In this study, we examined how misinformation framing, source authority, model persona, and omission of key clinical details affect the diagnostic accuracy and reliability of LLM outputs. We conducted two experiments: one introducing misleading external opinions with varying assertiveness (perturbation test), and another removing specific categories of patient information (ablation test). Using public datasets (MedQA and Medbullets), we evaluated proprietary models (GPT-4o, Claude 3.5 Sonnet, Claude 3.5 Haiku, Gemini 1.5 Pro, Gemini 1.5 Flash) and open-source models (LLaMA 3 8B, LLaMA 3 Med42 8B, DeepSeek R1 8B). All models were vulnerable to user-driven misinformation, with proprietary models especially affected by definitive and authoritative language. Assertive tone had the greatest negative impact on accuracy. In the ablation test, omitting physical exam findings and lab results caused the most significant performance drop. Although proprietary models had higher baseline accuracy, their performance declined sharply under misinformation. These results highlight the need for well-structured prompts and complete clinical context. Users should avoid authoritative framing of misinformation and provide full clinical details, especially for complex cases.

cs.CL↗

arXiv2025

Multi-Class Segmentation of Aortic Branches and Zones in Computed Tomography Angiography: The AortaSeg24 Challenge

Multi-class segmentation of the aorta in computed tomography angiography (CTA) scans is essential for diagnosing and planning complex endovascular treatments for patients with aortic dissections. However, existing methods reduce aortic segmentation to a binary problem, limiting their ability to measure diameters across different branches and zones. Furthermore, no open-source dataset is currently available to support the development of multi-class aortic segmentation methods. To address this gap, we organized the AortaSeg24 MICCAI Challenge, introducing the first dataset of 100 CTA volumes annotated for 23 clinically relevant aortic branches and zones. This dataset was designed to facilitate both model development and validation. The challenge attracted 121 teams worldwide, with participants leveraging state-of-the-art frameworks such as nnU-Net and exploring novel techniques, including cascaded models, data augmentation strategies, and custom loss functions. We evaluated the submitted algorithms using the Dice Similarity Coefficient (DSC) and Normalized Surface Distance (NSD), highlighting the approaches adopted by the top five performing teams. This paper presents the challenge design, dataset details, evaluation metrics, and an in-depth analysis of the top-performing algorithms. The annotated dataset, evaluation code, and implementations of the leading methods are publicly available to support further research. All resources can be accessed at https://aortaseg24.grand-challenge.org.

eess.IV↗

arXiv2024

LLM-driven Multimodal Target Volume Contouring in Radiation Oncology

Target volume contouring for radiation therapy is considered significantly more challenging than the normal organ segmentation tasks as it necessitates the utilization of both image and text-based clinical information. Inspired by the recent advancement of large language models (LLMs) that can facilitate the integration of the textural information and images, here we present a novel LLM-driven multimodal AI, namely LLMSeg, that utilizes the clinical text information and is applicable to the challenging task of target volume contouring for radiation therapy, and validate it within the context of breast cancer radiation therapy target volume contouring. Using external validation and data-insufficient environments, which attributes highly conducive to real-world applications, we demonstrate that the proposed model exhibits markedly improved performance compared to conventional unimodal AI models, particularly exhibiting robust generalization performance and data efficiency. To our best knowledge, this is the first LLM-driven multimodal AI model that integrates the clinical text information into target volume delineation for radiation oncology.

eess.IV↗

arXiv2024

Improving Cone-Beam CT Image Quality with Knowledge Distillation-Enhanced Diffusion Model in Imbalanced Data Settings

In radiation therapy (RT), the reliance on pre-treatment computed tomography (CT) images encounter challenges due to anatomical changes, necessitating adaptive planning. Daily cone-beam CT (CBCT) imaging, pivotal for therapy adjustment, falls short in tissue density accuracy. To address this, our innovative approach integrates diffusion models for CT image generation, offering precise control over data synthesis. Leveraging a self-training method with knowledge distillation, we maximize CBCT data during therapy, complemented by sparse paired fan-beam CTs. This strategy, incorporated into state-of-the-art diffusion-based models, surpasses conventional methods like Pix2pix and CycleGAN. A meticulously curated dataset of 2800 paired CBCT and CT scans, supplemented by 4200 CBCT scans, undergoes preprocessing and teacher model training, including the Brownian Bridge Diffusion Model (BBDM). Pseudo-label CT images are generated, resulting in a dataset combining 5600 CT images with corresponding CBCT images. Thorough evaluation using MSE, SSIM, PSNR and LPIPS demonstrates superior performance against Pix2pix and CycleGAN. Our approach shows promise in generating high-quality CT images from CBCT scans in RT.

cs.CV↗

arXiv2023

K-SMPC: Koopman Operator-Based Stochastic Model Predictive Control for Enhanced Lateral Control of Autonomous Vehicles

