Identifying symptom clusters and temporal interconnections in patients with lung tumors after CT-guided microwave ablation: A network analysis.

PURPOSE: To explore symptom clusters and interrelationships using a network analysis approach among symptoms in patients with lung tumors who underwent computed tomography (CT)-guided microwave ablation (MWA). METHODS: A longitudinal study was conducted, and 196 lung tumor patients undergoing MWA were recruited and were measured at 24 h, 48 h, and 72 h after MWA. The Chinese version of the MD Anderson Symptom Inventory and the Revised Lung Cancer Module were used to evaluate symptoms. Network analyses were performed to explore the symptom clusters and interrelationships among symptoms. RESULTS: Four stable symptom communities were identified within the networks. Distress, weight loss, and chest tightness were the central symptoms. Distress, and weight loss were also the most key bridge symptoms, followed by cough. Three symptom networks were temporally stable in terms of symptom centrality, global connectivity, and network structure. CONCLUSION: Our findings identified the central symptoms, bridge symptoms, and the stability of symptom networks of patients with lung tumors after MWA. These network results will have important implications for future targeted symptom management intervention development. Future research should focus on developing precise interventions for targeting central symptoms and bridge symptoms to promote patients' health.

Oxidative Transformation of Nafion-Related Fluorinated Ether Sulfonates: Comparison with Legacy PFAS Structures and Opportunities of Acidic Persulfate Digestion for PFAS Precursor Analysis.

The total oxidizable precursor (TOP) assay has been extensively used for detecting PFAS pollutants that do not have analytical standards. It uses hydroxyl radicals (HO•) from the heat activation of persulfate under alkaline pH to convert H-containing precursors to perfluoroalkyl carboxylates (PFCAs) for target analysis. However, the current TOP assay oxidation method does not apply to emerging PFAS because (i) many structures do not contain C-H bonds for HO• attack and (ii) the transformation products are not necessarily PFCAs. In this study, we explored the use of classic acidic persulfate digestion, which generates sulfate radicals (SO4-•), to extend the capability of the TOP assay. We examined the oxidation of Nafion-related ether sulfonates that contain C-H or -COO-, characterized the oxidation products, and quantified the F atom balance. The SO4-• oxidation greatly expanded the scope of oxidizable precursors. The transformation was initiated by decarboxylation, followed by various spontaneous steps, such as HF elimination and ester hydrolysis. We further compared the oxidation of legacy fluorotelomers using SO4-• versus HO•. The results suggest novel product distribution patterns, depending on the functional group and oxidant dose. The general trends and strategies were also validated by analyzing a mixture of 100000- or 10000-fold diluted aqueous film-forming foam (containing various fluorotelomer surfactants and organics) and a spiked Nafion precursor. Therefore, (1) the combined use of SO4-• and HO• oxidation, (2) the expanded list of standard chemicals, and (3) further elucidation of SO4-• oxidation mechanisms will provide more critical information to probe emerging PFAS pollutants.

Effect of multidisciplinary collaborative empowerment education on psychological distress and quality of life in patients with colorectal cancer undergoing chemotherapy.

PURPOSE: To evaluate the effects of multidisciplinary collaborative empowerment education on psychological distress and quality of life (QoL) in patients with colorectal cancer undergoing chemotherapy. METHODS: A quasi-experimental study was conducted using repeated measures at pre- and post-intervention in the fourth chemotherapy cycle. Sixty patients with colorectal cancer aged 36-84 years were allocated to the intervention and control groups. The intervention group received multidisciplinary empowerment education, while the control group received routine health education. Psychological distress involving depression and anxiety symptoms was assessed using The Kessler Psychological Distress Scale (K10) and QoL was measured using The European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTCQLQ-C30). Repeated-measures analysis of variance was used to examine intervention effects. Statistical analyses were performed using the SPSS software (version 26.0). RESULTS: Psychological distress was considerably lower and QoL was considerably better in patients following multidisciplinary empowerment education in the intervention group than those in the control group. In addition, psychological distress significantly decreased and QoL improved in the intervention group compared to baseline. CONCLUSION: Multidisciplinary collaborative empowerment education was effective in improving the psychological distress and QoL among patients with colorectal cancer undergoing chemotherapy. These findings suggest that the establishment of multidisciplinary collaborative empowerment education might be considered as an innovative means of clinical patient education during combination chemotherapy to improve health outcomes in patients with colorectal cancer. However, our results should be interpreted with caution because of the small sample size. Further validation in a larger sample or randomized controlled design is necessary in the future.

