ResearchPad - intelligence https://www.researchpad.co Default RSS Feed en-us © 2020 Newgen KnowledgeWorks <![CDATA[The impact of knowledge transfer performance on the artificial intelligence industry innovation network: An empirical study of Chinese firms]]> https://www.researchpad.co/article/elastic_article_15759 As a core driving force of the most recent round of industrial transformation, artificial intelligence has triggered significant changes in the world economic structure, profoundly changed our life and way of thinking, and achieved an overall leap in social productivity. This paper aims to examine the effect of knowledge transfer performance on the artificial intelligence industry innovation network and the path artificial intelligence enterprises can take to promote sustainable development through knowledge transfer in the above context. First, we construct a theoretical hypothesis and conceptual model of the innovation network knowledge transfer mechanism within the artificial intelligence industry. Then, we collect data from questionnaires distributed to Chinese artificial intelligence enterprises that participate in the innovation network. Moreover, we empirically analyze the impact of innovation network characteristics, organizational distance, knowledge transfer characteristics, and knowledge receiver characteristics on knowledge transfer performance and verify the hypotheses proposed in the conceptual model. The results indicate that innovation network centrality and organizational culture distance have a significant effect on knowledge transfer performance, with influencing factors including network scale, implicit knowledge transfer, receiver’s willingness to receive, and receiver’s capacity to absorb knowledge. For sustainable knowledge transfer performance on promoting Chinese artificial intelligence enterprises innovation, this paper finally delivers valuable insights and suggestions.

]]>
<![CDATA[ToyArchitecture: Unsupervised learning of interpretable models of the environment]]> https://www.researchpad.co/article/elastic_article_15730 Research in Artificial Intelligence (AI) has focused mostly on two extremes: either on small improvements in narrow AI domains, or on universal theoretical frameworks which are often uncomputable, or lack practical implementations. In this paper we attempt to follow a big picture view while also providing a particular theory and its implementation to present a novel, purposely simple, and interpretable hierarchical architecture. This architecture incorporates the unsupervised learning of a model of the environment, learning the influence of one’s own actions, model-based reinforcement learning, hierarchical planning, and symbolic/sub-symbolic integration in general. The learned model is stored in the form of hierarchical representations which are increasingly more abstract, but can retain details when needed. We demonstrate the universality of the architecture by testing it on a series of diverse environments ranging from audio/visual compression to discrete and continuous action spaces, to learning disentangled representations.

]]>
<![CDATA[OtoMatch: Content-based eardrum image retrieval using deep learning]]> https://www.researchpad.co/article/elastic_article_14747 Acute infections of the middle ear are the most commonly treated childhood diseases. Because complications affect children’s language learning and cognitive processes, it is essential to diagnose these diseases in a timely and accurate manner. The prevailing literature suggests that it is difficult to accurately diagnose these infections, even for experienced ear, nose, and throat (ENT) physicians. Advanced care practitioners (e.g., nurse practitioners, physician assistants) serve as first-line providers in many primary care settings and may benefit from additional guidance to appropriately determine the diagnosis and treatment of ear diseases. For this purpose, we designed a content-based image retrieval (CBIR) system (called OtoMatch) for normal, middle ear effusion, and tympanostomy tube conditions, operating on eardrum images captured with a digital otoscope. We present a method that enables the conversion of any convolutional neural network (trained for classification) into an image retrieval model. As a proof of concept, we converted a pre-trained deep learning model into an image retrieval system. We accomplished this by changing the fully connected layers into lookup tables. A database of 454 labeled eardrum images (179 normal, 179 effusion, and 96 tube cases) was used to train and test the system. On a 10-fold cross validation, the proposed method resulted in an average accuracy of 80.58% (SD 5.37%), and maximum F1 score of 0.90 while retrieving the most similar image from the database. These are promising results for the first study to demonstrate the feasibility of developing a CBIR system for eardrum images using the newly proposed methodology.