This paper proposes Koopman operator-based Stochastic Model Predictive Control (K-SMPC) for enhanced lateral control of autonomous vehicles. The Koopman operator is a linear map representing the nonlinear dynamics in an infinite-dimensional space. Thus, we use the Koopman operator to represent the nonlinear dynamics of a vehicle in dynamic lane-keeping situations. The Extended Dynamic Mode Decomposition (EDMD) method is adopted to approximate the Koopman operator in a finite-dimensional space for practical implementation. We consider the modeling error of the approximated Koopman operator in the EDMD method. Then, we design K-SMPC to tackle the Koopman modeling error, where the error is handled as a probabilistic signal. The recursive feasibility of the proposed method is investigated with an explicit first-step state constraint by computing the robust control invariant set. A high-fidelity vehicle simulator, i.e., CarSim, is used to validate the proposed method with a comparative study. From the results, it is confirmed that the proposed method outperforms other methods in tracking performance. Furthermore, it is observed that the proposed method satisfies the given constraints and is recursively feasible.

eess.SY↗

arXiv2023

Uncertainty Quantification of Autoencoder-based Koopman Operator

This paper proposes a method for uncertainty quantification of an autoencoder-based Koopman operator. The main challenge of using the Koopman operator is to design the basis functions for lifting the state. To this end, this paper builds an autoencoder to automatically search the optimal lifting basis functions with a given loss function. We approximate the Koopman operator in a finite-dimensional space with the autoencoder, while the approximated Koopman has an approximation uncertainty. To resolve the problem, we compute a robust positively invariant set for the approximated Koopman operator to consider the approximation error. Then, the decoder of the autoencoder is analyzed by robustness certification against approximation error using the Lipschitz constant in the reconstruction phase. The forced Van der Pol model is used to show the validity of the proposed method. From the numerical simulation results, we confirmed that the trajectory of the true state stays in the uncertainty set centered by the reconstructed state.

eess.SY↗

arXiv2023

RNN Controller for Lane-Keeping Systems with Robustness and Safety Verification

This paper proposes a Recurrent Neural Network (RNN) controller for lane-keeping systems, effectively handling model uncertainties and disturbances. First, quadratic constraints cover the nonlinearities brought by the RNN controller, and the linear fractional transformation method models the dynamics of system uncertainties. Second, we prove the robust stability of the lane-keeping system in the presence of uncertain vehicle speed using a linear matrix inequality. Then, we define a reachable set for the lane-keeping system. Finally, to confirm the safety of the lane-keeping system with tracking error bound, we formulate semidefinite programming to approximate the outer set of the reachable set. Numerical experiments demonstrate that this approach confirms the stabilizing RNN controller and validates the safety with an untrained dataset with untrained varying road curvatures.

eess.SY↗

arXiv2023

Classification Method of Road Surface Condition and Type with LiDAR Using Spatiotemporal Information

This paper proposes a spatiotemporal architecture with a deep neural network (DNN) for road surface conditions and types classification using LiDAR. It is known that LiDAR provides information on the reflectivity and number of point clouds depending on a road surface. Thus, this paper utilizes the information to classify the road surface. We divided the front road area into four subregions. First, we constructed feature vectors using each subregion's reflectivity, number of point clouds, and in-vehicle information. Second, the DNN classifies road surface conditions and types for each subregion. Finally, the output of the DNN feeds into the spatiotemporal process to make the final classification reflecting vehicle speed and probability given by the outcomes of softmax functions of the DNN output layer. To validate the effectiveness of the proposed method, we performed a comparative study with five other algorithms. With the proposed DNN, we obtained the highest accuracy of 98.0\% and 98.6\% for two subregions near the vehicle. In addition, we implemented the proposed method on the Jetson TX2 board to confirm that it is applicable in real-time.

eess.IV↗

arXiv2023

Reachable Set-based Path Planning for Automated Vertical Parking System

This paper proposes a local path planning method with a reachable set for Automated vertical Parking Systems (APS). First, given a parking lot layout with a goal position, we define an intermediate pose for the APS to accomplish reverse parking with a single maneuver, i.e., without changing the gear shift. Then, we introduce a reachable set which is a set of points consisting of the grid points of all possible intermediate poses. Once the APS approaches the goal position, it must select an intermediate pose in the reachable set. A minimization problem was formulated and solved to choose the intermediate pose. We performed various scenarios with different parking lot conditions. We used the Hybrid-A* algorithm for the global path planning to move the vehicle from the starting pose to the intermediate pose and utilized clothoid-based local path planning to move from the intermediate pose to the goal pose. Additionally, we designed a controller to follow the generated path and validated its tracking performance. It was confirmed that the tracking error in the mean root square for the lateral position was bounded within 0.06m and for orientation within 0.01rad.

cs.RO↗

arXiv2021

Abdominal synthetic CT reconstruction with intensity projection prior for MRI-only adaptive radiotherapy