Phase separation enabled silver nano-array.

The surface-supported silver nanoparticles have been studied and applied in various applications. Many unique nanostructures have been introduced into this field to improve the functionalities of the surfaces depending on application purposes. We created featured silver nano-array surfaces by utilizing the solvent-mediated phase transition on the surface grafted with poly (acrylic) acids polymer chains and taking advantage of the low temperature of argon gas discharged plasma as a reducing agent. The applied solvents and grafted polymer chain densities affected the phase transition and thus determined the outcome of surface nano-array patterns. However, the total loaded silver ions on the surface affected silver nano-array structures at the sub-micron levels. The featured silver patterned surfaces made in the optimal conditions present a favorable surface-enhanced Raman spectroscopy enhancement as well as recyclability for detection re-usage. This novel method prepares tunable silver nanopatterned surfaces and provides a new approach to various potential applications.

Plasma-initiated graft polymerization of carbon nanoparticles as nano-based drug delivery systems.

Plasma-initiated free radical polymerization was used to engineer carbon nanoparticles (CNPs) with tailored chemical and physical properties. Following surface modification, CNPs were loaded with a highly effective anti-infection agent called metal-free Russian propolis ethanol extract (MFRPEE), thus, creating nano-based drug delivery systems (NBDDSs). The loading of MFRPEE onto grafted CNPs occurred naturally through both electrostatic interactions and hydrogen bonding. When constructed under optimal experimental conditions, the NBDDSs were stable under physiologic conditions, and demonstrated enhanced anti-biofilm activity when compared with free MFRPEE. Mechanistic studies revealed that the enhanced anti-infectious activity of the NBDDSs was attributed to the modified surface chemistry of grafted CNPs. More specifically, the overall positive surface charge on grafted CNPs, which stems from quaternary ammonium polymer brushes covalently bound to the CNPs, provides NBDDSs with the ability to specifically target negatively charged components of biofilms. When studying the release profile of MFRPEE from the modified CNPs, acidic components produced by a biofilm triggered the release of MFRPEE bound to the NBDDS. Once in its free form, the anti-infectious properties of MFRPEE became activated and damaged the extracellular polymeric matrix (EPM) of the biofilm. Once the architecture of the biofilm became compromised, the EPM was no longer capable of protecting the bacteria encapsulated within the biofilm from the anti-infectious agent. Consequently, exposure of bacteria to MFRPEE led to bacterial cell death and biofilm inactivation. The results obtained from this study begin to examine the potential application of NBDDSs for the treatment of healthcare-associated infections (HCAIs).

Rapid Synthesis of Metal Nanoparticles Using Low-Temperature, Low-Pressure Argon Plasma Chemistry and Self-Assembly.

The synthesis of metal nanoparticles has become a priority for the advancement of nanotechnology. In attempts to create these nanoparticles, several different methods: chemistry, physics, and biology, have all been used. In this study, we report the reduction of cations using argon plasma chemistry to produce nanoparticles of gold (AuNPs), silver (AgNPs), and copper (CuNPs). Although other groups have used plasma-reduction methods to synthesize metal nanoparticles from their cation counterparts, these approaches often require plasma|liquid state interactions, high temperature, specific combinations of gases, and extended treatment times (>10 minutes), for which only specific cations (noble or non-noble) may be reduced. As a result, we have developed a non-thermal, low-pressure argon-plasma|solid state approach for the reduction of both noble and non-noble cations. More specifically, when 50-µL droplets of 2-mM solutions of gold(III) chloride, silver nitrate, or copper(II) sulfate are exposed to vacuum, they undergo an evaporation process. As the pressure in the chamber decreases to 220 mtorr, the droplets become completely evaporated, leaving behind a metal precursor. Nucleation and growth studies reveal that when the metal precursors of gold(III) chloride, silver nitrate, and copper(II) sulfate are treated with 80 watts of argon plasma for 5, 60, and 150 seconds, respectively, nanoparticles could be synthesized with efficiency rates of upwards of 98%. The size of nanoparticles synthesized in this work was studied using Scanning Electron Microscopy, and the scattering properties of the nanoparticles was studied using UV/Vis spectroscopy. Transmission Electron Microscopy with corresponding elemental analysis was also very useful in confirming the identity of the synthesized nanoparticles. The results from this study reveal that we have synthesized metal nanoparticles with distinct chemical and physical properties. Scanning Electron Microscopy depicts AgNPs with a round-shape and diameters from 40 - 80 nm, while AuNPs were hexagonal, with sizes from 40 - 80 nm, and CuNPs were rod-shaped, with dimensions 40 by 160 nm. Our findings demonstrate that the argon plasma approach used in this study is a rapid, green, and versatile reduction method for the synthesis of both noble and non-noble metal nanoparticles.

