Distinguishing nontuberculous mycobacterial lung disease and Mycobacterium tuberculosis lung disease on X-ray images using deep transfer learning.

BACKGROUND: Nontuberculous mycobacterial lung disease (NTM-LD) and Mycobacterium tuberculosis lung disease (MTB-LD) have similar clinical characteristics. Therefore, NTM-LD is sometimes incorrectly diagnosed with MTB-LD and treated incorrectly. To solve these difficulties, we aimed to distinguish the two diseases in chest X-ray images using deep learning technology, which has been used in various fields recently. METHODS: We retrospectively collected chest X-ray images from 3314 patients infected with Mycobacterium tuberculosis (MTB) or nontuberculosis mycobacterium (NTM). After selecting the data according to the diagnostic criteria, various experiments were conducted to create the optimal deep learning model. A performance comparison was performed with the radiologist. Additionally, the model performance was verified using newly collected MTB-LD and NTM-LD patient data. RESULTS: Among the implemented deep learning models, the ensemble model combining EfficientNet B4 and ResNet 50 performed the best in the test data. Also, the ensemble model outperformed the radiologist on all evaluation metrics. In addition, the accuracy of the ensemble model was 0.85 for MTB-LD and 0.78 for NTM-LD on an additional validation dataset consisting of newly collected patients. CONCLUSIONS: In previous studies, it was known that it was difficult to distinguish between MTB-LD and NTM-LD in chest X-ray images, but we have successfully distinguished the two diseases using deep learning methods. This study has the potential to aid clinical decisions if the two diseases need to be differentiated.

Highly efficient homology-directed repair using CRISPR/Cpf1-geminiviral replicon in tomato.

Genome editing via the homology-directed repair (HDR) pathway in somatic plant cells is very inefficient compared with error-prone repair by nonhomologous end joining (NHEJ). Here, we increased HDR-based genome editing efficiency approximately threefold compared with a Cas9-based single-replicon system via the use of de novo multi-replicon systems equipped with CRISPR/LbCpf1 in tomato and obtained replicon-free but stable HDR alleles. The efficiency of CRISPR/LbCpf1-based HDR was significantly modulated by physical culture conditions such as temperature and light. Ten days of incubation at 31 °C under a light/dark cycle after Agrobacterium-mediated transformation resulted in the best performance among the tested conditions. Furthermore, we developed our single-replicon system into a multi-replicon system that effectively increased HDR efficiency. Although this approach is still challenging, we showed the feasibility of HDR-based genome editing of a salt-tolerant SlHKT1;2 allele without genomic integration of antibiotic markers or any phenotypic selection. Self-pollinated offspring plants carrying the HKT1;2 HDR allele showed stable inheritance and germination tolerance in the presence of 100 mm NaCl. Our work may pave the way for transgene-free editing of alleles of interest in asexually and sexually reproducing plants.

Septal Integration Graft, a Modified Direct Type of Septal Extension Graft for the Treatment of Unilateral Cleft Lip Nose Deformity: An Anthropometric Analysis.

BACKGROUND: The purpose of this study was to evaluate the cosmetic outcomes and stability of septal integration grafts (SIGs), a modified direct type of septal extension graft, in patients with cleft lip nose deformity. METHODS: Seventy-three patients underwent SIGs between January 2002 and July 2017. The exclusion criteria were patients who lacked clinical photographs preoperatively or postoperatively. We reviewed data regarding the patient demographics, combined procedures, and complications as well as preoperative and postoperative clinical photography. Eleven measurement items were evaluated from basal and right lateral views with photogrammetry and 3-dimensional anthropometry using standardized clinical photographic techniques. RESULTS: Sixty-one patients were analyzed in this study. Twelve patients were excluded owing to a lack of postoperative photographs. The mean ± SD follow-up period was 26.70 ± 29.43 months (range, 6-134 months). Compared with the preoperative value, the nasal tip projection index, height of nose index, and nostril axis inclination significantly improved. The columellar length (Rt-Lt) index significantly decreased, which implied improved columellar deviation. Complications included hematoma (n = 1), wound dehiscence (n = 1), tearing of septal mucosa (n = 3), and silicone implant deviation (n = 2). CONCLUSIONS: We used SIGs to correct nasal deformity in patients with cleft lip nose deformity. The SIG can minimize the amount of the graft used for septal extension and can effectively correct nasal tip and columella deformities.