An MRI-only adaptive radiotherapy (ART) workflow is desirable for managing interfractional changes in anatomy, but producing synthetic CT (sCT) data through paired data-driven deep learning (DL) for abdominal dose calculations remains a challenge due to the highly variable presence of intestinal gas. We present the preliminary dosimetric evaluation of our novel approach to sCT reconstruction that is well suited to handling intestinal gas in abdominal MRI-only ART. We utilize a paired data DL approach enabled by the intensity projection prior, in which well-matching training pairs are created by propagating air from MRI to corresponding CT scans. Evaluations focus on two classes: patients with (1) little involvement of intestinal gas, and (2) notable differences in intestinal gas presence between corresponding scans. Comparisons between sCT-based plans and CT-based clinical plans for both classes are made at the first treatment fraction to highlight the dosimetric impact of the variable presence of intestinal gas. Class 1 patients ($n=13$) demonstrate differences in prescribed dose coverage of the PTV of $1.3 \pm 2.1\%$ between clinical plans and sCT-based plans. Mean DVH differences in all structures for Class 1 patients are found to be statistically insignificant. In Class 2 ($n=20$), target coverage is $13.3 \pm 11.0\%$ higher in the clinical plans and mean DVH differences are found to be statistically significant. Significant deviations in calculated doses arising from the variable presence of intestinal gas in corresponding CT and MRI scans may limit the effectiveness of adaptive dose escalation efforts. We have proposed a paired data-driven DL approach to sCT reconstruction for accurate dose calculations in abdominal ART enabled by the creation of a clinically unavailable training data set with well-matching representations of intestinal gas.

physics.med-ph↗

arXiv2021

Robust Control for Lane Keeping System Using Linear Parameter Varying Approach with Scheduling Variables Reduction

This paper presents a robust controller using a Linear Parameter Varying (LPV) model of the lane-keeping system with parameter reduction. Both varying vehicle speed and roll motion on a curved road influence the lateral vehicle model parameters, such as tire cornering stiffness. Thus, we use the LPV technique to take the parameter variations into account in vehicle dynamics. However, multiple varying parameters lead to a high number of scheduling variables and cause massive computational complexity. In this paper, to reduce the computational complexity, Principal Component Analysis (PCA)-based parameter reduction is performed to obtain a reduced model with a tighter convex set. We designed the LPV robust feedback controller using the reduced model solving a set of Linear Matrix Inequality (LMI). The effectiveness of the proposed system is validated with full vehicle dynamics from CarSim on an interchange road. From the simulation, we confirmed that the proposed method largely reduces the lateral offset error, compared with other controllers based on Linear Time-Invariant (LTI) system.

eess.SY↗

arXiv2021

Intentional Deep Overfit Learning (IDOL): A Novel Deep Learning Strategy for Adaptive Radiation Therapy

In this study, we propose a tailored DL framework for patient-specific performance that leverages the behavior of a model intentionally overfitted to a patient-specific training dataset augmented from the prior information available in an ART workflow - an approach we term Intentional Deep Overfit Learning (IDOL). Implementing the IDOL framework in any task in radiotherapy consists of two training stages: 1) training a generalized model with a diverse training dataset of N patients, just as in the conventional DL approach, and 2) intentionally overfitting this general model to a small training dataset-specific the patient of interest (N+1) generated through perturbations and augmentations of the available task- and patient-specific prior information to establish a personalized IDOL model. The IDOL framework itself is task-agnostic and is thus widely applicable to many components of the ART workflow, three of which we use as a proof of concept here: the auto-contouring task on re-planning CTs for traditional ART, the MRI super-resolution (SR) task for MRI-guided ART, and the synthetic CT (sCT) reconstruction task for MRI-only ART. In the re-planning CT auto-contouring task, the accuracy measured by the Dice similarity coefficient improves from 0.847 with the general model to 0.935 by adopting the IDOL model. In the case of MRI SR, the mean absolute error (MAE) is improved by 40% using the IDOL framework over the conventional model. Finally, in the sCT reconstruction task, the MAE is reduced from 68 to 22 HU by utilizing the IDOL framework.

cs.LG↗

arXiv2015

Feasibility Study of Neutron Dose for Real Time Image Guided Proton Therapy: A Monte Carlo Study

Two full rotating gantry with different nozzles (Multipurpose nozzle with MLC, Scanning Dedicated nozzle) with conventional cyclotron system is installed and under commissioning for various proton treatment options at Samsung Medical Center in Korea. The purpose of this study is to investigate neutron dose equivalent per therapeutic dose, H/D, to x-ray imaging equipment under various treatment conditions with monte carlo simulation. At first, we investigated H/D with the various modifications of the beam line devices (Scattering, Scanning, Multi-leaf collimator, Aperture, Compensator) at isocenter, 20, 40, 60 cm distance from isocenter and compared with other research groups. Next, we investigated the neutron dose at x-ray equipments used for real time imaging with various treatment conditions. Our investigation showed the 0.07 ~ 0.19 mSv/Gy at x-ray imaging equipments according to various treatment options and intestingly 50% neutron dose reduction effect of flat panel detector was observed due to multi- leaf collimator during proton scanning treatment with multipurpose nozzle. In future studies, we plan to investigate experimental measurement of neutron dose and validation of simulation data for x-ray imaging equipment with additional neutron dose reduction method.

physics.med-ph↗