nala: text mining natural language mutation mentions.

Bioinformatics. 2017 Feb 13;:

Authors: Miguel Cejuela J, Bojchevski A, Uhlig C, Bekmukhametov R, Kumar Karn S, Mahmuti S, Baghudana A, Dubey A, Satagopam VP, Rost B

PMID: 28200120 [PubMed - as supplied by publisher]

A MeSH-based text mining method for identifying novel prebiotics.

Medicine (Baltimore). 2016 Dec;95(49):e5585

Authors: Shan G, Lu Y, Min B, Qu W, Zhang C

Abstract
Prebiotics contribute to the well-being of their host by altering the composition of the gut microbiota. Discovering new prebiotics is a challenging and arduous task due to strict inclusion criteria; thus, highly limited numbers of prebiotic candidates have been identified. Notably, the large numbers of published studies may contain substantial information attached to various features of known prebiotics that can be used to predict new candidates. In this paper, we propose a medical subject headings (MeSH)-based text mining method for identifying new prebiotics with structured texts obtained from PubMed. We defined an optimal feature set for prebiotics prediction using a systematic feature-ranking algorithm with which a variety of carbohydrates can be accurately classified into different clusters in accordance with their chemical and biological attributes. The optimal feature set was used to separate positive prebiotics from other carbohydrates, and a cross-validation procedure was employed to assess the prediction accuracy of the model. Our method achieved a specificity of 0.876 and a sensitivity of 0.838. Finally, we identified a high-confidence list of candidates of prebiotics that are strongly supported by the literature. Our study demonstrates that text mining from high-volume biomedical literature is a promising approach in searching for new prebiotics.

PMID: 27930574 [PubMed - indexed for MEDLINE]

Chronic Non-infectious Uveitis in Patients with Juvenile Idiopathic Arthritis.

Ocul Immunol Inflamm. 2016 Aug;24(4):377-85

Authors: Kolomeyer AM, Tu Y, Miserocchi E, Ranjan M, Davidow A, Chu DS

Abstract
PURPOSE: To describe clinical findings and analyze treatment evolution of chronic, non-infectious uveitis in patients with juvenile idiopathic arthritis (JIA).
METHODS: A total of 82 patients (147 eyes) with JIA-related uveitis treated for ≥2 months were included (78% females; 79% bilateral uveitis; 74% anterior uveitis). Outcome measures were visual acuity (VA), inflammation control, side-effects, and surgical procedures.
RESULTS: Mean ± SD age at diagnosis was 4.9 ± 3.8 years; mean ± SD follow-up time was 8.7 ± 7.8 years. Mean VA did not significantly change throughout the study. Three (2%) eyes resulted in no light perception (NLP) vision. Thirty (37%) patients underwent 69 procedures. In total, 41 (50%) patients achieved inflammation control. TNF-α inhibitors were significantly associated with inflammation control. Seven (8.5%) patients stopped treatment due to side-effects.
CONCLUSIONS: JIA is a cause of significant ocular morbidity. TNF-α inhibitor use was associated with inflammation control. Prospective, randomized, double blind clinical trials in this regard are warranted.

PMID: 26902465 [PubMed - indexed for MEDLINE]

Mining peripheral arterial disease cases from narrative clinical notes using natural language processing.

J Vasc Surg. 2017 Feb 08;:

Authors: Afzal N, Sohn S, Abram S, Scott CG, Chaudhry R, Liu H, Kullo IJ, Arruda-Olson AM

