Search

Your search keyword '"Lee, Won-Hee"' showing total 1,171 results
Start Over Author "Lee, Won-Hee"
1,171 results on '"Lee, Won-Hee"'

1. Diversify, Contextualize, and Adapt: Efficient Entropy Modeling for Neural Image Codec

Author

Kim, Jun-Hyuk, Kim, Seungeon, Lee, Won-Hee, and Oh, Dokwan

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Artificial Intelligence, Computer Science - Multimedia
Abstract

Designing a fast and effective entropy model is challenging but essential for practical application of neural codecs. Beyond spatial autoregressive entropy models, more efficient backward adaptation-based entropy models have been recently developed. They not only reduce decoding time by using smaller number of modeling steps but also maintain or even improve rate--distortion performance by leveraging more diverse contexts for backward adaptation. Despite their significant progress, we argue that their performance has been limited by the simple adoption of the design convention for forward adaptation: using only a single type of hyper latent representation, which does not provide sufficient contextual information, especially in the first modeling step. In this paper, we propose a simple yet effective entropy modeling framework that leverages sufficient contexts for forward adaptation without compromising on bit-rate. Specifically, we introduce a strategy of diversifying hyper latent representations for forward adaptation, i.e., using two additional types of contexts along with the existing single type of context. In addition, we present a method to effectively use the diverse contexts for contextualizing the current elements to be encoded/decoded. By addressing the limitation of the previous approach, our proposed framework leads to significant performance improvements. Experimental results on popular datasets show that our proposed framework consistently improves rate--distortion performance across various bit-rate regions, e.g., 3.73% BD-rate gain over the state-of-the-art baseline on the Kodak dataset., Comment: NeurIPS 2024

Published
2024

2. MAD: Multi-Alignment MEG-to-Text Decoding

Author

Yang, Yiqian, Jo, Hyejeong, Duan, Yiqun, Zhang, Qiang, Zhou, Jinni, Lee, Won Hee, Xu, Renjing, and Xiong, Hui

Subjects
Computer Science - Computation and Language
Abstract

Deciphering language from brain activity is a crucial task in brain-computer interface (BCI) research. Non-invasive cerebral signaling techniques including electroencephalography (EEG) and magnetoencephalography (MEG) are becoming increasingly popular due to their safety and practicality, avoiding invasive electrode implantation. However, current works under-investigated three points: 1) a predominant focus on EEG with limited exploration of MEG, which provides superior signal quality; 2) poor performance on unseen text, indicating the need for models that can better generalize to diverse linguistic contexts; 3) insufficient integration of information from other modalities, which could potentially constrain our capacity to comprehensively understand the intricate dynamics of brain activity. This study presents a novel approach for translating MEG signals into text using a speech-decoding framework with multiple alignments. Our method is the first to introduce an end-to-end multi-alignment framework for totally unseen text generation directly from MEG signals. We achieve an impressive BLEU-1 score on the $\textit{GWilliams}$ dataset, significantly outperforming the baseline from 5.49 to 10.44 on the BLEU-1 metric. This improvement demonstrates the advancement of our model towards real-world applications and underscores its potential in advancing BCI research. Code is available at $\href{https://github.com/NeuSpeech/MAD-MEG2text}{https://github.com/NeuSpeech/MAD-MEG2text}$.

Published
2024

3. REVECA: Adaptive Planning and Trajectory-based Validation in Cooperative Language Agents using Information Relevance and Relative Proximity

Author

Seo, SeungWon, Noh, SeongRae, Lee, Junhyeok, Lim, SooBin, Lee, Won Hee, and Kang, HyeongYeop

Subjects
Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Computer Science - Computer Vision and Pattern Recognition, Computer Science - Multiagent Systems
Abstract

