Prediction of immunochemotherapy response for diffuse large B-cell lymphoma using artificial intelligence digital pathology.

Diffuse large B-cell lymphoma (DLBCL) is a heterogeneous and prevalent subtype of aggressive non-Hodgkin lymphoma that poses diagnostic and prognostic challenges, particularly in predicting drug responsiveness. In this study, we used digital pathology and deep learning to predict responses to immunochemotherapy in patients with DLBCL. We retrospectively collected 251 slide images from 216 DLBCL patients treated with rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP), with their immunochemotherapy response labels. The digital pathology images were processed using contrastive learning for feature extraction. A multi-modal prediction model was developed by integrating clinical data and pathology image features. Knowledge distillation was employed to mitigate overfitting on gigapixel histopathology images to create a model that predicts responses based solely on pathology images. Based on the importance derived from the attention mechanism of the model, we extracted histological features that were considered key textures associated with drug responsiveness. The multi-modal prediction model achieved an impressive area under the ROC curve of 0.856, demonstrating significant associations with clinical variables such as Ann Arbor stage, International Prognostic Index, and bulky disease. Survival analyses indicated their effectiveness in predicting relapse-free survival. External validation using TCGA datasets supported the model's ability to predict survival differences. Additionally, pathology-based predictions show promise as independent prognostic indicators. Histopathological analysis identified centroblastic and immunoblastic features to be associated with treatment response, aligning with previous morphological classifications and highlighting the objectivity and reproducibility of artificial intelligence-based diagnosis. This study introduces a novel approach that combines digital pathology and clinical data to predict the response to immunochemotherapy in patients with DLBCL. This model shows great promise as a diagnostic and prognostic tool for clinical management of DLBCL. Further research and genomic data integration hold the potential to enhance its impact on clinical practice, ultimately improving patient outcomes.

Removal of intrahepatic bile duct stone could reduce the risk of cholangiocarcinoma: A single-center retrospective study in South Korea.

BACKGROUND: Intrahepatic duct (IHD) stones are among the most important risk factors for cholangiocarcinoma (CCC). Approximately 10% of patients with IHD stones develop CCC; however, there are limited studies regarding the effect of IHD stone removal on CCC development. AIM: To investigate the association between IHD stone removal and CCC development. METHODS: We retrospectively analyzed 397 patients with IHD stones at a tertiary referral center between January 2011 and December 2020. RESULTS: CCC occurred in 36 of the 397 enrolled patients. In univariate analysis, chronic hepatitis B infection (11.1% vs 3.0%, P = 0.03), carbohydrate antigen 19-9 (CA19-9, 176.00 vs 11.96 II/mL, P = 0.010), stone located in left or both lobes (86.1% vs 70.1%, P = 0.042), focal atrophy (52.8% vs 26.9%, P = 0.001), duct stricture (47.2% vs 24.9%, P = 0.004), and removal status of IHD stone (33.3% vs 63.2%, P < 0.001) were significantly different between IHD stone patients with and without CCC. In the multivariate analysis, CA19-9 > upper normal limit, carcinoembryonic antigen > upper normal limit, stones located in the left or both lobes, focal atrophy, and complete removal of IHD stones without recurrence were independent factors influencing CCC development. However, the type of removal method was not associated with CCC risk. CONCLUSION: Complete removal of IHD stones without recurrence could reduce CCC risk.

Resource analysis and modifications of quantum computing with noisy qubits for elliptic curve discrete logarithms.

We estimate the number of physical qubits and execution time by decomposing an implementation of Shor's algorithm for elliptic curve discrete logarithms into universal gate units at the logical level when surface codes are used. We herein also present modified quantum circuits for elliptic curve discrete logarithms and compare our results with those of the original quantum circuit implementations at the physical level. Through the analysis, we show that the use of more logical qubits in quantum algorithms does not always lead to the use of more physical qubits. We assumed using rotated surface code and logical qubits with all-to-all connectivity. The number of physical qubits and execution time are expressed in terms of bit length, physical gate error rate, and probability of algorithm failure. In addition, we compare our results with the number of physical qubits and execution time of Shor's factoring algorithm to assess the risk of attack by quantum computers in RSA and elliptic curve cryptography.

Higher sclerostin is associated with pulmonary hypertension in pre-dialysis end-stage kidney disease patients: a cross-sectional prospective observational cohort study.