Abstract
OBJECTIVE: Lower extremity peripheral arterial disease (PAD) is highly prevalent and affects millions of individuals worldwide. We developed a natural language processing (NLP) system for automated ascertainment of PAD cases from clinical narrative notes and compared the performance of the NLP algorithm with billing code algorithms, using ankle-brachial index test results as the gold standard.
METHODS: We compared the performance of the NLP algorithm to (1) results of gold standard ankle-brachial index; (2) previously validated algorithms based on relevant International Classification of Diseases, Ninth Revision diagnostic codes (simple model); and (3) a combination of International Classification of Diseases, Ninth Revision codes with procedural codes (full model). A dataset of 1569 patients with PAD and controls was randomly divided into training (n = 935) and testing (n = 634) subsets.
RESULTS: We iteratively refined the NLP algorithm in the training set including narrative note sections, note types, and service types, to maximize its accuracy. In the testing dataset, when compared with both simple and full models, the NLP algorithm had better accuracy (NLP, 91.8%; full model, 81.8%; simple model, 83%; P < .001), positive predictive value (NLP, 92.9%; full model, 74.3%; simple model, 79.9%; P < .001), and specificity (NLP, 92.5%; full model, 64.2%; simple model, 75.9%; P < .001).
CONCLUSIONS: A knowledge-driven NLP algorithm for automatic ascertainment of PAD cases from clinical notes had greater accuracy than billing code algorithms. Our findings highlight the potential of NLP tools for rapid and efficient ascertainment of PAD cases from electronic health records to facilitate clinical investigation and eventually improve care by clinical decision support.

PMID: 28189359 [PubMed - as supplied by publisher]

Challenges facing European agriculture and possible biotechnological solutions.

Crit Rev Biotechnol. 2016 Oct;36(5):875-83

Authors: Ricroch A, Harwood W, Svobodová Z, Sági L, Hundleby P, Badea EM, Rosca I, Cruz G, Salema Fevereiro MP, Marfà Riera V, Jansson S, Morandini P, Bojinov B, Cetiner S, Custers R, Schrader U, Jacobsen HJ, Martin-Laffon J, Boisron A, Kuntz M

Abstract
Agriculture faces many challenges to maximize yields while it is required to operate in an environmentally sustainable manner. In the present study, we analyze the major agricultural challenges identified by European farmers (primarily related to biotic stresses) in 13 countries, namely Belgium, Bulgaria, the Czech Republic, France, Germany, Hungary, Italy, Portugal, Romania, Spain, Sweden, UK and Turkey, for nine major crops (barley, beet, grapevine, maize, oilseed rape, olive, potato, sunflower and wheat). Most biotic stresses (BSs) are related to fungi or insects, but viral diseases, bacterial diseases and even parasitic plants have an important impact on yield and harvest quality. We examine how these challenges have been addressed by public and private research sectors, using either conventional breeding, marker-assisted selection, transgenesis, cisgenesis, RNAi technology or mutagenesis. Both national surveys and scientific literature analysis followed by text mining were employed to evaluate genetic engineering (GE) and non-GE approaches. This is the first report of text mining of the scientific literature on plant breeding and agricultural biotechnology research. For the nine major crops in Europe, 128 BS challenges were identified with 40% of these addressed neither in the scientific literature nor in recent European public research programs. We found evidence that the private sector was addressing only a few of these "neglected" challenges. Consequently, there are considerable gaps between farmer's needs and current breeding and biotechnology research. We also provide evidence that the current political situation in certain European countries is an impediment to GE research in order to address these agricultural challenges in the future. This study should also contribute to the decision-making process on future pertinent international consortia to fill the identified research gaps.

PMID: 26133365 [PubMed - indexed for MEDLINE]

Long-term operation performance and variation of substrate tolerance ability in an anammox attached film expanded bed (AAFEB) reactor.

Bioresour Technol. 2016 Jul;211:31-40

Authors: Zhang Y, Niu Q, Ma H, He S, Kubota K, Li YY

Abstract
An anammox attached film expanded bed (AAFEB) reactor was operated to study the long-term performance and the variation of substrate tolerance ability. The results indicated that the nitrogen loading potential (NLP) was significantly enhanced from 13.56gN·(L·d)(-)(1) to 20.95gN·(L·d)(-)(1) during the stable operation period. The inhibitory concentration of 10% (IC10) for free ammonia (FA), free nitrous acid (FNA) and SNinf (diluted substrate concentration) increased from 18mg/L, 12μgL(-1) and 370mgNL(-)(1) to 31mg/L, 19μgL(-1) and 670mgNL(-)(1), respectively. However, the substrate shock of 2500mgNL(-)(1) for 24h terribly weakened the treatment performance and substrate tolerance ability of the reactor. The results of batch tests indicated that the existence of lag phase made the AAFEB reactor more vulnerable to substrate variation. The SNinf was accurate to be used to monitor the reactor performance and should be maintained below 320mgNL(-)(1) to ensure the absolute stable operation.