]]>
<![CDATA[Innovative machine learning approach and evaluation campaign for predicting the subjective feeling of work-life balance among employees]]> https://www.researchpad.co/article/elastic_article_14744 At present, many researchers see hope that artificial intelligence, machine learning in particular, will improve several aspects of the everyday life for individuals, cities and whole nations alike. For example, it has been speculated that the so-called machine learning could soon relieve employees of part of the duties, which may improve processes or help to find the most effective ways of performing tasks. Consequently, in the long run, it would help to enhance employees’ work-life balance. Thus, workers’ overall quality of life would improve, too. However, what would happen if machine learning as such were employed to try and find the ways of achieving work-life balance? This is why the authors of the paper decided to utilize a machine learning tool to search for the factors that influence the subjective feeling of one’s work-life balance. The possible results could help to predict and prevent the occurrence of work-life imbalance in the future. In order to do so, the data provided by an exceptionally sizeable group of 800 employees was utilised; it was one of the largest sample groups in similar studies in Poland so far. Additionally, this was one of the first studies where so many employees had been analysed using an artificial neural network. In order to enable replicability of the study, the specific setup of the study and the description of the dataset are provided. Having analysed the data and having conducted several experiments, the correlations between some factors and work-life balance have indeed been identified: it has been found that the most significant was the relation between the feeling of balance and the actual working hours; shifting it resulted in the tool predicting the switch from balance to imbalance, and vice versa. Other factors that proved significant for the predicted WLB are the amount of free time a week the employee has for themselves, working at weekends only, being self-employed and the subjective assessment of one’s financial status. In the study the dataset gets balanced, the most important features are selected with the selectKbest algorithm, an artificial neural network of 2 hidden layers with 50 and 25 neurons, ReLU and ADAM is constructed and trained on 90% of the dataset. In tests, it predicts WLB based on the prepared dataset and selected features with 81% accuracy.

]]>
<![CDATA[Neural networks for open and closed Literature-based Discovery]]> https://www.researchpad.co/article/elastic_article_14696 Literature-based Discovery (LBD) aims to discover new knowledge automatically from large collections of literature. Scientific literature is growing at an exponential rate, making it difficult for researchers to stay current in their discipline and easy to miss knowledge necessary to advance their research. LBD can facilitate hypothesis testing and generation and thus accelerate scientific progress. Neural networks have demonstrated improved performance on LBD-related tasks but are yet to be applied to it. We propose four graph-based, neural network methods to perform open and closed LBD. We compared our methods with those used by the state-of-the-art LION LBD system on the same evaluations to replicate recently published findings in cancer biology. We also applied them to a time-sliced dataset of human-curated peer-reviewed biological interactions. These evaluations and the metrics they employ represent performance on real-world knowledge advances and are thus robust indicators of approach efficacy. In the first experiments, our best methods performed 2-4 times better than the baselines in closed discovery and 2-3 times better in open discovery. In the second, our best methods performed almost 2 times better than the baselines in open discovery. These results are strong indications that neural LBD is potentially a very effective approach for generating new scientific discoveries from existing literature. The code for our models and other information can be found at: https://github.com/cambridgeltl/nn_for_LBD.

