Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
20,539
result(s) for
"Kim, Young Jin"
Sort by:
Development of machine learning model for diagnostic disease prediction based on laboratory tests
The use of deep learning and machine learning (ML) in medical science is increasing, particularly in the visual, audio, and language data fields. We aimed to build a new optimized ensemble model by blending a DNN (deep neural network) model with two ML models for disease prediction using laboratory test results. 86 attributes (laboratory tests) were selected from datasets based on value counts, clinical importance-related features, and missing values. We collected sample datasets on 5145 cases, including 326,686 laboratory test results. We investigated a total of 39 specific diseases based on the International Classification of Diseases, 10th revision (ICD-10) codes. These datasets were used to construct light gradient boosting machine (LightGBM) and extreme gradient boosting (XGBoost) ML models and a DNN model using TensorFlow. The optimized ensemble model achieved an F1-score of 81% and prediction accuracy of 92% for the five most common diseases. The deep learning and ML models showed differences in predictive power and disease classification patterns. We used a confusion matrix and analyzed feature importance using the SHAP value method. Our new ML model achieved high efficiency of disease prediction through classification of diseases. This study will be useful in the prediction and diagnosis of diseases.
Journal Article
Electric Energy Consumption Prediction by Deep Learning with State Explainable Autoencoder
As energy demand grows globally, the energy management system (EMS) is becoming increasingly important. Energy prediction is an essential component in the first step to create a management plan in EMS. Conventional energy prediction models focus on prediction performance, but in order to build an efficient system, it is necessary to predict energy demand according to various conditions. In this paper, we propose a method to predict energy demand in various situations using a deep learning model based on an autoencoder. This model consists of a projector that defines an appropriate state for a given situation and a predictor that forecasts energy demand from the defined state. The proposed model produces consumption predictions for 15, 30, 45, and 60 min with 60-min demand to date. In the experiments with household electric power consumption data for five years, this model not only has a better performance with a mean squared error of 0.384 than the conventional models, but also improves the capacity to explain the results of prediction by visualizing the state with t-SNE algorithm. Despite unsupervised representation learning, we confirm that the proposed model defines the state well and predicts the energy demand accordingly.
Journal Article
Anti‐Inflammatory Effect of Quercetin on RAW 264.7 Mouse Macrophages Induced with Polyinosinic‐Polycytidylic Acid
Quercetin (3,3′,4′,5,6‐pentahydroxyflavone) is a well‐known antioxidant and a flavonol found in many fruits, leaves, and vegetables. Quercetin also has known anti‐inflammatory effects on lipopolysaccharide‐induced macrophages. However, the effects of quercetin on virus‐induced macrophages have not been fully reported. In this study, the anti‐inflammatory effect of quercetin on double‐stranded RNA (dsRNA)‐induced macrophages was examined. Quercetin at concentrations up to 50 μM significantly inhibited the production of NO, IL‐6, MCP‐1, IP‐10, RANTES, GM‐CSF, G‐CSF, TNF‐α, LIF, LIX, and VEGF as well as calcium release in dsRNA (50 μg/mL of polyinosinic‐polycytidylic acid)‐induced RAW 264.7 mouse macrophages (p < 0.05). Quercetin at concentrations up to 50 μM also significantly inhibited mRNA expression of signal transducer and activated transcription 1 (STAT1) and STAT3 in dsRNA‐induced RAW 264.7 cells (p < 0.05). In conclusion, quercetin had alleviating effects on viral inflammation based on inhibition of NO, cytokines, chemokines, and growth factors in dsRNA‐induced macrophages via the calcium‐STAT pathway.
Journal Article
Resistive-Based Gas Sensors Using Quantum Dots: A Review
Quantum dots (QDs) are used progressively in sensing areas because of their special electrical properties due to their extremely small size. This paper discusses the gas sensing features of QD-based resistive sensors. Different types of pristine, doped, composite, and noble metal decorated QDs are discussed. In particular, the review focus primarily on the sensing mechanisms suggested for these gas sensors. QDs show a high sensing performance at generally low temperatures owing to their extremely small sizes, making them promising materials for the realization of reliable and high-output gas-sensing devices.
Journal Article
Highly efficient oxygen evolution reaction via facile bubble transport realized by three-dimensionally stack-printed catalysts
Despite highly promising characteristics of three-dimensionally (3D) nanostructured catalysts for the oxygen evolution reaction (OER) in polymer electrolyte membrane water electrolyzers (PEMWEs), universal design rules for maximizing their performance have not been explored. Here we show that woodpile (WP)-structured Ir, consisting of 3D-printed, highly-ordered Ir nanowire building blocks, improve OER mass activity markedly. The WP structure secures the electrochemically active surface area (ECSA) through enhanced utilization efficiency of the extended surface area of 3D WP catalysts. Moreover, systematic control of the 3D geometry combined with theoretical calculations and various electrochemical analyses reveals that facile transport of evolved O
2
gas bubbles is an important contributor to the improved ECSA-specific activity. The 3D nanostructuring-based improvement of ECSA and ECSA-specific activity enables our well-controlled geometry to afford a 30-fold higher mass activity of the OER catalyst when used in a single-cell PEMWE than conventional nanoparticle-based catalysts.