PMID: 26995619 [PubMed - indexed for MEDLINE]

P316 New approaches for IBD management based on text mining of digitalised medical reports and latent class modelling.

J Crohns Colitis. 2017 Feb 01;11(suppl_1):S237-S238

Authors: Bergey F, Saccenti E, Jonkers D, van den Heuvel T, Jeuring S, Pierik M, Martins Dos Santos V

PMID: 28172927 [PubMed - in process]

Extraction of Left Ventricular Ejection Fraction Information from Various Types of Clinical Reports.

J Biomed Inform. 2017 Feb 02;:

Authors: Kim Y, Garvin JH, Goldstein MK, Hwang TS, Redd A, Bolton D, Heidenreich PA, Meystre SM

Abstract
Efforts to improve the treatment of congestive heart failure, a common and serious medical condition, include the use of quality measures to assess guideline-concordant care. The goal of this study is to identify left ventricular ejection fraction (LVEF) information from various types of clinical notes, and to then use this information for heart failure quality measurement. We analyzed the annotation differences between a new corpus of clinical notes from the Echocardiography, Radiology, and Text Integrated Utility package and other corpora annotated for natural language processing (NLP) research in the Department of Veterans Affairs. These reports contain varying degrees of structure. To examine whether existing LVEF extraction modules we developed in prior research improve the accuracy of LVEF information extraction from the new corpus, we created two sequence-tagging NLP modules trained with a new data set, with or without predictions from the existing LVEF extraction modules. We also conducted a set of experiments to examine the impact of training data size on information extraction accuracy. We found that less training data is needed when reports are highly structured, and that combining predictions from existing LVEF extraction modules improves information extraction when reports have less structured formats and a rich set of vocabulary.

PMID: 28163196 [PubMed - as supplied by publisher]

A New Data Representation Based on Training Data Characteristics to Extract Drug Name Entity in Medical Text.

Comput Intell Neurosci. 2016;2016:3483528

Authors: Sadikin M, Fanany MI, Basaruddin T

Abstract
One essential task in information extraction from the medical corpus is drug name recognition. Compared with text sources come from other domains, the medical text mining poses more challenges, for example, more unstructured text, the fast growing of new terms addition, a wide range of name variation for the same drug, the lack of labeled dataset sources and external knowledge, and the multiple token representations for a single drug name. Although many approaches have been proposed to overwhelm the task, some problems remained with poor F-score performance (less than 0.75). This paper presents a new treatment in data representation techniques to overcome some of those challenges. We propose three data representation techniques based on the characteristics of word distribution and word similarities as a result of word embedding training. The first technique is evaluated with the standard NN model, that is, MLP. The second technique involves two deep network classifiers, that is, DBN and SAE. The third technique represents the sentence as a sequence that is evaluated with a recurrent NN model, that is, LSTM. In extracting the drug name entities, the third technique gives the best F-score performance compared to the state of the art, with its average F-score being 0.8645.

PMID: 27843447 [PubMed - indexed for MEDLINE]

Cognitive-Enhancing Herbal Formulae in Korean Medicine: Identification of Candidates by Text Mining and Literature Review.

J Altern Complement Med. 2016 May;22(5):413-8

Authors: Pae SB, Yun BC, Han YK, Choi BT, Shin HK, Baek JU

Abstract
OBJECTIVE: The main aims of this study were to identify candidates for cognitive-enhancing herbal formulae from the Korean medicine literature and to obtain preliminary data that experimental and clinical researchers could use to develop new cognitive-enhancing drugs.
METHODS: The authors systematically searched for terms related to cognitive enhancement in Dongui Bogam (or Dongyi Baojian), a seminal Korean medicine book. They also reviewed the existing literature on the effects of candidates for cognitive-enhancing herbal formulae and their main constituents.
RESULTS AND CONCLUSIONS: Twenty-three candidates were selected for cognitive-enhancing herbal formulae and their main constituents. For 14 herbal formulae among the 23 candidates, on average 5.6 published research papers per herbal formula describing cognitive-enhancing effects were found. In addition, some published papers were identified for 5 main constituents most frequently used to make up the 23 candidates.