]]>
<![CDATA[Deep learning assisted detection of glaucomatous optic neuropathy and potential designs for a generalizable model]]> https://www.researchpad.co/article/elastic_article_14620 To evaluate ways to improve the generalizability of a deep learning algorithm for identifying glaucomatous optic neuropathy (GON) using a limited number of fundus photographs, as well as the key features being used for classification.MethodsA total of 944 fundus images from Taipei Veterans General Hospital (TVGH) were retrospectively collected. Clinical and demographic characteristics, including structural and functional measurements of the images with GON, were recorded. Transfer learning based on VGGNet was used to construct a convolutional neural network (CNN) to identify GON. To avoid missing cases with advanced GON, an ensemble model was adopted in which a support vector machine classifier would make final classification based on cup-to-disc ratio if the CNN classifier had low-confidence score. The CNN classifier was first established using TVGH dataset, and then fine-tuned by combining the training images of TVGH and Drishti-GS datasets. Class activation map (CAM) was used to identify key features used for CNN classification. Performance of each classifier was determined through area under receiver operating characteristic curve (AUC) and compared with the ensemble model by diagnostic accuracy.ResultsIn 187 TVGH test images, the accuracy, sensitivity, and specificity of the CNN classifier were 95.0%, 95.7%, and 94.2%, respectively, and the AUC was 0.992 compared to the 92.8% accuracy rate of the ensemble model. For the Drishti-GS test images, the accuracy of the CNN, the fine-tuned CNN and ensemble model was 33.3%, 80.3%, and 80.3%, respectively. The CNN classifier did not misclassify images with moderate to severe diseases. Class-discriminative regions revealed by CAM co-localized with known characteristics of GON.ConclusionsThe ensemble model or a fine-tuned CNN classifier may be potential designs to build a generalizable deep learning model for glaucoma detection when large image databases are not available. ]]> <![CDATA[Prediction of hypertension, hyperglycemia and dyslipidemia from retinal fundus photographs via deep learning: A cross-sectional study of chronic diseases in central China]]> https://www.researchpad.co/article/elastic_article_14561 Retinal fundus photography provides a non-invasive approach for identifying early microcirculatory alterations of chronic diseases prior to the onset of overt clinical complications. Here, we developed neural network models to predict hypertension, hyperglycemia, dyslipidemia, and a range of risk factors from retinal fundus images obtained from a cross-sectional study of chronic diseases in rural areas of Xinxiang County, Henan, in central China. 1222 high-quality retinal images and over 50 measurements of anthropometry and biochemical parameters were generated from 625 subjects. The models in this study achieved an area under the ROC curve (AUC) of 0.880 in predicting hyperglycemia, of 0.766 in predicting hypertension, and of 0.703 in predicting dyslipidemia. In addition, these models can predict with AUC>0.7 several blood test erythrocyte parameters, including hematocrit (HCT), mean corpuscular hemoglobin concentration (MCHC), and a cluster of cardiovascular disease (CVD) risk factors. Taken together, deep learning approaches are feasible for predicting hypertension, dyslipidemia, diabetes, and risks of other chronic diseases.

]]>
<![CDATA[Disentangling sequential from hierarchical learning in Artificial Grammar Learning: Evidence from a modified Simon Task]]> https://www.researchpad.co/article/elastic_article_14558 In this paper we probe the interaction between sequential and hierarchical learning by investigating implicit learning in a group of school-aged children. We administered a serial reaction time task, in the form of a modified Simon Task in which the stimuli were organised following the rules of two distinct artificial grammars, specifically Lindenmayer systems: the Fibonacci grammar (Fib) and the Skip grammar (a modification of the former). The choice of grammars is determined by the goal of this study, which is to investigate how sensitivity to structure emerges in the course of exposure to an input whose surface transitional properties (by hypothesis) bootstrap structure. The studies conducted to date have been mainly designed to investigate low-level superficial regularities, learnable in purely statistical terms, whereas hierarchical learning has not been effectively investigated yet. The possibility to directly pinpoint the interplay between sequential and hierarchical learning is instead at the core of our study: we presented children with two grammars, Fib and Skip, which share the same transitional regularities, thus providing identical opportunities for sequential learning, while crucially differing in their hierarchical structure. More particularly, there are specific points in the sequence (k-points), which, despite giving rise to the same transitional regularities in the two grammars, support hierarchical reconstruction in Fib but not in Skip. In our protocol, children were simply asked to perform a traditional Simon Task, and they were completely unaware of the real purposes of the task. Results indicate that sequential learning occurred in both grammars, as shown by the decrease in reaction times throughout the task, while differences were found in the sensitivity to k-points: these, we contend, play a role in hierarchical reconstruction in Fib, whereas they are devoid of structural significance in Skip. More particularly, we found that children were faster in correspondence to k-points in sequences produced by Fib, thus providing an entirely new kind of evidence for the hypothesis that implicit learning involves an early activation of strategies of hierarchical reconstruction, based on a straightforward interplay with the statistically-based computation of transitional regularities on the sequences of symbols.