We address the challenge of multi-agent cooperation, where agents achieve a common goal by cooperating with decentralized agents under complex partial observations. Existing cooperative agent systems often struggle with efficiently processing continuously accumulating information, managing globally suboptimal planning due to lack of consideration of collaborators, and addressing false planning caused by environmental changes introduced by other collaborators. To overcome these challenges, we propose the RElevance, Proximity, and Validation-Enhanced Cooperative Language Agent (REVECA), a novel cognitive architecture powered by GPT-4o-mini. REVECA enables efficient memory management, optimal planning, and cost-effective prevention of false planning by leveraging Relevance Estimation, Adaptive Planning, and Trajectory-based Validation. Extensive experimental results demonstrate REVECA's superiority over existing methods across various benchmarks, while a user study reveals its potential for achieving trustworthy human-AI cooperation., Comment: v2 is the AAAI'25 camera-ready version, including the appendix, which has been enhanced based on the reviewers' comments

Published
2024

4. Are EEG-to-Text Models Working?

Author

Jo, Hyejeong, Yang, Yiqian, Han, Juhyeok, Duan, Yiqun, Xiong, Hui, and Lee, Won Hee

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence
Abstract

This work critically analyzes existing models for open-vocabulary EEG-to-Text translation. We identify a crucial limitation: previous studies often employed implicit teacher-forcing during evaluation, artificially inflating performance metrics. Additionally, they lacked a critical benchmark - comparing model performance on pure noise inputs. We propose a methodology to differentiate between models that truly learn from EEG signals and those that simply memorize training data. Our analysis reveals that model performance on noise data can be comparable to that on EEG data. These findings highlight the need for stricter evaluation practices in EEG-to-Text research, emphasizing transparent reporting and rigorous benchmarking with noise inputs. This approach will lead to more reliable assessments of model capabilities and pave the way for robust EEG-to-Text communication systems. Code is available at https://github.com/NeuSpeech/EEG-To-Text

Published
2024

6. NeuSpeech: Decode Neural signal as Speech

Author

Yang, Yiqian, Duan, Yiqun, Zhang, Qiang, Jo, Hyejeong, Zhou, Jinni, Lee, Won Hee, Xu, Renjing, and Xiong, Hui

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence
Abstract

Decoding language from brain dynamics is an important open direction in the realm of brain-computer interface (BCI), especially considering the rapid growth of large language models. Compared to invasive-based signals which require electrode implantation surgery, non-invasive neural signals (e.g. EEG, MEG) have attracted increasing attention considering their safety and generality. However, the exploration is not adequate in three aspects: 1) previous methods mainly focus on EEG but none of the previous works address this problem on MEG with better signal quality; 2) prior works have predominantly used $``teacher-forcing"$ during generative decoding, which is impractical; 3) prior works are mostly $``BART-based"$ not fully auto-regressive, which performs better in other sequence tasks. In this paper, we explore the brain-to-text translation of MEG signals in a speech-decoding formation. Here we are the first to investigate a cross-attention-based ``whisper" model for generating text directly from MEG signals without teacher forcing. Our model achieves impressive BLEU-1 scores of 60.30 and 52.89 without pretraining $\&$ teacher-forcing on two major datasets ($\textit{GWilliams}$ and $\textit{Schoffelen}$). This paper conducts a comprehensive review to understand how speech decoding formation performs on the neural decoding tasks, including pretraining initialization, training $\&$ evaluation set splitting, augmentation, and scaling law. Code is available at https://github.com/NeuSpeech/NeuSpeech1$.

Published
2024

7. Mitochondrial DNA mosaicism in normal human somatic cells

Author

An, Jisong, Nam, Chang Hyun, Kim, Ryul, Lee, Yunah, Won, Hyein, Park, Seongyeol, Lee, Won Hee, Park, Hansol, Yoon, Christopher J., An, Yohan, Kim, Jie-Hyun, Jun, Jong Kwan, Bae, Jeong Mo, Shin, Eui-Cheol, Kim, Bun, Cha, Yong Jun, Kwon, Hyun Woo, Oh, Ji Won, Park, Jee Yoon, Kim, Min Jung, and Ju, Young Seok

Published
2024
Full Text
View/download PDF

11. Neural Video Compression with Temporal Layer-Adaptive Hierarchical B-frame Coding

Author

Kim, Yeongwoong, Bahk, Suyong, Kim, Seungeon, Lee, Won Hee, Oh, Dokwan, and Kim, Hui Yong

Subjects
Computer Science - Computer Vision and Pattern Recognition, Electrical Engineering and Systems Science - Image and Video Processing
Abstract