BACKGROUND: Pulmonary hypertension (PH) is a complication of chronic kidney disease (CKD) that contributes to mortality. Sclerostin, a SOST gene product that reduces osteoblastic bone formation by inhibiting Wnt/ß-catenin signaling, is involved in arterial stiffness and CKD-bone mineral disease, but scanty evidence to PH. This study explored the relationship between sclerostin and PH in CKD 5, pre-dialysis end-stage kidney disease (ESKD) patients. METHODS: This cross-sectional prospective observational cohort study included 44 pre-dialysis ESKD patients between May 2011 and May 2015. Circulating sclerostin levels were measured using an enzyme-linked immunosorbent assay. PH was defined as an estimated pulmonary artery systolic pressure > 35 mmHg on echocardiography. RESULTS: Patients with higher sclerostin levels ≥ 218.18pmol/L had echocardiographic structural cardiac abnormalities, especially PH (P < 0.01). On multivariate logistic analysis, sclerostin over 218.19pmol/L was significantly associated with PH (odds ratio [OR], 41.14; 95% confidence interval [CI], 4.53-373.89, P < 0.01), but multivariate Cox regression analysis showed the systemic vascular calcification score over 1 point (Hazard ratio [HR] 11.49 95% CI 2.48-53.14, P = 0.002) and PH ([HR] 5.47, 95% CI 1.30-23.06, P = 0.02) were risk factors for all-cause mortality in pre-dialysis ESKD patients. CONCLUSIONS: Serum sclerostin and PH have a positive correlation in predialysis ESKD patients. The higher systemic vascular calcification score and PH have an association to increase all-cause mortality in pre-dialysis ESKD patients.

High-dimensional single photon based quantum secure direct communication using time and phase mode degrees.

Quantum secure direct communication (QSDC) can guarantee security using the characteristics of quantum mechanics even when a message is directly transmitted through a quantum channel without using a secret key. However, the transmission rate of the QSDC is limited by the dead time of a single photon detector (SPD) as well as channel loss over the distance. To overcome this limited transmission rate, we propose a high-dimensional single photon-based QSDC protocol that applies two optical degrees of freedom: time and phase state. First, an N-dimensional time and phase state generation method that considers the dead time is proposed to minimize the measurement loss of a transmitted message. Second, among the two types of quantum states, the phase state with relatively low measurement efficiency is used only for eavesdropping detection, and the time state is used for sending messages with differential delay time bin-based encoding techniques. Lastly, we propose an efficient method for measuring N-dimensional time and phase-based quantum states and recovering classical bit information. This study performs security analysis against various attacks, and verifies the transmission rate improvement effect through simulation. The result indicates that our proposal can guarantee higher security and transmission rates compared to the conventional DL04 QSDC.

Ensemble Deep Learning Model to Predict Lymphovascular Invasion in Gastric Cancer.

Lymphovascular invasion (LVI) is one of the most important prognostic factors in gastric cancer as it indicates a higher likelihood of lymph node metastasis and poorer overall outcome for the patient. Despite its importance, the detection of LVI(+) in histopathology specimens of gastric cancer can be a challenging task for pathologists as invasion can be subtle and difficult to discern. Herein, we propose a deep learning-based LVI(+) detection method using H&E-stained whole-slide images. The ConViT model showed the best performance in terms of both AUROC and AURPC among the classification models (AUROC: 0.9796; AUPRC: 0.9648). The AUROC and AUPRC of YOLOX computed based on the augmented patch-level confidence score were slightly lower (AUROC: -0.0094; AUPRC: -0.0225) than those of the ConViT classification model. With weighted averaging of the patch-level confidence scores, the ensemble model exhibited the best AUROC, AUPRC, and F1 scores of 0.9880, 0.9769, and 0.9280, respectively. The proposed model is expected to contribute to precision medicine by potentially saving examination-related time and labor and reducing disagreements among pathologists.

DLM-DTI: a dual language model for the prediction of drug-target interaction with hint-based learning.

The drug discovery process is demanding and time-consuming, and machine learning-based research is increasingly proposed to enhance efficiency. A significant challenge in this field is predicting whether a drug molecule's structure will interact with a target protein. A recent study attempted to address this challenge by utilizing an encoder that leverages prior knowledge of molecular and protein structures, resulting in notable improvements in the prediction performance of the drug-target interactions task. Nonetheless, the target encoders employed in previous studies exhibit computational complexity that increases quadratically with the input length, thereby limiting their practical utility. To overcome this challenge, we adopt a hint-based learning strategy to develop a compact and efficient target encoder. With the adaptation parameter, our model can blend general knowledge and target-oriented knowledge to build features of the protein sequences. This approach yielded considerable performance enhancements and improved learning efficiency on three benchmark datasets: BIOSNAP, DAVIS, and Binding DB. Furthermore, our methodology boasts the merit of necessitating only a minimal Video RAM (VRAM) allocation, specifically 7.7GB, during the training phase (16.24% of the previous state-of-the-art model). This ensures the feasibility of training and inference even with constrained computational resources.