]]>
<![CDATA[ECG-based prediction algorithm for imminent malignant ventricular arrhythmias using decision tree]]> https://www.researchpad.co/article/elastic_article_14548 Spontaneous prediction of malignant ventricular arrhythmia (MVA) is useful to avoid delay in rescue operations. Recently, researchers have developed several algorithms to predict MVA using various features derived from electrocardiogram (ECG). However, there are several unresolved issues regarding MVA prediction such as the effect of number of ECG features on a prediction remaining unclear, possibility that an alert for occurring MVA may arrive very late and uncertainty in the performance of the algorithm predicting MVA minutes before onset. To overcome the aforementioned problems, this research conducts an in-depth study on the number and types of ECG features that are implemented in a decision tree classifier. In addition, this research also investigates an algorithm’s execution time before the occurrence of MVA to minimize delays in warnings for MVA. Lastly, this research aims to study both the sensitivity and specificity of an algorithm to reveal the performance of MVA prediction algorithms from time to time. To strengthen the results of analysis, several classifiers such as support vector machine and naive Bayes are also examined for the purpose of comparison study. There are three phases required to achieve the objectives. The first phase is literature review on existing relevant studies. The second phase deals with design and development of four modules for predicting MVA. Rigorous experiments are performed in the feature selection and classification modules. The results show that eight ECG features with decision tree classifier achieved good prediction performance in terms of execution time and sensitivity. In addition, the results show that the highest percentage for sensitivity and specificity is 95% and 90% respectively, in the fourth 5-minute interval (15.1 minutes–20 minutes) that preceded the onset of an arrhythmia event. Such results imply that the fourth 5-minute interval would be the best time to perform prediction.

]]>
<![CDATA[Inference on dengue epidemics with Bayesian regime switching models]]> https://www.researchpad.co/article/elastic_article_14505 Dengue, a mosquito-borne infectious disease caused by the dengue viruses, is present in many parts of the tropical and subtropical regions of the world. All four serotypes of dengue viruses are endemic in Singapore, an equatorial city-state. Frequent outbreaks occur, sometimes leading to national epidemics. However, few studies have attempted to characterize breakpoints which precede large rises in dengue case counts. In this paper, Bayesian regime switching (BRS) models were employed to infer epidemic and endemic regimes of dengue transmissions, each containing regime specific processes which drive the growth and decline of dengue cases, estimated using a custom built multi-move Gibbs sampling algorithm. Assessments against various baseline showed that BRS performs better in characterizing dengue transmissions. The dengue regimes estimated by BRS are characterized by their persistent nature. Next, climate analysis showed no short nor long term associations between classified regimes with climate. Lastly, fitting BRS to simulated disease data generated from a mechanistic model, we showed links between disease infectivity and regimes classified using BRS. The model proposed could be applied to other localities and diseases under minimal data requirements where transmission counts over time are collected.