Active Substances from Callicarpa nudiflora Exert Anti-Cervicitis Effects and Regulate NLRP3-Associated Inflammation.

Luohuazizhu suppository is a Traditional Chinese Medicine used in clinic to treat cervicitis, which is prepared from Callicarpa nudiflora Hook. et Arn (C. nudiflora), an herbal Chinese medicine named Luohuazizhu. This study aimed to figure out the active constituents of C. nudiflora and the potential mechanism for its anti-cervicitis effect. The ethanol extract in C. nudiflora (CNE) and the different fractions of CNE extracted by petroleum ether (CNE-p), dichloromethane (CNE-d), and n-butanol (CNE-b) were tested in vivo for their anti-cervicitis effects. Then the isolated compounds from the CNE-p were tested in vitro for their anti-inflammatory activities. The results displayed that CNE-p, CNE-d, and CNE-b exhibited adequate anti-cervicitis effects, with CNE-p showing the highest efficacy. Further experiment demonstrated that CNE-p could significantly inhibit the expression of NLRP3 in vitro. Six diterpenoids obtained from the CNE-p showed the ability to regulate inflammatory factor levels in vitro. Among these compounds, compounds 1 (callicarpic acid A) and 2 (syn-3,4-seco-12S-hydroxy-15,16-epoxy-4(18),8(17),3(16),14(15)-labdatetraen-3-oic acid) were the most effective agents, and they also inhibited the expression level of NLRP3 in vitro. The results confirmed that C. nudiflora has significant anti-cervicitis effects and the diterpenoids were most likely to be its active components. These data provide scientific support for the clinic usage of Luohuazizhu suppository and the development of new agents in treating cervicitis.

The deep learning model combining CT image and clinicopathological information for predicting ALK fusion status and response to ALK-TKI therapy in non-small cell lung cancer patients.

PURPOSE: This study aimed to investigate the deep learning model (DLM) combining computed tomography (CT) images and clinicopathological information for predicting anaplastic lymphoma kinase (ALK) fusion status in non-small cell lung cancer (NSCLC) patients. MATERIALS AND METHODS: Preoperative CT images, clinicopathological information as well as the ALK fusion status from 937 patients in three hospitals were retrospectively collected to train and validate the DLM for the prediction of ALK fusion status in tumors. Another cohort of patients (n = 91) received ALK tyrosine kinase inhibitor (TKI) treatment was also included to evaluate the value of the DLM in predicting the clinical outcomes of the patients. RESULTS: The performances of the DLM trained only by CT images in the primary and validation cohorts were AUC = 0.8046 (95% CI 0.7715-0.8378) and AUC = 0.7754 (95% CI 0.7199-0.8310), respectively, while the DLM trained by both CT images and clinicopathological information exhibited better performance for the prediction of ALK fusion status (AUC = 0.8540, 95% CI 0.8257-0.8823 in the primary cohort, p < 0.001; AUC = 0.8481, 95% CI 0.8036-0.8926 in the validation cohort, p < 0.001). In addition, the deep learning scores of the DLMs showed significant differences between the wild-type and ALK infusion tumors. In the ALK-target therapy cohort (n = 91), the patients predicted as ALK-positive by the DLM showed better performance of progression-free survival than the patients predicted as ALK-negative (16.8 vs. 7.5 months, p = 0.010). CONCLUSION: Our findings showed that the DLM trained by both CT images and clinicopathological information could effectively predict the ALK fusion status and treatment responses of patients. For the small size of the ALK-target therapy cohort, larger data sets would be collected to further validate the performance of the model for predicting the response to ALK-TKI treatment.

Simultaneously learning affinity matrix and data representations for machine fault diagnosis.