A robust model training strategy using hard negative mining in a weakly labeled dataset for lymphatic invasion in gastric cancer.

Gastric cancer is a significant public health concern, emphasizing the need for accurate evaluation of lymphatic invasion (LI) for determining prognosis and treatment options. However, this task is time-consuming, labor-intensive, and prone to intra- and interobserver variability. Furthermore, the scarcity of annotated data presents a challenge, particularly in the field of digital pathology. Therefore, there is a demand for an accurate and objective method to detect LI using a small dataset, benefiting pathologists. In this study, we trained convolutional neural networks to classify LI using a four-step training process: (1) weak model training, (2) identification of false positives, (3) hard negative mining in a weakly labeled dataset, and (4) strong model training. To overcome the lack of annotated datasets, we applied a hard negative mining approach in a weakly labeled dataset, which contained only final diagnostic information, resembling the typical data found in hospital databases, and improved classification performance. Ablation studies were performed to simulate the lack of datasets and severely unbalanced datasets, further confirming the effectiveness of our proposed approach. Notably, our results demonstrated that, despite the small number of annotated datasets, efficient training was achievable, with the potential to extend to other image classification approaches used in medicine.

Metastatic pancreatic solitary fibrous tumor: A case report.

BACKGROUND: Solitary fibrous tumor (SFT) is a rare mesenchymal tumor that is most commonly found in the pleura but can also originate from non-pleural sites. Among the non-pleural localizations, the pancreas is extremely rare. In particular, metastasis to the pancreas from the central nervous system (CNS) is extremely rare, with only two cases reported so far. We report a case of recurrence in the pancreas 14 years after the initial complete surgical removal of a tumor in the CNS. CASE SUMMARY: A 68-year-old man with a past medical history of recurrent meningeal hemangiopericytoma, currently referred to as SFT, presented to the hospital with jaundice. The patient was first diagnosed with an 8cm-sized meningeal hemangiopericytoma fourteen years ago and underwent osteoplastic craniotomy. After 16 mo, imaging showed recurrence and he underwent gamma knife radiosurgery (GKRS). 2 years later, imaging showed recurrence again leading to a second GKRS. 5 years later, recurrence was again suspected leading to a second craniotomy. Since then 3 years had passed, and imaging showed a 3.5cm-sized mass lesion on the pancreatic head with obstruction of the pancreatic and bile ducts. Endosonography with fine-needle aspiration biopsy was done preoperatively and aided in the diagnosis of SFT. The patient underwent pylorus-preserving pancreaticoduodenectomy. Pathologic findings of the resected pancreatic specimen, consistent with the previously resected brain specimen, confirmed the diagnosis of SFT. CONCLUSION: The rarity and lack of knowledge about SFTs make suspecting and diagnosing this disease challenging. We believe that a report of metastatic pancreatic SFT from the CNS will contribute to a better understanding of this rare disease.

Evidence integration on health damage for humidifier disinfectant exposure and legal presumption of causation.

OBJECTIVES: Inhalation exposure to humidifier disinfectants has resulted to various types of health damages in Korea. To determine the epidemiological correlation necessary for presuming the legal causation, we aimed to develop a method to synthesize the entire evidence. METHODS: Epidemiological and toxicological studies are systematically reviewed. Target health problems are selected by criteria such as frequent complaints of claimants. Relevant epidemiologic studies are reviewed and the risk of bias and confidence level of the total evidence are evaluated. Toxicological literature reviews are conducted on three lines of evidence including hazard information, animal studies, and mechanistic studies, considering the source-to-exposure-to-outcome continuum. The confidence level of the body of evidence is then translated into the toxicological evidence levels for the causality between humidifier disinfectant exposure and health effects. Finally, the levels of epidemiological and toxicological evidence are synthesized. RESULTS: Under the Special Act revised in 2020, if the history of exposure and the disease occurred/worsened after exposure were approved, and the epidemiological correlation between the exposure and disease was verified, the legal causation is presumed unless the company proves the evidence against it. The epidemiological correlation can be verified through epidemiological investigations, health monitoring, cohort investigations and/or toxicological studies. It is not simply as statistical association as understood in judicial precedents, but a general causation established by the evidence as a whole, i.e., through weight-of-the-evidence approach. CONCLUSIONS: The weight-of-the-evidence approach differs from the conclusive single study approach and this systematic evidence integration can be used in presumption of causation.