PMID: 27058606 [PubMed - indexed for MEDLINE]

Measuring adherence to a Choosing Wisely recommendation in a regional oncology clinic.

J Clin Oncol. 2016 Mar;34(7_suppl):196

Authors: Lyman GH, Kreizenbeck KL, Fedorenko CR, Alfiler A, Noble H, Kusnir-Wong T, Mohedano A, Stewart FM, Greer BE, Ramsey SD

Abstract
196 Background: Natural language processing (NLP) has the potential to significantly ease the burden of manual abstraction of unstructured electronic text when measuring adherence to national guidelines. We incorporated NLP into standard data processing techniques such as manual abstraction and database queries in order to more efficiently evaluate a regional oncology clinic's adherence to ASCO's Choosing Wisely colony stimulating factor (CSF) recommendation using clinical, billing, and cancer registry data.
METHODS: Database queries on the clinic's cancer registry yielded the study population of patients with stage II-IV breast, non-small cell lung (NSCL), and colorectal cancer. We manually abstracted chemotherapy regimens from paper prescription records. CSF orders were collected through queries on the clinic's facility billing data, when available; otherwise through a custom NLP program and manual abstraction of the electronic medical record. The NLP program was designed to identify clinical note text containing CSF information, which was then manually abstracted.
RESULTS: Out of 31,725 clinical notes for the eligible population, the NLP program identified 1,487 clinical notes with CSF-related language, effectively reducing the number of notes requiring abstraction by up to 95%. Between 1/1/2012-12/31/2014, adherence to the ASCO CW CSF recommendation at the regional oncology clinic was 89% for a population of 322 patients.
CONCLUSIONS: NLP significantly reduced the burden of manual abstraction by singling out relevant clinical text for abstractors. Abstraction is often necessary due to the complexity of data collection tasks or the use of paper records. However, NLP is a valuable addition to the suite of data processing techniques traditionally used to measure adherence to national guidelines.

PMID: 28152842 [PubMed - in process]

iPTMnet: Integrative Bioinformatics for Studying PTM Networks.

Methods Mol Biol. 2017;1558:333-353

Authors: Ross KE, Huang H, Ren J, Arighi CN, Li G, Tudor CO, Lv M, Lee JY, Chen SC, Vijay-Shanker K, Wu CH

Abstract
Protein post-translational modification (PTM) is an essential cellular regulatory mechanism, and disruptions in PTM have been implicated in disease. PTMs are an active area of study in many fields, leading to a wealth of PTM information in the scientific literature. There is a need for user-friendly bioinformatics resources that capture PTM information from the literature and support analyses of PTMs and their functional consequences. This chapter describes the use of iPTMnet ( http://proteininformationresource.org/iPTMnet/ ), a resource that integrates PTM information from text mining, curated databases, and ontologies and provides visualization tools for exploring PTM networks, PTM crosstalk, and PTM conservation across species. We present several PTM-related queries and demonstrate how they can be addressed using iPTMnet.

PMID: 28150246 [PubMed - in process]

Analysis of Protein Phosphorylation and Its Functional Impact on Protein-Protein Interactions via Text Mining of the Scientific Literature.