Multi-Population Genetic Algorithm for Multilabel Feature Selection Based on Label Complementary Communication.

Multilabel feature selection is an effective preprocessing step for improving multilabel classification accuracy, because it highlights discriminative features for multiple labels. Recently, multi-population genetic algorithms have gained significant attention with regard to feature selection studies. This is owing to their enhanced search capability when compared to that of traditional genetic algorithms that are based on communication among multiple populations. However, conventional methods employ a simple communication process without adapting it to the multilabel feature selection problem, which results in poor-quality final solutions. In this paper, we propose a new multi-population genetic algorithm, based on a novel communication process, which is specialized for the multilabel feature selection problem. Our experimental results on 17 multilabel datasets demonstrate that the proposed method is superior to other multi-population-based feature selection methods.

Molecular-Level Understanding of Continuous Growth from Iron-Oxo Clusters to Iron Oxide Nanoparticles.

The formation of inorganic nanoparticles has been understood based on the classical crystallization theory described by a burst of nucleation, where surface energy is known to play a critical role, and a diffusion-controlled growth process. However, this nucleation and growth model may not be universally applicable to the entire nanoparticle systems because different precursors and surface ligands are used during their synthesis. Their intrinsic chemical reactivity can lead to a formation pathway that deviates from a classical nucleation and growth model. The formation of metal oxide nanoparticles is one such case because of several distinct chemical aspects during their synthesis. Typical carboxylate surface ligands, which are often employed in the synthesis of oxide nanoparticles, tend to continuously remain on the surface of the nanoparticles throughout the growth process. They can also act as an oxygen source during the growth of metal oxide nanoparticles. Carboxylates are prone to chemical reactions with different chemical species in the synthesis such as alcohol or amine. Such reactions can frequently leave reactive hydroxyl groups on the surface. Herein, we track the entire growth process of iron oxide nanoparticles synthesized from conventional iron precursors, iron-oleate complexes, with strongly chelating carboxylate moieties. Mass spectrometry studies reveal that the iron-oleate precursor is a cluster comprising a tri-iron-oxo core and carboxylate ligands rather than a mononuclear complex. A combinatorial analysis shows that the entire growth, regulated by organic reactions of chelating ligands, is continuous without a discrete nucleation step.

A prospective 1-year comparative study of endoscopic thyroidectomy via a retroauricular approach versus conventional open thyroidectomy at a single institution.

OBJECTIVE: The objective of this study was to evaluate the feasibility and safety of performing an endoscopic thyroidectomy (ETE) via a retroauricular approach. METHODS: Forty-seven patients who underwent ETE via a retroauricular approach were included, and a total of 47 patients who underwent conventional open thyroid lobectomy in the same period were analyzed as a control group. All patients underwent prospective functional evaluations before the operation and 1 week, and 1, 3, 6, and 12 months postoperatively using a comprehensive battery of functional assessments. RESULTS: The mean total operative time was 152 ± 48 min, with a mean endoscopic procedure time of 58 ± 18 min. One patient developed temporary vocal fold paralysis. Although most of the parameters for the functional outcome were worse in the ETE group, these differences were transient. Subjective worsening on the voice handicap index and dysphagia handicap index normalized by 3 months postoperatively. The average pain score on a visual analog scale at 1 week after surgery was 2.84, representing a tolerable range of discomfort. The mean paresthesia/hyperesthesia score was worse in the ETE group than the open surgery group by postoperative month 6; however, these differences eventually disappeared. Thirty-six of the 47 patients in the ETE group were satisfied or extremely satisfied with the retroauricular incision by 6 months after surgery. CONCLUSIONS: ETE via a retroauricular approach is a safe, feasible, and cosmetically desirable treatment option, with outcomes comparable to conventional open thyroidectomy in the longer term.