Improved design of three-dimensionally nanostructured catalysts for oxygen evolution reaction (OER) can play a key role in maximizing the catalytic performance. Here, the authors show that woodpile-structured iridium consisting of 3D-printed, highly-ordered nanowire building blocks significantly improve OER mass activity.
Journal Article
Deep Reinforcement Learning-Based Adaptive Scheduling for Wireless Time-Sensitive Networking
Time-sensitive networking (TSN) technologies have garnered attention for supporting time-sensitive communication services, with recent interest extending to the wireless domain. However, adapting TSN to wireless areas faces challenges due to the competitive channel utilization in IEEE 802.11, necessitating exclusive channels for low-latency services. Additionally, traditional TSN scheduling algorithms may cause significant transmission delays due to dynamic wireless characteristics, which must be addressed. This paper proposes a wireless TSN model of IEEE 802.11 networks for the exclusive channel access and a novel time-sensitive traffic scheduler, named the wireless intelligent scheduler (WISE), based on deep reinforcement learning. We designed a deep reinforcement learning (DRL) framework to learn the repetitive transmission patterns of time-sensitive traffic and address potential latency issues from changing wireless conditions. Within this framework, we identified the most suitable DRL model, presenting the WISE algorithm with the best performance. Experimental results indicate that the proposed mechanisms meet up to 99.9% under the various wireless communication scenarios. In addition, they show that the processing delay is successfully limited within the specific time requirements and the scalability of TSN streams is guaranteed by the proposed mechanisms.
Journal Article
Deflection of light by a Coulomb charge in Born–Infeld electrodynamics
We study the propagation of light under a strong electric field in Born–Infeld electrodynamics. The nonlinear effect can be described by the effective indices of refraction. Because the effective indices of refraction depend on the background electric field, the path of light can be bent when the background field is non-uniform. We compute the bending angle of light by a Born–Infeld-type Coulomb charge in the weak lensing limit using the trajectory equation based on geometric optics. We also compute the deflection angle of light by the Einstein–Born–Infeld black hole using the geodesic equation and confirm that the contribution of the electric charge to the total bending angle agree.
Journal Article
Free-space transfer of comb-rooted optical frequencies over an 18 km open-air link
Phase-coherent transfer of optical frequencies over a long distance is required for diverse photonic applications, including optical clock dissemination and physical constants measurement. Several demonstrations were made successfully over fiber networks, but not much work has been done yet through the open air where atmospheric turbulence prevails. Here, we use an 18 km outdoor link to transmit multiple optical carriers extracted directly from a frequency comb of a 4.2 THz spectral width. In stabilization to a high-finesse cavity with a 1.5 Hz linewidth, the comb-rooted optical carriers are simultaneously transferred with collective suppression of atmospheric phase noise to −80 dBc Hz
−1
. Microwaves are also delivered by pairing two separate optical carriers bound with inter-comb-mode coherence, for example a 10 GHz signal with phase noise of −105 dBc Hz
−1
at 1 Hz offset. Lastly, an add-on demonstration is given for multi-channel coherent optical communications with the potential of multi-Tbps data transmission in free space.
Phase-coherent transfer of optical frequencies over long open-air paths is necessary in photonic applications. Here the authors demonstrate the parallel transmission of multiple optical carriers in air up to 18 km using a stable near-infrared frequency comb.
Journal Article
Roll-to-roll production of 30-inch graphene films for transparent electrodes
The outstanding electrical
1
, mechanical
2
,
3
and chemical
4
,
5
properties of graphene make it attractive for applications in flexible electronics
6
,
7
,
8
. However, efforts to make transparent conducting films from graphene have been hampered by the lack of efficient methods for the synthesis, transfer and doping of graphene at the scale and quality required for applications. Here, we report the roll-to-roll production and wet-chemical doping of predominantly monolayer 30-inch graphene films grown by chemical vapour deposition onto flexible copper substrates. The films have sheet resistances as low as ∼125 Ω □
−1
with 97.4% optical transmittance, and exhibit the half-integer quantum Hall effect, indicating their high quality. We further use layer-by-layer stacking to fabricate a doped four-layer film and measure its sheet resistance at values as low as ∼30 Ω □
−1
at ∼90% transparency, which is superior to commercial transparent electrodes such as indium tin oxides. Graphene electrodes were incorporated into a fully functional touch-screen panel device capable of withstanding high strain.
Graphene films with electrical and optical characteristics superior to indium tin oxide are produced in a roll-to-roll process and used to construct devices with flexible touch-screen panels.
Journal Article