Methods Mol Biol. 2017;1558:213-232

Authors: Wang Q, Ross KE, Huang H, Ren J, Li G, Vijay-Shanker K, Wu CH, Arighi CN

Abstract
Post-translational modifications (PTMs) are one of the main contributors to the diversity of proteoforms in the proteomic landscape. In particular, protein phosphorylation represents an essential regulatory mechanism that plays a role in many biological processes. Protein kinases, the enzymes catalyzing this reaction, are key participants in metabolic and signaling pathways. Their activation or inactivation dictate downstream events: what substrates are modified and their subsequent impact (e.g., activation state, localization, protein-protein interactions (PPIs)). The biomedical literature continues to be the main source of evidence for experimental information about protein phosphorylation. Automatic methods to bring together phosphorylation events and phosphorylation-dependent PPIs can help to summarize the current knowledge and to expose hidden connections. In this chapter, we demonstrate two text mining tools, RLIMS-P and eFIP, for the retrieval and extraction of kinase-substrate-site data and phosphorylation-dependent PPIs from the literature. These tools offer several advantages over a literature search in PubMed as their results are specific for phosphorylation. RLIMS-P and eFIP results can be sorted, organized, and viewed in multiple ways to answer relevant biological questions, and the protein mentions are linked to UniProt identifiers.

PMID: 28150240 [PubMed - in process]

Dietary Nanosized Lactobacillus plantarum Enhances the Anticancer Effect of Kimchi on Azoxymethane and Dextran Sulfate Sodium-Induced Colon Cancer in C57BL/6J Mice.

J Environ Pathol Toxicol Oncol. 2016;35(2):147-59

Authors: Lee HA, Kim H, Lee KW, Park KY

Abstract
This study was undertaken to evaluate enhancement of the chemopreventive properties of kimchi by dietary nanosized Lactobacillus (Lab.)plantarum (nLp) in an azoxymethane (AOM)/dextran sulfate sodium (DSS)-induced colitis-associated colorectal cancer C57BL/6J mouse model. nLp is a dead, shrunken, processed form of Lab. Plantarum isolated from kimchi that is 0.5-1.0 µm in size. The results obtained showed that animals fed kimchi with nLp (K-nLp) had longer colons and lower colon weights/length ratios and developed fewer tumors than mice fed kimchi alone (K). In addition, K-nLp administration reduced levels of proinflammatory cytokine serum levels and mediated the mRNA and protein expressions of inflammatory, apoptotic, and cell-cycle markers to suppress inflammation and induce tumor-cell apoptosis and cell-cycle arrest. Moreover, it elevated natural killer-cell cytotoxicity. The study suggests adding nLp to kimchi could improve the suppressive effect of kimchi on AOM/DSS-induced colorectal cancer. These findings indicate nLp has potential use as a functional chemopreventive ingredient in the food industry.

PMID: 27481492 [PubMed - indexed for MEDLINE]

Performance of a Machine Learning Classifier of Knee MRI Reports in Two Large Academic Radiology Practices: A Tool to Estimate Diagnostic Yield.

AJR Am J Roentgenol. 2017 Jan 31;:1-4

Authors: Hassanpour S, Langlotz CP, Amrhein TJ, Befera NT, Lungren MP

Abstract
OBJECTIVE: The purpose of this study is to evaluate the performance of a natural language processing (NLP) system in classifying a database of free-text knee MRI reports at two separate academic radiology practices.
MATERIALS AND METHODS: An NLP system that uses terms and patterns in manually classified narrative knee MRI reports was constructed. The NLP system was trained and tested on expert-classified knee MRI reports from two major health care organizations. Radiology reports were modeled in the training set as vectors, and a support vector machine framework was used to train the classifier. A separate test set from each organization was used to evaluate the performance of the system. We evaluated the performance of the system both within and across organizations. Standard evaluation metrics, such as accuracy, precision, recall, and F1 score (i.e., the weighted average of the precision and recall), and their respective 95% CIs were used to measure the efficacy of our classification system.
RESULTS: The accuracy for radiology reports that belonged to the model's clinically significant concept classes after training data from the same institution was good, yielding an F1 score greater than 90% (95% CI, 84.6-97.3%). Performance of the classifier on cross-institutional application without institution-specific training data yielded F1 scores of 77.6% (95% CI, 69.5-85.7%) and 90.2% (95% CI, 84.5-95.9%) at the two organizations studied.
CONCLUSION: The results show excellent accuracy by the NLP machine learning classifier in classifying free-text knee MRI reports, supporting the institution-independent reproducibility of knee MRI report classification. Furthermore, the machine learning classifier performed well on free-text knee MRI reports from another institution. These data support the feasibility of multiinstitutional classification of radiologic imaging text reports with a single machine learning classifier without requiring institution-specific training data.