Neural video compression (NVC) is a rapidly evolving video coding research area, with some models achieving superior coding efficiency compared to the latest video coding standard Versatile Video Coding (VVC). In conventional video coding standards, the hierarchical B-frame coding, which utilizes a bidirectional prediction structure for higher compression, had been well-studied and exploited. In NVC, however, limited research has investigated the hierarchical B scheme. In this paper, we propose an NVC model exploiting hierarchical B-frame coding with temporal layer-adaptive optimization. We first extend an existing unidirectional NVC model to a bidirectional model, which achieves -21.13% BD-rate gain over the unidirectional baseline model. However, this model faces challenges when applied to sequences with complex or large motions, leading to performance degradation. To address this, we introduce temporal layer-adaptive optimization, incorporating methods such as temporal layer-adaptive quality scaling (TAQS) and temporal layer-adaptive latent scaling (TALS). The final model with the proposed methods achieves an impressive BD-rate gain of -39.86% against the baseline. It also resolves the challenges in sequences with large or complex motions with up to -49.13% more BD-rate gains than the simple bidirectional extension. This improvement is attributed to the allocation of more bits to lower temporal layers, thereby enhancing overall reconstruction quality with smaller bits. Since our method has little dependency on a specific NVC model architecture, it can serve as a general tool for extending unidirectional NVC models to the ones with hierarchical B-frame coding.

Published
2023

21. Graph Convolutional Network with Morphometric Similarity Networks for Schizophrenia Classification

Author

Park, Hye Won, Kim, Seo Yeong, Lee, Won Hee, Goos, Gerhard, Founding Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Greenspan, Hayit, editor, Madabhushi, Anant, editor, Mousavi, Parvin, editor, Salcudean, Septimiu, editor, Duncan, James, editor, Syeda-Mahmood, Tanveer, editor, and Taylor, Russell, editor

Published
2023
Full Text
View/download PDF

25. Modeling Cardiovascular Risks of E-Cigarettes With Human-Induced Pluripotent Stem Cell–Derived Endothelial Cells

Author

Lee, Won Hee, Ong, Sang-Ging, Zhou, Yang, Tian, Lei, Bae, Hye Ryeong, Baker, Natalie, Whitlatch, Adam, Mohammadi, Leila, Guo, Hongchao, Nadeau, Kari C, Springer, Matthew L, Schick, Suzaynn F, Bhatnagar, Aruni, and Wu, Joseph C

Subjects
Medical Biotechnology, Biomedical and Clinical Sciences, Tobacco Smoke and Health, Tobacco, Cardiovascular, Electronic Nicotine Delivery Systems, Stem Cell Research - Induced Pluripotent Stem Cell - Human, Stem Cell Research, Stem Cell Research - Induced Pluripotent Stem Cell, 2.2 Factors relating to the physical environment, Good Health and Well Being, Adult, Cardiovascular Diseases, Case-Control Studies, Cytokines, Endothelial Cells, Humans, Induced Pluripotent Stem Cells, Leukocyte Count, Macrophages, Nicotine, Reactive Oxygen Species, Smokers, Vaping, e-cigarette aerosol, e-liquid flavoring, endothelial dysfunction, iPSC-ECs, Cardiorespiratory Medicine and Haematology, Public Health and Health Services, Cardiovascular System & Hematology, Cardiovascular medicine and haematology
Abstract

BackgroundElectronic cigarettes (e-cigarettes) have experienced a tremendous increase in use. Unlike cigarette smoking, the effects of e-cigarettes and their constituents on mediating vascular health remain understudied. However, given their increasing popularity, it is imperative to evaluate the health risks of e-cigarettes, including the effects of their ingredients, especially nicotine and flavorings.ObjectivesThe purpose of this study was to investigate the effects of flavored e-cigarette liquids (e-liquids) and serum isolated from e-cigarette users on endothelial health and endothelial cell-dependent macrophage activation.MethodsHuman-induced pluripotent stem cell-derived endothelial cells (iPSC-ECs) and a high-throughput screening approach were used to assess endothelial integrity following exposure to 6 different e-liquids with varying nicotine concentrations and to serum from e-cigarette users.ResultsThe cytotoxicity of the e-liquids varied considerably, with the cinnamon-flavored product being most potent and leading to significantly decreased cell viability, increased reactive oxygen species (ROS) levels, caspase 3/7 activity, and low-density lipoprotein uptake, activation of oxidative stress-related pathway, and impaired tube formation and migration, confirming endothelial dysfunction. Upon exposure of ECs to e-liquid, conditioned media induced macrophage polarization into a pro-inflammatory state, eliciting the production of interleukin-1β and -6, leading to increased ROS. After exposure of human iPSC-ECs to serum of e-cigarette users, increased ROS linked to endothelial dysfunction was observed, as indicated by impaired pro-angiogenic properties. There was also an observed increase in inflammatory cytokine expression in the serum of e-cigarette users.ConclusionsAcute exposure to flavored e-liquids or e-cigarette use exacerbates endothelial dysfunction, which often precedes cardiovascular diseases.