Recently, preserving geometry information of data while learning representations have attracted increasing attention in intelligent machine fault diagnosis. Existing geometry preserving methods require to predefine the similarities between data points in the original data space. The predefined affinity matrix, which is also known as the similarity matrix, is then used to preserve geometry information during the process of representations learning. Hence, the data representations are learned under the assumption of a fixed and known prior knowledge, i.e., similarities between data points. However, the assumed prior knowledge is difficult to precisely determine the real relationships between data points, especially in high dimensional space. Also, using two separated steps to learn affinity matrix and data representations may not be optimal and universal for data classification. In this paper, based on the extreme learning machine autoencoder (ELM-AE), we propose to learn the data representations and the affinity matrix simultaneously. The affinity matrix is treated as a variable and unified in the objective function of ELM-AE. Instead of predefining and fixing the affinity matrix, the proposed method adjusts the similarities by taking into account its capability of capturing the geometry information in both original data space and non-linearly mapped representation space. Meanwhile, the geometry information of original data can be preserved in the embedded representations with the help of the affinity matrix. Experimental results on several benchmark datasets demonstrate the effectiveness of the proposed method, and the empirical study also shows it is an efficient tool on machine fault diagnosis.

Polydopamine-Decorated Orlistat-Loaded Hollow Capsules with an Enhanced Cytotoxicity against Cancer Cell Lines.

Orlistat, an FDA-approved antiobesity drug, has recently been shown to have anticancer effects. However, orlistat is extremely hydrophobic with low absorption. Therefore, new approaches are needed to effectively deliver orlistat for cancer therapy. Herein, we developed a fast and simple method to use polydopamine-coated hollow capsule (PHC) as a drug nanocarrier for enhancing the therapeutic effects of orlistat. Orlistat-loaded PHC had an average size of 200 nm, which was characterized by using dynamic light scattering and scanning electron microscope. Furthermore, the polydopamine layer provided an excellent control of orlistat release because it was extremely sensitive to pH values. The cellular uptake and cytotoxicity experiments were performed to show that orlistat packaged in PHC could be endocytosed into cells and then significantly improved the cytotoxic activity against cancer cell lines in a short time compared with free orlistat. Moreover, dynamic study of cell membrane lysis was performed by staining with the LIVE/DEAD kit to demonstrate the cancer-killing mechanism. The size of the cell surface area has also been proven to be a key parameter which affected drug efficacy. Taken all together, these results present that orlistat-loaded PHC is a very promising formula for cancer treatments.

Label-free detection of Staphylococcus aureus bacteria using long-period fiber gratings with functional polyelectrolyte coatings.

Highly sensitive long-period fiber gratings (LPFG) was developed for label-free and rapid detection of Staphylococcus aureus (S. aureus). Specifically, the LPFG was functionalized with antibody and nanopitted polyelectrolyte coatings to facilitate bacterial adhesion and thus enhance the sensitivity of bacteria detection. The kinetics of S. aureus adhesion on functional coatings were tracked by surface morphology evolution and time-resolved resonance wavelength shift of the coated LPFG at a flow rate of 30 µl/ml and 37 °C in the concentration range of 104-108 colony forming unit (CFU)/ml. S. aureus detection at concentrations as low as 224 CFU/ml can be achieved within a short time span of 30 min. The LPFG-based biosensor can be readily adapted to a variety of biophotonic platforms, for applications such as food safety inspection, environmental monitoring, clinical diagnostics, and medical applications.

Adhesive Gold Nanoparticles for Easy and Controlled Surface Coating.

Gold nanoparticles (Au NPs) are one of the most important nanomaterials due to their unique properties and broad applications. Among these applications, decorating Au NPs on universal surfaces is highly desired. Herein, we report adhesive Au NPs functionalized by borated dopamine dithiocarbamate. Such Au NPs are nonreactive in colloidal solution but can be activated at an acidic pH to produce adhesive Au NPs and initiate spontaneous surface coating through deprotected catechol-mediated reactions. Easy and controllable surface coating was achieved on materials with distinguished chemical and physical properties because of the high reactivity of catechol. Adhesive Au NPs represent new surface coating method with wide application potentials.

The dataset of scanning electron microscope images of silver nanoparticles formed in situ by dopamine chemistry.

The mussel inspired chemistry of dopamine oxidation to form polydopamine (PDA) and in situ reduction of metal ions in solution to form metal nanoparticles have widely opened the application of metal nanoparticles surface modification technology. This article contains the dataset of the scanning electron microscope (SEM) images of silver nanoparticles coated on polyethylene terephthalate (PET) films utilizing dopamine chemistry alone or combined with polyvinylpyrrolidone or glucose. The Ag NPs formed in various environments present round, cubic, or triangle shape. Mendeley Data, http://dx.doi.org/10.17632/bjjrt2dwbn.1.

Graphene-kaolin composite sponge for rapid and riskless hemostasis.