PMID: 28140627 [PubMed - as supplied by publisher]

Volar locking plate (VLP) versus non-locking plate (NLP) in the treatment of die-punch fractures of the distal radius, an observational study.

Int J Surg. 2016 Oct;34:142-147

Authors: Zhang X, Hu C, Yu K, Bai J, Tian D, Xu Y, Zhang B

Abstract
PURPOSE: This study aims to evaluate whether volar locking plate was superior over non-locking plate in the treatment of die-punch fractures of the distal radius.
METHODS: A total of 57 patients with closed die-punch fractures of the distal radius were included and analyzed. Of them, 32 were treated by non-locking plate (NLP) and the remaining 25 were treated by volar locking plate (VLP). Preoperative radiographs, computer tomographs and three-dimensional reconstruction, radiographs taken at immediate postoperation and at last follow-up were extracted and evaluated. Patients' electronic medical records were inquired and related demographic and medical data were documented. The documented contents were volar tilt, radial inclination, ulnar variance, grip strength, Disabilities of the Arm, Shoulder, and Hand (DASH) and visual analog scale (VAS) scores and complications.
RESULTS: VLP group demonstrated a significantly reduced radial subsidence of 1.5 mm (0.7 versus 2.2 mm), during the interval of bony union (P < 0.001), compared to NLP group. Larger proportion of patients (88% versus 62.5%) in VLP group gained acceptable joint congruity (step-off <2 mm) at the final follow-up (P = 0.037). No significant differences were observed between the groups in the measurements of volar tilt, radial inclination, DASH, VAS and grip strength recovery at the last follow-up. There was a trend of fewer overall complications (5/25 versus 10/32) and major complications that required surgery interventions (1/25 versus 4/32) in VLP than NLP groups, although the difference did not approach to significance (P = 0.339, 0.372).
CONCLUSIONS: VLP leaded to significantly better results of reduction maintainance and the final joint congruity than NLP, while reducing overall and major complications. However, the results should be treated in the context of limitations and the clinical significance of the difference required further studies to investigate.

PMID: 27593172 [PubMed - indexed for MEDLINE]

A tale of two countries: International comparison of online doctor reviews between China and the United States.

Int J Med Inform. 2017 Mar;99:37-44

Authors: Hao H, Zhang K, Wang W, Gao G

Abstract
BACKGROUND: Worldwide, patients have posted millions of online reviews for their doctors. The rich textual information in the online reviews holds the potential to generate insights into how patients' experience with their doctors differ across nations and how should we use them to improve our health service.
OBJECTIVE: We apply customized text mining techniques to compare online doctor reviews from China and the United States, in order to measure the systematic differences in patient reviews between the two countries, and assess the potential insights that can be derived from this large volume of online text data.
METHODS: We compare the textual reviews of obstetrics and gynecology (OBGYN) doctors from the two most popular online doctor rating websites in the U.S. and China, respectively: RateMDs.com and Haodf.com. We apply a customized text mining technique, Latent Dirichlet Allocation (LDA) topic modeling to identify the major topics in positive and negative reviews of those two countries. We then compare their similarities and differences.
RESULTS: Among the positive reviews, both Chinese and American patients talked about medical treatment, bedside manner, and appreciation/recommendation, but Chinese patients commented more about medical treatment while American patients focused more on recommendation. Also, reviews about bedside manner from Chinese patients were more related to doctors while on the American side, they were more about staff. This reflects the difference between the two countries' health systems. Further, among the negative reviews, both countries' patients talked about medical treatment, bedside manner, and logistics. However, Chinese patients focus more on the registration process, while American patients are more related to the staff, wait time, and insurance, which further shows the differences between the two nations' health systems.
CONCLUSIONS: Online doctor reviews contain valuable information that can generate insights on the similarities and differences of patient experience across nations. They are useful assets to assist healthcare consumers, providers, and administrators in moving toward a patient-centered care. In this age of big data, online doctor reviews can be a valuable source for international perspectives on healthcare systems.