Published
2019

26. Single-Cell RNA Sequencing of Human Embryonic Stem Cell Differentiation Delineates Adverse Effects of Nicotine on Embryonic Development.

Author

Guo, Hongchao, Tian, Lei, Zhang, Joe Z, Kitani, Tomoya, Paik, David T, Lee, Won Hee, and Wu, Joseph C

Subjects
Myocytes, Cardiac, Humans, Calcium, Reactive Oxygen Species, Nicotine, Gene Expression Profiling, Sequence Analysis, RNA, Cell Communication, Cell Cycle, Cell Differentiation, Gene Expression Regulation, Developmental, Embryonic Development, Single-Cell Analysis, Transcriptome, Human Embryonic Stem Cells, differentiation, embryonic development, embryonic stem cells, nicotine, single-cell RNA sequencing, smoking, Gene Expression Regulation, Developmental, Myocytes, Cardiac, Sequence Analysis, RNA, Biochemistry and Cell Biology, Clinical Sciences
Abstract

Nicotine, the main chemical constituent of tobacco, is highly detrimental to the developing fetus by increasing the risk of gestational complications and organ disorders. The effects of nicotine on human embryonic development and related mechanisms, however, remain poorly understood. Here, we performed single-cell RNA sequencing (scRNA-seq) of human embryonic stem cell (hESC)-derived embryoid body (EB) in the presence or absence of nicotine. Nicotine-induced lineage-specific responses and dysregulated cell-to-cell communication in EBs, shedding light on the adverse effects of nicotine on human embryonic development. In addition, nicotine reduced cell viability, increased reactive oxygen species (ROS), and altered cell cycling in EBs. Abnormal Ca2+ signaling was found in muscle cells upon nicotine exposure, as verified in hESC-derived cardiomyocytes. Consequently, our scRNA-seq data suggest direct adverse effects of nicotine on hESC differentiation at the single-cell level and offer a new method for evaluating drug and environmental toxicity on human embryonic development in utero.

Published
2019

27. Association of clozapine with structural and resting-state functional abnormalities of the hippocampus in chronic schizophrenia.

Author

Joo, Sung Woo, Kim, Sang Kyoung, Lee, Won Hee, Kim, Se Hyun, and Lee, Jungsun

Subjects
GRANULE cells, FUNCTIONAL magnetic resonance imaging, DENTATE gyrus, FUNCTIONAL connectivity, FUNCTIONAL groups
Abstract

Introduction: Abnormalities in the hippocampus have been extensively reported in schizophrenia research. However, inconsistent findings exist, and how structural and functional abnormalities of the hippocampus are associated with clinical symptoms in schizophrenia, especially concerning clozapine treatment, remains uncertain. Methods: We recruited 52 patients with schizophrenia, each with an illness duration of at least 5 years, and categorized them based on clozapine treatment. T1-weighted images and resting-state functional magnetic resonance imaging scans were obtained and analyzed to perform group comparisons of the structural and functional changes in the hippocampus. Volumes of the hippocampal subregions, as well as resting-state functional connectivity maps from these areas were compared between the groups. Associations with clinical symptoms, including the severity of psychiatric symptoms and cognitive functions, were investigated. Results: The clozapine group (n=23) exhibited smaller volumes in several hippocampal subregions, including the CA1, CA4, granule cell and molecular layers of the dentate gyrus, compared to the non-clozapine group (n=29). Seven clusters with significant group differences in functional connectivity with these hippocampal subregions were identified, with six of these clusters showing increased functional connectivity in the clozapine group. The reduced volumes of the hippocampal subregions were moderately associated with the severity of negative symptoms, general intelligence, and executive function. Discussion: Patients with schizophrenia undergoing clozapine treatment exhibited smaller volumes in the hippocampal subregions, which were moderately associated with negative symptoms and cognitive functions, compared to those without clozapine treatment. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