New insight into biomagnification factor of mercury based on food web structure using stable isotopes of amino acids.

Although many attempts have been carried out to elaborate trophic magnification factor (TMF) and biomagnification factor (BMF), such as normalizing the concentration of pollutants and averaging diet sources, the uncertainty of the indexes still need to be improved to assess the bioaccumulation of pollutants. This study first suggests an improved BMF (i.e., BMF') applied to mercury bioaccumulation in freshwater fish from four sites before and after rainfall. The diet source and TP of each fish were identified using nitrogen stable isotope of amino acids (δ15NAAs) combined with bulk carbon stable isotope (δ13C). The BMF' was calculated normalizing with TP and diet contributions derived from MixSIAR. The BMF' values (1.3-27.2 and 1.2-27.8), which are representative of the entire food web, were generally higher than TMF (1.5-13.9 and 1.5-14.5) for both total mercury and methyl mercury, respectively. The BMF' implying actual mercury transfer pathway is more reliable index than relatively underestimated TMF for risk assessment. The ecological approach for BMF calculations provides novel insight into the behavior and trophic transfer of pollutants like mercury.

Diagnostic performance of the fibrosis-4 index and the NAFLD fibrosis score for screening at-risk individuals in a health check-up setting.

BACKGROUND: The fibrosis-4 index (FIB-4) and the NAFLD fibrosis score (NFS) have been used as noninvasive screening methods for advanced fibrosis in patients with NAFLD. However, their diagnostic performance has not been evaluated in at-risk individuals regardless of hepatic steatosis. This study evaluated the performance of the FIB-4 and NFS in at-risk groups of health check-up examinees at mass screening centers. METHODS: This retrospective, cross-sectional study included 8545 participants who underwent voluntary magnetic resonance elastography at a discounted fee during their regular health check-ups at 13 mass screening centers nationwide. The at-risk group was defined as those with any of the following conditions: NAFLD, 2 or more metabolic abnormalities, diabetes mellitus, or abnormal aminotransferase levels. A magnetic resonance elastography cutoff of ≥3.6 kPa was used to define conventional advanced fibrosis. RESULTS: According to the proposed criteria, the proportion of at-risk individuals was 67.4%-80.2% in the health check-up cohort without viral or alcohol-associated liver disease. The prevalence of individuals with advanced hepatic fibrosis in each at-risk group was ~2.3%-2.8% according to various criteria. It was higher in patients without NAFLD than in those with NAFLD. A total of 28.2%-39.6% of those in each at-risk group did not show hepatic steatosis on ultrasonography. The performance of FIB-4 for advanced fibrosis in the at-risk group was comparable with that in the NAFLD group. FIB-4 showed a better area under the receiver operating characteristic curve and sensitivity than NFS in the at-risk group. CONCLUSIONS: FIB-4 demonstrated superior performance compared with the NFS, and its performance in at-risk individuals was similar to that observed for patients with NAFLD.

Evolution of plume geometry, dilution and reactive mixing in porous media under highly transient flow fields at the surface water-groundwater interface.

Highly transient boundary conditions affect mixing of dissolved solutes in groundwater. An example of these transient boundary conditions occurs at the surface water-groundwater interface, where the water level in rivers can change rapidly due to the operation of hydropower plants, leading to a regime known as hydropeaking. Inspired by this phenomenon, this work studies at laboratory scale the effects of fluctuating surface water bodies on solute transport in aquifers. We performed flow-through experiments at two different flow velocities and under steady and transient flow conditions where a conservative tracer was injected in the system and its concentration measured with optical imaging methods. The experimental results were quantitatively interpreted with numerical simulations implementing a non-linear velocity-dependent dispersive transport model. We estimated plume dilution by computing the dilution index and its evolution as well as two key geometrical metrics of the transient plumes: the perimeter and the area. We further investigated reactive mixing and mixing enhancement considering mixing-controlled bimolecular reactions using different critical mixing ratios. In general, highly transient boundary conditions lead to a larger area, perimeter and plume dilution and the results show greater relative enhancement for the scenarios with low groundwater flow velocity. A linear relationship was observed between the evolution of the area and the dilution index of the plumes for the transient flow scenarios investigated. Considering reactive transport and mixing-limited reactions at the surface water-groundwater interface, we identified different dilution and reaction dominated regimes, characterized, respectively, by increasing and decreasing plume entropies at different mixing ratios of the reactants. Furthermore, reactive mixing was enhanced by transient flows leading to a faster degradation of contaminant plumes compared to corresponding steady flow conditions.