Optimum Thickness Conditions of TiO2 Nanotubes Layer for Efficient Electrochemical Luminescence Cells Application.

We report a TiO2 nanotubes (NTs)-based Electrochemical luminescence (ECL) cell. The ECL cell was fabricated using the electrode of TiO2 NTs and Ru(II) complex (Ru(bpy)2+(3)) as a luminescence materials. The fabricated ECL cell is composed of F-doped SnO2 (FTO) glass/Ru(II)/TiO2 NTs/Ti plate. At a bias voltage of 3 V, the measured ECL efficiencies were 0 Im/W for cell without NTs, 0.03 Im/W for NTs-6.5 µm, 0.07 Im/W for NTs-8 µm and 0.1 Im/W for NTs-10 µm, respectively. The use of Ti02 NTs increases ECL intensities by about 2 times compared to the typical ECL cell without the use of TiO2 NTs.

Suppressing instability of liquid thin films by a fibril network and its application to micropatterning without a residual layer.

This study proposes a method to coat thin films of non-volatile solvents on substrates. A small amount of crystalline polymer dissolved in solvents forms a network of crystalline fibrils during the coating process. The network suppresses dewetting of the solvent liquid and helps the liquid film sustaining on the substrate. This strategy can be used in soft lithography to generate micropatterns of diverse materials without having a residual layer. This process does not request etching for achieving residual layer-free micropatterns, which has been a long challenge in soft lithography. As examples, we demonstrate micropatterns of polymer hydrogels and metal oxides (ZnO, In2O3).

Growth of long triisopropylsilylethynyl pentacene (TIPS-PEN) nanofibrils in a polymer thin film during spin-coating.

This study demonstrates the growth of long triisopropylsilyethynyl pentacene (TIPS-PEN) nanofibrils in a thin film of a crystalline polymer, poly(ε-caprolactone) (PCL). During spin-coating, TIPS-PEN molecules are locally extracted around the PCL grain boundaries and they crystallize into [010] direction forming long nanofibrils. Molecular weight of PCL and weight fraction (α) of TIPS-PEN in PCL matrix are key factors to the growth of nanofibrils. Long high-quality TIPS-PEN nanofibrils are obtained with high-molecular-weight PCL and at the α values in the range of 0.03-0.1. The long nanofibrils are used as an active layer in a field-effect organic transistor.

Widely tunable band gaps of graphdiyne: an ab initio study.

Functionalization of graphdiyne, a two-dimensional atomic layer of sp-sp(2) hybrid carbon networks, was investigated through first-principles calculations. Hydrogen or halogen atoms preferentially adsorb on sp-bonded carbon atoms rather than on sp(2)-bonded carbon atoms, forming sp(2)- or sp(3)-hybridization. The energy band gap of graphdiyne is increased from ~0.5 eV to ~5.2 eV through the hydrogenation or halogenation. Unlike graphene, segregation of adsorbing atoms is energetically unfavourable. Our results show that hydrogenation or halogenation can be utilized for modifying the electronic properties of graphdiyne for applications to nano-electronics and -photonics.

Optimizing an airborne mass-balance methodology for accurate emission rate quantification of industrial facilities: A case study of industrial facilities in South Korea.