28. Assessing Radiation Exposure and Contrast Agent Utilization: A Comparative Analysis of the Woven EndoBridge Device and Stent-Assisted Coil Embolization for Managing Unruptured Wide-Neck Bifurcation Aneurysms.

Author

Baek, Jinwook, Jeong, Hae Woong, Han, Ji-Yeon, Heo, Young Jin, Yun, Sooyoung, Lee, Won Hee, and Kim, Sung Tae

Subjects
INTERNAL carotid artery, CONTRAST media, INTRACRANIAL aneurysms, ENDOVASCULAR surgery, THERAPEUTIC embolization, FLUOROSCOPY
Abstract

Purpose: In this study, we determined whether there were significant differences in procedure time, radiation dose, fluoroscopy time, and total contrast media dose when unruptured wideneck bifurcation aneurysms (WNBAs) were treated with the Woven EndoBridge (WEB) device and stent-assisted coil (SAC) embolization. Materials and Methods: The WEB device and SAC embolization (14:17) were used to treat 31 cases of internal carotid artery bifurcation, anterior communicating artery, middle cerebral artery bifurcation, and basilar bifurcation aneurysms between August 2021 and December 2022. The procedure time, radiation dose, fluoroscopy time, and total contrast medium dose between the 2 treatment groups were compared and analyzed. In the WEB device group, the results between operators were compared, and the follow-up radiologic outcomes were investigated. Results: The procedure and fluoroscopy times were significantly shorter in the WEB device group. Radiation and total contrast media dose were also significantly smaller in the WEB device, but there was no significant difference in results between operators. The follow-up radiological outcome showed adequate occlusion in 83.3% (10/12) of cases. Conclusion: The WEB device can be used as an alternative treatment method among the available endovascular treatment methods for WNBAs to reduce radiation exposure and the dose of contrast media when used adequately with appropriate indications. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

36. A Feature Fusion Model Based on Temporal Convolutional Network for Automatic Sleep Staging Using Single-Channel EEG

Author

Bao, Jiameng, Wang, Guangming, Wang, Tianyu, Wu, Ning, Hu, Shimin, Lee, Won Hee, Lo, Sio-Long, Yan, Xiangguo, Zheng, Yang, and Wang, Gang

Abstract

Sleep staging is a crucial task in sleep monitoring and diagnosis, but clinical sleep staging is both time-consuming and subjective. In this study, we proposed a novel deep learning algorithm named feature fusion temporal convolutional network (FFTCN) for automatic sleep staging using single-channel EEG data. This algorithm employed a one-dimensional convolutional neural network (1D-CNN) to extract temporal features from raw EEG, and a two-dimensional CNN (2D-CNN) to extract time-frequency features from spectrograms generated through continuous wavelet transform (CWT) at the epoch level. These features were subsequently fused and further fed into a temporal convolutional network (TCN) to classify sleep stages at the sequence level. Moreover, a two-step training strategy was used to enhance the model's performance on an imbalanced dataset. Our proposed method exhibits superior performance in the 5-class classification task for healthy subjects, as evaluated on the SHHS-1, Sleep-EDF-153, and ISRUC-S1 datasets. This work provided a straightforward and promising method for improving the accuracy of automatic sleep staging using only single-channel EEG, and the proposed method exhibited great potential for future applications in professional sleep monitoring, which could effectively alleviate the workload of sleep technicians.