PMID: 28118920 [PubMed - in process]

E-Cigarette Topics Shared by Medical Professionals: A Comparison of Tweets from the United States and United Kingdom.

Cyberpsychol Behav Soc Netw. 2017 Jan 24;:

Authors: Glowacki EM, Lazard AJ, Wilcox GB

Abstract
Medical professionals are now relying on social media platforms like Twitter to express their recommendations for the use or avoidance of products like electronic cigarettes (e-cigs), which may have long-term health consequences for users. The goal of this study is to compare how physicians from the United States and the United Kingdom talk about e-cigs on Twitter and identify the topics that these groups perceive as salient. Comparing tweets from the U.S. and U.K. will allow for a better understanding of how medical professionals from these countries differ in their attitudes toward e-cigs. This information can be also used to inform policies designed to regulate the use of e-cigs. Using a text-mining program, we analyzed approximately 3,800 original tweets sent by physicians from the U.S. and the U.K. within a 1-year time span (June 2015 through June 2016). The program clustered the tweets by topics, which allowed us to categorize the topics by importance. Both sets of tweets contained debates about the degree to which e-cigs pose a threat to health, but the U.S. tweets emphasized the dangers of e-cig use for teens, while the U.K. tweets focused more on the potential that e-cigs have to be used as a smoking cessation aid. Doctors are using Twitter to share timely information about the potential risks, benefits, and regulations associated with e-cigs. Evaluating these tweets allows researchers to collect information about topics that doctors find important and make comparisons about how medical professionals from the U.S. and the U.K. regard e-cigs.

PMID: 28118024 [PubMed - as supplied by publisher]

Optimal synthesis and design of the number of cycles in the leaching process for surimi production.

J Food Sci Technol. 2016 Dec;53(12):4325-4335

Authors: Reinheimer MA, Scenna NJ, Mussati SF

Abstract
Water consumption required during the leaching stage in the surimi manufacturing process strongly depends on the design and the number and size of stages connected in series for the soluble protein extraction target, and it is considered as the main contributor to the operating costs. Therefore, the optimal synthesis and design of the leaching stage is essential to minimize the total annual cost. In this study, a mathematical optimization model for the optimal design of the leaching operation is presented. Precisely, a detailed Mixed Integer Nonlinear Programming (MINLP) model including operating and geometric constraints was developed based on our previous optimization model (NLP model). Aspects about quality, water consumption and main operating parameters were considered. The minimization of total annual costs, which considered a trade-off between investment and operating costs, led to an optimal solution with lesser number of stages (2 instead of 3 stages) and higher volumes of the leaching tanks comparing with previous results. An analysis was performed in order to investigate how the optimal solution was influenced by the variations of the unitary cost of fresh water, waste treatment and capital investment.

PMID: 28115773 [PubMed - in process]

Efficient Exact Inference With Loss Augmented Objective in Structured Learning.

IEEE Trans Neural Netw Learn Syst. 2016 Aug 19;:

Authors: Bauer A, Nakajima S, Muller KR

Abstract
Structural support vector machine (SVM) is an elegant approach for building complex and accurate models with structured outputs. However, its applicability relies on the availability of efficient inference algorithms--the state-of-the-art training algorithms repeatedly perform inference to compute a subgradient or to find the most violating configuration. In this paper, we propose an exact inference algorithm for maximizing nondecomposable objectives due to special type of a high-order potential having a decomposable internal structure. As an important application, our method covers the loss augmented inference, which enables the slack and margin scaling formulations of structural SVM with a variety of dissimilarity measures, e.g., Hamming loss, precision and recall, Fβ-loss, intersection over union, and many other functions that can be efficiently computed from the contingency table. We demonstrate the advantages of our approach in natural language parsing and sequence segmentation applications.

PMID: 28113643 [PubMed - as supplied by publisher]