Accurate estimation of emissions from industrial point sources is crucial in understanding the effectiveness of reduction efforts and establishing reliable emission inventories. In this study, we employ an airborne Chemical Ionization Mass Spectrometry (CIMS) instrument to quantify sulfur dioxide (SO2) emissions from prominent industrial facilities in South Korea, including power plants, a steel mill, and a petrochemical facility. Our analysis utilizes the box mass balance technique to derive SO2 emissions and associated uncertainty. We evaluate the interpolation methods between 2D kriging and 3D radial basis function. The results demonstrate that the total uncertainty of the box mass balance technique ranges from 5 % to 28 %, with an average of 20 %. Mixing ratio ground extrapolation from the lowest altitude of the airborne sampling to the ground emerges as the dominant source of uncertainty, followed by the determination of the boundary layer height. Adequate sampling at multiple altitudes is found to be essential in reducing the overall uncertainty by capturing the full extent of the plume. Furthermore, we assess the uncertainty of the single-height transect mass balance method commonly employed in previous studies. Our findings reveal an average precision of 47 % for this method, with the potential for overestimating emissions by up to 206 %. Samplings at fewer altitudes or with larger altitude gaps increase the risk of under-sampling and elevate method uncertainties. Therefore, this study provides a quantitative basis to evaluate previously airborne observational emission constraints.

Identifying the source rookery of green turtles (Chelonia mydas) found in feeding grounds around the Korean Peninsula.

Determining the genetic diversity and source rookeries of sea turtles collected from feeding grounds can facilitate effective conservation initiatives. To ascertain the genetic composition and source rookery, we examined a partial sequence of the mitochondrial control region (CR, 796 bp) of 40 green turtles (Chelonia mydas) collected from feeding grounds around the Korean Peninsula between 2014 and 2022. We conducted genetic and mixed-stock analyses (MSA) and identified 10 CR haplotypes previously reported in Japanese populations. In the haplotype network, six, three, and one haplotype(s) grouped with the Japan, Indo-Pacific, and Central South Pacific clades, respectively. The primary rookeries of the green turtles were two distantly remote sites, Ogasawara (OGA) and Central Ryukyu Island (CRI), approximately 1,300 km apart from each other. Comparing three parameters (season, maturity, and specific feeding ground), we noted that OGA was mainly associated with summer and the Jeju Sea, whereas CRI was with fall and the East (Japan) Sea ground. The maturity did not show a distinct pattern. Our results indicate that green turtles in the feeding grounds around the Korean Peninsula originate mainly from the Japan MU and have genetic origins in the Japan, Indo-Pacific, and Central South Pacific clades. Our results provide crucial insights into rookeries and MUs, which are the focus of conservation efforts of the Republic of Korea and potential parties to collaborate for green turtle conservation.

Convergence Gas Sensors with One-Dimensional Nanotubes and Pt Nanoparticles Based on Ultraviolet Photonic Energy for Room-Temperature NO2 Gas Sensing.

Zinc oxide (ZnO) is a promising material for nitrogen dioxide (NO2) gas sensors because of its nontoxicity, low cost, and small size. We fabricated one-dimensional (1D) and zero-dimensional (0D) convergence gas sensors activated via ultraviolet (UV) photonic energy to sense NO2 gas at room temperature. One-dimensional ZnO nanorod (ZNR)-based and ZnO nanotube (ZNT)-based gas sensors were synthesized using a simple hydrothermal method. All the sensors were tested under UV irradiation (365 nm) so that they could be operated at room temperature rather than a high temperature. In addition, we decorated 0D Pt nanoparticles (NPs) on the gas sensors to further improve their sensing responsivity. The NO2-sensing response of the ZNT/Pt NP convergence gas sensor was 2.93 times higher than that of the ZNR gas sensor. We demonstrated the complex effects of UV radiation on 1D ZnO nanostructures and 0D metal nanostructures in NO2 gas sensing.

Interfacial Capillary Spooling of Conductive Polyurethane-Silver Core-Sheath (PU@Ag) Microfibers for Highly Stretchable Interconnects.