Published
2024
Full Text
View/download PDF

39. Alterations in the multilayer network in patients with rapid eye movement sleep behaviour disorder.

Author

Lee, Dong Ah, Lee, Won Hee, Lee, Ho‐Joon, and Park, Kang Min

Subjects
*PARASOMNIAS, *DIFFUSION tensor imaging, *MAGNETIC resonance imaging, *CINGULATE cortex, *BEHAVIOR disorders, *RAPID eye movement sleep
Abstract

Summary: This study aimed to reveal the pathophysiology of isolated rapid eye movement sleep behaviour disorder (RBD) in patients using multilayer network analysis. Participants eligible for isolated RBD were included and verified via polysomnography. Both iRBD patients and healthy controls underwent brain MRI, including T1‐weighted imaging and diffusion tensor imaging. Grey matter matrix was derived from T1‐weighted images using a morphometric similarity network. White matter matrix was formed from diffusion tensor imaging‐based structural connectivity. Multilayer network analysis of grey and white matter was performed using graph theory. We studied 29 isolated RBD patients and 30 healthy controls. Patients exhibited a higher average overlap degree (27.921 vs. 23.734, p = 0.002) and average multilayer clustering coefficient (0.474 vs. 0.413, p = 0.002) compared with controls. Additionally, several regions showed significant differences in the degree of overlap and multilayer clustering coefficient between patients with isolated RBD and healthy controls at the nodal level. The degree of overlap in the left medial orbitofrontal, left posterior cingulate, and right paracentral nodes and the multilayer clustering coefficients in the left lateral occipital, left rostral middle frontal, right fusiform, right inferior posterior parietal, and right parahippocampal nodes were higher in patients with isolated RBD than in healthy controls. We found alterations in the multilayer network at the global and nodal levels in patients with isolated RBD, and these changes may be associated with the pathophysiology of isolated RBD. Multilayer network analysis can be used widely to explore the mechanisms underlying various neurological disorders. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

47. Comparison of Magnetic Resonance Imaging and Serum Biomarkers for Detection of Human Pluripotent Stem Cell-Derived Teratomas

Author

Riegler, Johannes, Ebert, Antje, Qin, Xulei, Shen, Qi, Wang, Mouer, Ameen, Mohamed, Kodo, Kazuki, Ong, Sang-Ging, Lee, Won Hee, Lee, Grace, Neofytou, Evgenios, Gold, Joseph D, Connolly, Andrew J, and Wu, Joseph C

Subjects
Biochemistry and Cell Biology, Biological Sciences, Biomedical Imaging, Stem Cell Research - Induced Pluripotent Stem Cell - Human, Heart Disease, Stem Cell Research - Induced Pluripotent Stem Cell, Cardiovascular, Regenerative Medicine, Stem Cell Research, Clinical Research, Heart Disease - Coronary Heart Disease, Stem Cell Research - Embryonic - Human, Animals, Biomarkers, Tumor, Cell Differentiation, Cell Proliferation, Gadolinium, Heart, Humans, Induced Pluripotent Stem Cells, Inflammation, Lentivirus, Magnetic Resonance Imaging, Male, Myocytes, Cardiac, Phenotype, Pluripotent Stem Cells, Rats, Nude, Teratoma, Tumor Burden, magnetic resonance imaging, microRNA biomarker, pluripotent stem cells, serum biomarker, tumorigenicity, ultrasound imaging, Clinical Sciences, Biochemistry and cell biology
Abstract

The use of cells derived from pluripotent stem cells (PSCs) for regenerative therapies confers a considerable risk for neoplastic growth and teratoma formation. Preclinical and clinical assessment of such therapies will require suitable monitoring strategies to understand and mitigate these risks. Here we generated human-induced pluripotent stem cells (iPSCs), selected clones that continued to express reprogramming factors after differentiation into cardiomyocytes, and transplanted these cardiomyocytes into immunocompromised rat hearts post-myocardial infarction. We compared magnetic resonance imaging (MRI), cardiac ultrasound, and serum biomarkers for their ability to delineate teratoma formation and growth. MRI enabled the detection of teratomas with a volume >8 mm(3). A combination of three plasma biomarkers (CEA, AFP, and HCG) was able to detect teratomas with a volume >17 mm(3) and with a sensitivity of more than 87%. Based on our findings, a combination of serum biomarkers with MRI screening may offer the highest sensitivity for teratoma detection and tracking.

Published
2016

Catalog

Books, media, physical & digital resources
See catalog results