Integrative Korean Medicine Treatments for Traumatic Facial Palsy Following Mandibular Fracture: A Case Report and Literature Review.

Prior studies exploring the effectiveness of traditional Korean medicine (TKM) treatment for facial palsy have mainly focused on Bell's palsy, and there are few studies on the effectiveness of TKM treatments for traumatic facial palsy following mandibular fracture. The patient was a 24-year-old Korean man with left-sided facial paralysis following a left mandibular fracture. Surgery was performed for the fracture and the facial palsy was treated using conventional medicine (CM) treatments for approximately 3 months, but there was no improvement observed in the patient's condition. Subsequently, the patient underwent an integrative Korean medicine treatment regimen consisting of acupuncture, pharmacopuncture, cupping, moxibustion, and herbal medication for a duration of 2 months. After 2 months of treatments, the House-Brackmann facial grading scale changed from Ⅴ to II and Yanagihara's unweighted grading score increased from 9 to 34. This case presentation and previous studies of traumatic facial palsy using TKM treatment show that TKM treatment may be considered a complementary or alternative treatment method to CM treatment in patients with traumatic facial palsy. PROSPERO registration number: CRD42023445051.

The cross-linguistic comparison of perceptual strength norms for Korean, English and L2 English.

This study aimed to establish perceptual strength norms for 1,000 words in the languages of Korean, English, and L2 English, in order to investigate the similarity and difference across languages as well as the influence of the environment on semantic processing. The perceptual strength norms, which are a collection of word profiles that summarize how a word is experienced through different sensory modalities including the five common senses and interoception, provide a valuable tool for testing embodiment cognition theory. The results of this study demonstrated that language users had parallel sensory experiences with concepts, and that L2 learners were also able to associate their sensory experiences with linguistic concepts. Additionally, the results highlighted the importance of incorporating interoception as a sensory modality in the development of perceptual strength norms, as it had a negative correlation with both vision and concreteness. This study was the first to establish norms for Korean and L2 English and directly compare languages using the identical and translation-equivalent word list.

Diagnostic Performance of the Fibrosis-4 Index and Nonalcoholic Fatty Liver Disease Fibrosis Score in Lean Adults With Nonalcoholic Fatty Liver Disease.

IMPORTANCE: The diagnostic performance of the fibrosis-4 index (FIB-4) and nonalcoholic fatty liver disease (NAFLD) fibrosis score (NFS) for advanced fibrosis in lean patients with NAFLD is limited. OBJECTIVE: To evaluate the diagnostic performance of the FIB-4 and NFS in lean individuals with NAFLD. DESIGN, SETTING, AND PARTICIPANTS: This diagnostic study included adults with biopsy-proven NAFLD from 6 referral centers in Asia from 1995 to 2019. Cohorts were matched by age and sex between the lean and nonlean groups. All statistical analyses were executed from October 2022 to March 2023. MAIN OUTCOMES AND MEASURES: The diagnostic performance of the FIB-4 and NFS at the current cutoff for advanced hepatic fibrosis in lean (body mass index [BMI] below 23 [calculated as weight in kilograms divided by height in meters squared]) and nonlean (BMI above 23) patients were evaluated. RESULTS: A total of 1501 patients were included in analysis (mean [SD] age, 46.1 [16.4] years); 788 male (52.5%), 115 lean (7.7%), 472 (30.2%) Korean, 821 (48.7%) Japanese, and 341 (21.3%) Taiwanese. Among the age- and sex-matched cohort, the mean (SD) age was 52.3 (15.1) years and 41.2% (47 of 114) were male. The diagnostic performance and areas under the operating characteristic curve of the FIB-4 (lean, 0.807 vs nonlean, 0.743; P = .28) and NFS (lean, 0.790 vs nonlean, 0.755; P = .54) between the 2 groups were comparable in the age- and sex-matched cohort. The sensitivity and specificity of the NFS showed increasing and decreasing tendency according to the BMI quartiles (P for trend < .001), while those of the FIB-4 did not (P for trend = .05 and P = .20, respectively). Additionally, although the areas under the operating characteristic curve of the FIB-4 and NFS were not significantly different in the lean group (0.807 vs 0.790; P = .09), the sensitivity of the current NFS cutoff values was lower in the lean group than in that of FIB-4 (54.4% vs 81.8%; P = .03). CONCLUSIONS AND RELEVANCE: In this cohort study, the performance of the FIB-4 and NFS in diagnosing advanced fibrosis did not differ significantly between the 2 groups overall. However, in lean NAFLD, while the sensitivity for diagnosing advanced hepatic fibrosis remained reasonable at the current cutoff level, the sensitivity of NFS at the current cutoff was too low to be an adequate screening tool.