Conductive fibers are core materials in textile electronics for the sustainable operation of devices under mechanical stimuli. Conventional polymer-metal core-sheath fibers were employed as stretchable electrical interconnects. However, their electrical conductivity is severely degraded by the rupture of metal sheaths at low strains. Because the core-sheath fibers are not intrinsically stretchable, designing a stretchable architecture of interconnects based on the fibers is essential. Herein, we introduce nonvolatile droplet-conductive microfiber arrays as stretchable interconnects by employing interfacial capillary spooling, motivated by the reversible spooling of capture threads in a spider web. Polyurethane (PU)-Ag core-sheath (PU@Ag) fibers were prepared by wet-spinning and thermal evaporation. When the fiber was placed on a silicone droplet, a capillary force was generated at their interface. The highly soft PU@Ag fibers were fully spooled within the droplet and reversibly uncoiled when a tensile force was applied. Without mechanical failures of the Ag sheaths, an excellent conductivity of 3.9 × 104 S cm-1 was retained at a strain of 1200% for 1000 spooling-uncoiling cycles. A light-emitting diode connected to a multiarray of droplet-PU@Ag fibers exhibited stable operation during spooling-uncoiling cycles.

Quality improvement of outpatient clinical chemistry tests through a novel middleware-laboratory information system solution.

OBJECTIVES: Rapid and accurate laboratory tests are essential to support clinical decision-making. Despite the various efforts to control quality in the laboratory, our outpatient chemistry turnaround time (TAT) has deteriorated since 2018. Moreover, these difficulties have accelerated further due to the COVID-19 pandemic. Therefore, we aimed to improve laboratory work efficiency by identifying and eliminating the causes of reduced laboratory work efficiency. DESIGN & METHODS: We surveyed to identify tasks that reduce work efficiency. Based on our survey, a new-concept of work assistance middleware linked to laboratory information system (LIS) was developed. The middleware supports test end-time prediction, automatic real-time TAT monitoring, and urgent test requests so that medical technologists can focus on their chemistry tests. The developed middleware was used for 6 months in laboratory and outpatient clinics, and its effectiveness was evaluated. RESULTS: The median TAT for outpatient chemistry tests was reduced by 6.6 min, from 72.4 min to 65.8 min. And not only did the maximum TAT for the sample decrease from 353 min to 214 min, but the proportion of samples exceeding the TAT target (120 min) also decreased by 77%; from 2.00% in 2010 (1,905 out of 94,989 samples) to 0.46% in 2021 (453 out of 98,117 samples). 2,199 samples were urgently requested through middleware, and they were processed about 15% faster than other samples, effectively performing urgent tests. The test end-time prediction showed an error of 8.6 min in the evaluation using the MAE (Mean Absolute Error) index. CONCLUSIONS: Through this study, the quality and efficiency of the laboratory were improved, and while reducing the workload of medical staff, it contributed to enhancing patient safety and satisfaction.

Electrochemiluminescence cell fabrication using TiO2 nanotube produced by anodic oxidation.

Electrochemiluminescence (ECL) cell using the electrode of TiO2 nanotube (NT) and Ru(ll) complex Ru(bpy)3(PF6)2 as a luminacence substance was fabricated. TiO2 NT were produced from the membrane of TiO2 NT arrays fabricated by anodic oxidation of approximately 100 microm thick Ti-plate. TiO2 NT arrays inject increasing number of electrons to the Ru(II) complex at the interface of TiO2 NTs. It allows the increasing exergonic oxidation/reduction reaction of Ru(II) complex. The ECL cell emits approximately 600 nm light in orange color. The cell structure is composed of a glass/F-doped SnO2(FTO)/TiO2 NT/Ru(II) complex in propylene carbonate/FTO/glass. The ECL efficiency of the cell consisting of the layer of TiO2 NT was approximately 255 cd/m2 at a bias of 4 V. The use of TiO2 NT increases ECL intensities by 5 times compared to the typical ECL cell without the use of TiO2 NT.

