Jae-Ho Han

Background: Affinity-maturation phage-display next-generation sequencing (NGS) yields more paired single-chain variable fragment clones than can be characterized experimentally, creating a fixed-budget prioritization problem. Read counts provide empirical support rather than direct affinity labels. We developed AbPACER (Antibody Parent-Aware Contextual Evidence Ranker), an affinity-label-blind neural ranker combining parent-relative mutation descriptors, frozen antibody-language-model context, and NGS evidence from related clones. AbPACER is campaign-adaptive rather than zero-shot: for each campaign, it is fitted to paired sequences and round-resolved R1–R3 counts before returning a 384-candidate assay list. We evaluated it in two retrospective phage-display campaigns and separately assessed its supervised mean-squared-error adaptation on AlphaSeq, denoted AbPACER-MSE. Results: From frozen top-5% candidate sets containing 16,323 Fas-associated factor 1 (FAF1) and 7,487 vascular endothelial growth factor receptor (VEGFR) clones, each method ranked the complete target-specific set and selected 384 candidates. In FAF1, AbPACER recovered 2.00 ± 0.00 of seven retrospective panel clones, recovering two in every seed, compared with 1 / 7 by total count, 1.00 ± 0.00 by Ens-Grad CNN, 1.67 ± 1.15 by A2Binder-HL, and 1.33 ± 0.58 by AbAffinity. In VEGFR, AbPACER recovered 2.33 ± 0.58 of three panel clones, the highest observed learned-method mean, whereas total count recovered 3 / 3. No learned method was uniformly best at broader hypothetical budgets. On the public AlphaSeq common split of 11,670 fixed-test variants, AbPACER-MSE recovered 187.0 ± 2.6 of the true top-384, closely matching AbAffinity (188.0 ± 2.6) and exceeding A2Binder (175.7 ± 6.4) and Ens-Grad CNN (154.0 ± 6.1). AbPACER-MSE updated 1.378 million task-specific parameters, compared with 651.04 million for AbAffinity, and achieved Pearson 0.687 ± 0.003 and Spearman 0.652 ± 0.002. Conclusions: AbPACER provides a campaign-specific, parent-aware framework for fixed-budget prioritization from affinity-label-blind phage-display NGS data. At the 384-candidate endpoint, it showed the highest mean recovery among learned methods in both retrospective campaigns. AbPACER-MSE closely matched AbAffinity in true top-384 recovery while updating substantially fewer task-specific parameters. These results motivate prospective evaluation of sequence-conditioned reranking as a complement to count-based prioritization.

In surveillance environments, detecting anomalies requires understanding the contextual dynamics of the environment, human behaviors, and movements within a scene. Effective anomaly detection must address both the where and what of events, but existing approaches such as unimodal action-based methods or LLM-integrated multimodal frameworks have limitations. These methods either rely on implicit scene information, making it difficult to localize where anomalies occur, or fail to adapt to surveillance specific challenges such as view changes, subtle actions, low light conditions, and crowded scenes. As a result, these challenges hinder accurate detection of what occurs. To overcome these limitations, our system takes advantage of features from a lightweight scene classification model to discern where an event occurs, acquiring explicit location-based context. To identify what events occur, it focuses on atomic actions, which remain underexplored in this field and are better suited to interpreting intricate abnormal behaviors than conventional abstract action features. To achieve robust anomaly detection, the proposed Temporal-Semantic Relationship Network (TSRN) models spatio-temporal relationships among multimodal features and employs a Segment-selective Focal Margin loss (SFML) to effectively address class imbalance, outperforming conventional MIL-based methods. Compared to existing methods, experimental results demonstrate that our system significantly reduces false alarms while maintaining robustness across diverse scenarios. Quantitative and qualitative evaluations on public datasets validate the practical effectiveness of the proposed method for real-world surveillance applications.