]]>
<![CDATA[Improvement of electrocardiographic diagnostic accuracy of left ventricular hypertrophy using a Machine Learning approach]]> https://www.researchpad.co/article/elastic_article_14491 The electrocardiogram (ECG) is the most common tool used to predict left ventricular hypertrophy (LVH). However, it is limited by its low accuracy (<60%) and sensitivity (30%). We set forth the hypothesis that the Machine Learning (ML) C5.0 algorithm could optimize the ECG in the prediction of LVH by echocardiography (Echo) while also establishing ECG-LVH phenotypes. We used Echo as the standard diagnostic tool to detect LVH and measured the ECG abnormalities found in Echo-LVH. We included 432 patients (power = 99%). Of these, 202 patients (46.7%) had Echo-LVH and 240 (55.6%) were males. We included a wide range of ventricular masses and Echo-LVH severities which were classified as mild (n = 77, 38.1%), moderate (n = 50, 24.7%) and severe (n = 75, 37.1%). Data was divided into a training/testing set (80%/20%) and we applied logistic regression analysis on the ECG measurements. The logistic regression model with the best ability to identify Echo-LVH was introduced into the C5.0 ML algorithm. We created multiple decision trees and selected the tree with the highest performance. The resultant five-level binary decision tree used only six predictive variables and had an accuracy of 71.4% (95%CI, 65.5–80.2), a sensitivity of 79.6%, specificity of 53%, positive predictive value of 66.6% and a negative predictive value of 69.3%. Internal validation reached a mean accuracy of 71.4% (64.4–78.5). Our results were reproduced in a second validation group and a similar diagnostic accuracy was obtained, 73.3% (95%CI, 65.5–80.2), sensitivity (81.6%), specificity (69.3%), positive predictive value (56.3%) and negative predictive value (88.6%). We calculated the Romhilt-Estes multilevel score and compared it to our model. The accuracy of the Romhilt-Estes system had an accuracy of 61.3% (CI95%, 56.5–65.9), a sensitivity of 23.2% and a specificity of 94.8% with similar results in the external validation group. In conclusion, the C5.0 ML algorithm surpassed the accuracy of current ECG criteria in the detection of Echo-LVH. Our new criteria hinge on ECG abnormalities that identify high-risk patients and provide some insight on electrogenesis in Echo-LVH.

]]>
<![CDATA[The role of frontal and parietal cortex in the performance of gifted and average adolescents in a mental rotation task]]> https://www.researchpad.co/article/elastic_article_14471 Visual-spatial abilities are usually neglected in academic settings, even though several studies have shown that their predictive power in science, technology, engineering, and mathematics domains exceeds that of math and verbal ability. This neglect means that many spatially talented youths are not identified and nurtured, at a great cost to society. In the present work, we aim to identify behavioral and electrophysiological markers associated with visual spatial-ability in intellectually gifted adolescents (N = 15) compared to age-matched controls (N = 15). The participants performed a classic three-dimensional mental rotation task developed by Shepard and Metzler (1971) [33] while event-related potentials were measured in both frontal and parietal regions of interest. While response time was similar in the two groups, gifted subjects performed the test with greater accuracy. There was no indication of interhemispheric asymmetry of ERPs over parietal regions in both groups, although interhemispheric differences were observed in the frontal lobes. Moreover, intelligence quotient and working memory measures predicted variance in ERP’s amplitude in the right parietal and frontal hemispheres. We conclude that while gifted adolescents do not display a different pattern of electroencephalographic activity over the parietal cortex while performing the mental rotation task, their performance is correlated with the amplitude of ERPs in the frontal cortex during the execution of this task.

]]>
<![CDATA[Insight into the protein solubility driving forces with neural attention]]> https://www.researchpad.co/article/elastic_article_13832 The solubility of proteins is a crucial biophysical aspect when it comes to understanding many human diseases and to improve the industrial processes for protein production. Due to its relevance, computational methods have been devised in order to study and possibly optimize the solubility of proteins. In this work we apply a deep-learning technique, called neural attention to predict protein solubility while “opening” the model itself to interpretability, even though Machine Learning models are usually considered black boxes. Thank to the attention mechanism, we show that i) our model implicitly learns complex patterns related to emergent, protein folding-related, aspects such as to recognize β-amyloidosis regions and that ii) the N-and C-termini are the regions with the highes signal fro solubility prediction. When it comes to enhancing the solubility of proteins, we, for the first time, propose to investigate the synergistic effects of tandem mutations instead of “single” mutations, suggesting that this could minimize the number of required proposed mutations.