Kaolin is an effective and safe hemostatic agent for hemostasis. However, its ontic powder is difficult to use in actual practice. To develop a wieldy and powerful hemostat, composite strategy is usually a good choice. Herein, we developed a graphene-kaolin composite sponge (GKCS), synthesized with graphene oxide sheets, linker molecules and kaolin powders through a facile hydrothermal reaction. SEM observations support that GKCS has a porous structure, and EDS mapping further confirms that kaolin powders are embedded in graphene sheets. Once GKCS is exposed to bleeding, plasma is quickly absorbed inside the sponge, meanwhile blood cells are gathered at the interface. The gathered blood cells are in favor of accelerating clotting due to multi stimulations, including concentration, surface charge and activation of hemostatic factors, originating from both kaolin powders and graphene sponge. As a result, GKCS could stop bleeding in approximately 73 s in rabbit artery injury test. Besides, cytotoxicity and hemolysis assessments highlight that GKCS has a good biocompatibility. These remarkable properties suggest that GKCS is a potential riskless hemostatic agent for trauma treatment.

Downregulation of DNMT3A by miR-708-5p Inhibits Lung Cancer Stem Cell-like Phenotypes through Repressing Wnt/ß-catenin Signaling.

Purpose: Lung cancer is the leading cause of cancer-related death in the world, and emerging evidences suggest that lung cancer stem cells (CSC) are associated with its poor prognosis, tumor recurrence, and therapy resistance. Here we reveal a novel role for miR-708-5p in inhibiting lung CSC-like features.Experimental Design: Phenotypic effects of miR-708-5p on the lung CSC-like properties were examined by in vitro sphere formation assay and in xenografted animal models. Immunoblotting, dual luciferase reporter, and immunocytochemistry were performed to determine the target of miR-708-5p. DNA methylation of CDH1 promoter region was tested using bisulfate sequencing. Genome-wide miRNA sequencing data of 990 patients from The Cancer Genome Atlas (TCGA) dataset and 148 patients from China cohort were analyzed to excavate the pathogenic implications of miR-708-5p.Results: Expression of miR-708-5p inhibits the CSC traits of NSCLC cells in vitro while antagonizing miR-708-5p promotes tumorigenesis in vivo miR-708-5p directly suppresses the translation of DNMT3A, which results in a substantial reduction of global DNA methylation and the upregulated expression of tumor suppressor CDH1. The upregulation of CDH1 decreased the activity of Wnt/ß-catenin signaling and then impaired the stemness characteristics of NSCLC cells. Clinically, patients with high miR-708-5p expression show significantly better survival and lower recurrence. Furthermore, miR-708-5p has a promising potential to apply to differentiating histologic subtypes in NSCLC.Conclusions: Our findings support that miR-708-5p suppresses NSCLC initiation, development, and stemness through interfering DNMT3A-dependent DNA methylation. miR-708-5p may function as a novel diagnostic and prognostic biomarker in NSCLC. Clin Cancer Res; 24(7); 1748-60. ©2017 AACR.

Are Russian propolis ethanol extracts the future for the prevention of medical and biomedical implant contaminations?

BACKGROUND: Most studies reveal that the mechanism of action of propolis against bacteria is functional rather than structural and is attributed to a synergism between the compounds in the extracts. HYPOTHESIS/PURPOSE: Propolis is said to inhibit bacterial adherence, division, inhibition of water-insoluble glucan formation, and protein synthesis. However, it has been shown that the mechanism of action of Russian propolis ethanol extracts is structural rather than functional and may be attributed to the metals found in propolis. If the metals found in propolis are removed, cell lysis still occurs and these modified extracts may be used in the prevention of medical and biomedical implant contaminations. STUDY DESIGN: The antibacterial activity of metal-free Russian propolis ethanol extracts (MFRPEE) on two biofilm forming bacteria: penicillin-resistant Staphylococcus aureus and Escherichia coli was evaluated using MTT and a Live/Dead staining technique. Toxicity studies were conducted on mouse osteoblast (MC-3T3) cells using the same viability assays. METHODS: In the MTT assay, biofilms were incubated with MTT at 37°C for 30min. After washing, the purple formazan formed inside the bacterial cells was dissolved by SDS and then measured using a microplate reader by setting the detecting and reference wavelengths at 570nm and 630nm, respectively. Live and dead distributions of cells were studied by confocal laser scanning microscopy. RESULTS: Complete biofilm inactivation was observed when biofilms were treated for 40h with 2µg/ml of MFRPEE. Results indicate that the metals present in propolis possess antibacterial activity, but do not have an essential role in the antibacterial mechanism of action. Additionally, the same concentration of metals found in propolis samples, were toxic to tissue cells. Comparable to samples with metals, metal free samples caused damage to the cell membrane structures of both bacterial species, resulting in cell lysis. CONCLUSION: Results suggest that the structural mechanism of action of Russian propolis ethanol extracts stem predominate from the organic compounds. Further studies revealed drastically reduced toxicity to mammalian cells when metals were removed from Russian propolis ethanol extracts, suggesting a potential for medical and biomedical applications.