Risk Factors for Early- and Late-Onset Cholecystitis after Y-Configured Metal Stent Placement in Patients with Malignant Hilar Biliary Obstruction: A Single-Center Study.

This study evaluated the prevalence and risk factors of early- (within 7 days of placement) and late-onset (after 7 days of placement) cholecystitis after Y-configured metal stent placement. Between June 2005 and August 2020, 109 patients who had been treated with Y-configured metal stents for malignant hilar obstruction were enrolled in the study. We retrospectively analyzed the potential risk factors for post-stent cholecystitis. The presence of diabetes (p = 0.042), the length of the common part of the Y-stent (p = 0.017), filling of the gallbladder with contrast medium during the procedure (p = 0.040), and tumor invasion of the cystic duct accompanied by filling the gallbladder with contrast medium during metal stent placement (p = 0.001) were identified as important risk factors. In cases of late-onset cholecystitis, stent obstruction (p = 0.004) and repeated endoscopic procedures due to stent malfunction (p = 0.024) were significant risk factors. In the multivariate logistic regression analysis, significant risk factors were the length of the common part of the Y-stent (p = 0.032) in early-onset cholecystitis and stent obstruction (p = 0.007) in late-onset cholecystitis. This study demonstrated that early-onset cholecystitis may occur in patients according to the length of the common portion of the Y-stent. In contrast, late-onset cholecystitis may occur in patients with stent obstruction.

Diagnostic Performance of Noninvasive Tests for Advanced Hepatic Fibrosis in Young Age Population.

BACKGROUND & AIMS: Most noninvasive tests (NITs) for hepatic fibrosis are designed for middle-aged patients with chronic liver disease. We compared the diagnostic performance of major NITs (aspartate aminotransferase-to-platelet ratio index [APRI], Fibrosis-4 index, and nonalcoholic fatty liver disease fibrosis score) for a community-based cohort. METHODS: This cross-sectional study analyzed 8775 participants who underwent magnetic resonance elastography at community health check-up centers. Advanced hepatic fibrosis (≥F3) was defined by magnetic resonance elastography thresholds of 3.6 kPa. The diagnostic performance of 3 NITs was evaluated according to the etiology of liver disease, sex, metabolic syndrome, obesity, and increased aminotransferase levels in 4 age groups. RESULTS: The APRI generally showed the best area under the receiver operating characteristic curve in patients aged 45 years or younger, and it was statistically significant in patients with chronic viral hepatitis and alcoholic fatty liver disease (P < .043). The best APRI cut-off value for detecting advanced hepatic fibrosis was 0.4, with a sensitivity and specificity of 75.8% and 73.5%, respectively, in the community-based cohort. The APRI showed balanced sensitivity and specificity across all age groups, whereas the other metrics showed low sensitivity in those aged <45 and low specificity in those >65 years. CONCLUSIONS: The APRI showed better sensitivity and negative predictive value than the Fibrosis-4 index and the nonalcoholic fatty liver disease fibrosis score in community-based populations with mixed etiology, and, thus, can be performed as the primary test in young adults (age, ≤45 y).

Induction Motor Fault Diagnosis Using Support Vector Machine, Neural Networks, and Boosting Methods.

Induction motors are robust and cost effective; thus, they are commonly used as power sources in various industrial applications. However, due to the characteristics of induction motors, industrial processes can stop when motor failures occur. Thus, research is required to realize the quick and accurate diagnosis of faults in induction motors. In this study, we constructed an induction motor simulator with normal, rotor failure, and bearing failure states. Using this simulator, 1240 vibration datasets comprising 1024 data samples were obtained for each state. Then, failure diagnosis was performed on the acquired data using support vector machine, multilayer neural network, convolutional neural network, gradient boosting machine, and XGBoost machine learning models. The diagnostic accuracies and calculation speeds of these models were verified via stratified K-fold cross validation. In addition, a graphical user interface was designed and implemented for the proposed fault diagnosis technique. The experimental results demonstrate that the proposed fault diagnosis technique is suitable for diagnosing faults in induction motors.