Nanocrystalline indium tin oxide fabricated via sol-gel combustion for electrochemical luminescence cells.

Nanoporous indium tin oxide (ITO) was synthesized via a sol-gel combustion hybrid method using Ketjenblack as a fuel. The effects of the sol-gel combustion conditions on the structures and morphology of the ITO particles were studied. The size of the nanoporous powder was found to be 20-30 nm in diameter. The layer of the nanoporous ITO electrode (-10 microm thickness) with large surface area (-360 m2/g) was fabricated for an electrochemical luminescence (ECL) cell. At 4 V bias, the ECL efficiency of the cell consisting of the nanoporous ITO layer was approximately 1050 cd/m2, which is significantly higher than the cell using only the FTO electrode (450 cd/m2). The nanoporous ITO layer was effective in increasing the ECL intensities.

Effects of substrate heating on the photovoltaic characteristics of dye-sensitized solar cells during two-step Ti film deposition by RF magnetron sputtering.

Nanoporous Ti metal film electrodes for use as photoanodes in dye-sensitized solar cells (DSSCs) were deposited directly on the nanoporous TiO2 layer using the two-step RF magnetron sputtering technique. The Ti film electrode replaces the transparent conducting oxide (TCO) layer. The effect of substrate heating during the deposition of the Ti film was studied to improve the porosity and columnar array of the film pores and the resulting cell efficiency. The porous Ti layer (-41 microm) with low sheet resistance (-1.7 omega/sq) was obtained by deposition at 250 degrees C. The porous Ti layer collects electrons from the TiO2 layer and allows the diffusion of I-/I3(-) through the holes. The DSSC efficiency (eta) using porous Ti layers with highly columnar structures was measured with the highest conversion efficiency of -5.77%; the other photovoltaic properties were ff: 0.76, V(oc): 0.72 V, and J(sc): 10.6 mA/cm2.

Initial work-up for cervical lymphadenopathy: back to basics.

The aim of this study was to evaluate the value of FNA, Tb-PCR, and CBC as an initial work-up protocol for cervical lymphadenopathy to reiterate the importance of CBC in terms of predicting the clinical course. In this consecutive case series, 158 patients with cervical lymphadenopathy were enrolled. All patients underwent FNA and CBC, with or without Tb-PCR. The validity of combined FNA ± Tb-PCR and CBC in the diagnosis of diseases requiring definitive treatment was evaluated. Final diagnoses were self-limiting disease in 110 (69.6%), malignancy in 19 (12.0%), and tuberculosis in 26 (16.5%). Sensitivity of FNA ± Tb-PCR was 66.7% and of added CBC profile was 97.9%. Patients with neutropenia or lymphocytosis were found to have a higher chance of spontaneous recovery than patients with a normal WBC profile. FNA and Tb-PCR were found to be important in patients aged more than 20. The results indicate that FNA, Tb-PCR, and CBC are basic and essential in initial work-up for cervical lymphadenopathy. In particular, CBC was found to aid in detecting critical diseases and predicting the likelihood of open biopsy in patients with negative FNA and Tb-PCR. Patient age was also found to be an important determinant of the work-up protocol.

Fake review identification and utility evaluation model using machine learning.

Due to the structural growth of e-commerce platforms, the frequency of exchange of opinions and the number of online reviews of platform participants related to products are increasing. However, given the growth of fake reviews, the corresponding growth in the quality of online reviews seems to be slow, at best. The number of cases of harm to retailers and customers caused by malicious false reviews is steadily increasing every year. In this context, it is becoming difficult for users to determine useful reviews amid a flood of information. As a result, the intrinsic value of online reviews that reduce uncertainty in pre-purchase decisions is blurred, and e-commerce platforms are on the verge of losing credibility and traffic. Through this study, we intend to present solutions related to review filtering and classification by constructing a model for judging the authenticity and usefulness of online reviews using machine learning.