The proliferation of portable and wearable electroencephalography (EEG) devices has encouraged EEG research in various areas. These devices, while convenient, often come with limited computational capabilities. However, the challenge of minimizing network complexity for such edge devices was not fully addressed in previous studies. To tackle this, a scalable hybrid network is proposed to classify EEG signals with different demographic factors on edge devices. This model blends a convolutional neural network (CNN) with a self‐attention mechanism in a hybrid block structure. This design alternates between CNN layers and self‐attention layers to efficiently capture both local and global features. In this study, EEG signals acquired using a portable EEG device during gaming session is classified particularly into pre‐puberty and puberty stages. The developed scalable hybrid network (SH‐Net) has shown promising results in distinguishing between pre‐puberty and puberty EEG signals. As a result, the first stage of this model showed higher accuracy compared to other models in 10‐fold cross‐validation: 93.57% for four channels, 89.04% for the frontal lobe channels (AF), and 81.46% for the temporal lobe channels (TP). Notably, the third stage of this model, using the AF channel, achieved higher accuracy compared to other evaluated models that utilized four channels.

As games have been applied across various fields, including education and healthcare, numerous new games tailored to each field have emerged. Therefore, understanding user behavior has become crucial in securing the right players for each type of game. This study provides valuable insights for improving game development by measuring the electroencephalography (EEG) of game users and classifying the frequency of game usage. The multimodal mobile brain-body imaging (MOBI) dataset was employed for this study, and the frequency of game usage was categorized into ”often” and ”sometimes”. To achieve decent classification accuracy, a novel bimodal Transformer architecture featuring dedicated channels for the frontal (AF) and temporal (TP) lobes is introduced, wherein convolutional layers, self-attention mechanisms, and cross-attention mechanisms are integrated into a unified model. The model, designed to differentiate between AF and TP channels, exhibits functional differences between brain regions, allowing for a detailed analysis of inter-channel correlations. Evaluated through five-fold cross-validation (CV) and leave-one-subject-out cross-validation (LOSO CV), the proposed model demonstrates classification accuracies of 88.86% and 85.11%, respectively. By effectively classifying gameplay frequency, this methodology provides valuable insights for targeted game participation and contributes to strategic efforts to develop and design customized games for player acquisition.

Several techniques have been developed to overcome the limitation of sensor bandwidth for 2D signals [1] . Though compressive sensing is an attractive technique that reduces the number of measurements required to record information on a sparse signal basis [2] , [3] , recording information beyond the Nyquist frequency remains difficult when working with nonsparse signals. Given this constraint, this article focuses on the use of the physical bandwidth of a coherent signal in the complex form instead of its intensity form. The resulting trick combines holographic multiplexing with sampling scheme optimization to obtain the information in a 2D coherent signal from beyond the Nyquist frequency range. The prerequisites for understanding this article are a knowledge of basic algebra and the Fourier transform. Familiarity with holography is also beneficial.

Ocular axial length (AL) measurement is important in ophthalmology because it should be considered prior to operations, such as strabismus surgery or cataract surgery, and the automation of AL measurement with easily obtained retinal fundus images has been studied. However, the performance of deep learning methods inevitably depends on distribution of the data set used, and the lack of data is an issue that needs to be addressed. In this study, we propose a framework for generating pairs of fundus images and their corresponding ALs to improve the AL inference. The generator’s encoder was trained independently using metric learning based on the AL information. A random vector and zero padding were incorporated into the generator to increase data creation flexibility, after which AL information was inserted as conditional information. We verified the effectiveness of this framework by evaluating the performance of AL inference models after training them on a combined data set comprising privately collected actual data and data generated by the proposed method. Compared to using only the actual data set, the mean absolute error and standard deviation of the proposed method decreased from 10.23 and 2.56 to 3.96 and 0.23, respectively, even with a smaller number of layers in the AL prediction models.

Herein, we propose a novel model-based simultaneous multi-slice (SMS) reconstruction method by exploiting data-driven parameter modeling for highly accelerated T1 parameter quantification. We assume that the predefined slice-specific null space operator remains invariant along the parameter dimension. We incorporate the parameter dimension into SMS-HSL to exploit Hankel-structured and Casorati matrices. Given this consideration, the SMS signal is reformulated in k-p space as a constrained optimization problem that exploits rank deficiency for the Hankel-structured matrix and a finite-dimensional basis for a subspace containing slowly evolving signals in the parameter direction. The proposed model-based SMS reconstruction method is validated on in vivo data and compared with state-of-the-art methods with slice acceleration factors of 3 and 5, including an in-plane acceleration factor of 2. The experimental results demonstrate that the proposed method performs effective slice unfolding and signal recovery in reconstructed images and T1 maps with high precision as compared to the state-of-the-art methods.