Self-Polymerization of Dopamine in Acidic Environments without Oxygen.

An alkaline environment and the presence of oxygen are essential requirements for dopamine polymerization. In this study, we are the first to demonstrate the self-polymerization of dopamine through plasma-activated water (PAW) under acidic environments (pH < 5.5). Resulting poly(dopamine) (PDA) was characterized using Nanosizer, SEM, FTIR, UV-vis, 1H NMR, and fluorescence spectrophotometers and proved to have similar physical and chemical properties to those polymerized under a basic condition, except that the PDA particles formed in PAW were more stable and hardly aggregated at varied pHs. The PAW polymerization method avoids alkaline solutions and the presence of oxygen and thus extends the applications of dopamine polymerization, particularly in biomedical and pharmaceutical sciences.

MicroRNA-708-5p acts as a therapeutic agent against metastatic lung cancer.

MicroRNAs (miRNAs) have recently been recognized as targets for anti-metastatic therapy against cancer malignancy. Development of effective miRNA mediated therapies remains a challenge to both basic research and clinical practice. Here we presented the evidence for a miR-708-5p mediated replacement therapy against metastatic lung cancer. Expression of miR-708-5p was substantially reduced in metastatic lung cancer samples and cancer cell lines when compared to non-metastatic counterparts. Expression of the miRNA suppressed cell survival and metastasis in vitro through its direct target p21, and inhibited the PI3K/AKT pathway and stem cell-like characteristics of lung cancer cells. Systemic administration of this miRNA in a mouse model of NSCLC using polyethylenimine (PEI)-mediated delivery of unmodified miRNA mimics induced tumor specific apoptosis. It also effectively protected the tested animals from developing metastatic malignancy without causing any observed toxicity. The findings strongly support miR-708-5p as a novel and effective therapeutic agent against metastatic malignancy of non-small cell lung cancer.

Manifold ranking based scoring system with its application to cardiac arrest prediction: A retrospective study in emergency department patients.

BACKGROUND: The recently developed geometric distance scoring system has shown the effectiveness of scoring systems in predicting cardiac arrest within 72h and the potential to predict other clinical outcomes. However, the geometric distance scoring system predicts scores based on only local structure embedded by the data, thus leaving much room for improvement in terms of prediction accuracy. METHODS: We developed a novel scoring system for predicting cardiac arrest within 72h. The scoring system was developed based on a semi-supervised learning algorithm, manifold ranking, which explores both the local and global consistency of the data. System evaluation was conducted on emergency department patients׳ data, including both vital signs and heart rate variability (HRV) parameters. Comparison of the proposed scoring system with previous work was given in terms of sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV). RESULTS: Out of 1025 patients, 52 (5.1%) met the primary outcome. Experimental results show that the proposed scoring system was able to achieve higher area under the curve (AUC) on both the balanced dataset (0.907 vs. 0.824) and the imbalanced dataset (0.774 vs. 0.734) compared to the geometric distance scoring system. CONCLUSIONS: The proposed scoring system improved the prediction accuracy by utilizing the global consistency of the training data. We foresee the potential of extending this scoring system, as well as manifold ranking algorithm, to other medical decision making problems. Furthermore, we will investigate the parameter selection process and other techniques to improve performance on the imbalanced dataset.

Discriminative clustering via extreme learning machine.

Discriminative clustering is an unsupervised learning framework which introduces the discriminative learning rule of supervised classification into clustering. The underlying assumption is that a good partition (clustering) of the data should yield high discrimination, namely, the partitioned data can be easily classified by some classification algorithms. In this paper, we propose three discriminative clustering approaches based on Extreme Learning Machine (ELM). The first algorithm iteratively trains weighted ELM (W-ELM) classifier to gradually maximize the data discrimination. The second and third methods are both built on Fisher's Linear Discriminant Analysis (LDA); but one approach adopts alternative optimization, while the other leverages kernel k-means. We show that the proposed algorithms can be easily implemented, and yield competitive clustering accuracy on real world data sets compared to state-of-the-art clustering methods.