]]>
<![CDATA[Scedar: A scalable Python package for single-cell RNA-seq exploratory data analysis]]> https://www.researchpad.co/article/elastic_article_13837 In single-cell RNA-seq (scRNA-seq) experiments, the number of individual cells has increased exponentially, and the sequencing depth of each cell has decreased significantly. As a result, analyzing scRNA-seq data requires extensive considerations of program efficiency and method selection. In order to reduce the complexity of scRNA-seq data analysis, we present scedar, a scalable Python package for scRNA-seq exploratory data analysis. The package provides a convenient and reliable interface for performing visualization, imputation of gene dropouts, detection of rare transcriptomic profiles, and clustering on large-scale scRNA-seq datasets. The analytical methods are efficient, and they also do not assume that the data follow certain statistical distributions. The package is extensible and modular, which would facilitate the further development of functionalities for future requirements with the open-source development community. The scedar package is distributed under the terms of the MIT license at https://pypi.org/project/scedar.

]]>
<![CDATA[Forecasting the monthly incidence rate of brucellosis in west of Iran using time series and data mining from 2010 to 2019]]> https://www.researchpad.co/article/elastic_article_13811 The identification of statistical models for the accurate forecast and timely determination of the outbreak of infectious diseases is very important for the healthcare system. Thus, this study was conducted to assess and compare the performance of four machine-learning methods in modeling and forecasting brucellosis time series data based on climatic parameters.MethodsIn this cohort study, human brucellosis cases and climatic parameters were analyzed on a monthly basis for the Qazvin province–located in northwestern Iran- over a period of 9 years (2010–2018). The data were classified into two subsets of education (80%) and testing (20%). Artificial neural network methods (radial basis function and multilayer perceptron), support vector machine and random forest were fitted to each set. Performance analysis of the models were done using the Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Root Error (MARE), and R2 criteria.ResultsThe incidence rate of the brucellosis in Qazvin province was 27.43 per 100,000 during 2010–2019. Based on our results, the values of the RMSE (0.22), MAE (0.175), MARE (0.007) criteria were smaller for the multilayer perceptron neural network than their values in the other three models. Moreover, the R2 (0.99) value was bigger in this model. Therefore, the multilayer perceptron neural network exhibited better performance in forecasting the studied data. The average wind speed and mean temperature were the most effective climatic parameters in the incidence of this disease.ConclusionsThe multilayer perceptron neural network can be used as an effective method in detecting the behavioral trend of brucellosis over time. Nevertheless, further studies focusing on the application and comparison of these methods are needed to detect the most appropriate forecast method for this disease. ]]> <![CDATA[Any unique image biomarkers associated with COVID-19?]]> https://www.researchpad.co/article/elastic_article_13335 To define the uniqueness of chest CT infiltrative features associated with COVID-19 image characteristics as potential diagnostic biomarkers.MethodsWe retrospectively collected chest CT exams including n = 498 on 151 unique patients RT-PCR positive for COVID-19 and n = 497 unique patients with community-acquired pneumonia (CAP). Both COVID-19 and CAP image sets were partitioned into three groups for training, validation, and testing respectively. In an attempt to discriminate COVID-19 from CAP, we developed several classifiers based on three-dimensional (3D) convolutional neural networks (CNNs). We also asked two experienced radiologists to visually interpret the testing set and discriminate COVID-19 from CAP. The classification performance of the computer algorithms and the radiologists was assessed using the receiver operating characteristic (ROC) analysis, and the nonparametric approaches with multiplicity adjustments when necessary.ResultsOne of the considered models showed non-trivial, but moderate diagnostic ability overall (AUC of 0.70 with 99% CI 0.56–0.85). This model allowed for the identification of 8–50% of CAP patients with only 2% of COVID-19 patients.ConclusionsProfessional or automated interpretation of CT exams has a moderately low ability to distinguish between COVID-19 and CAP cases. However, the automated image analysis is promising for targeted decision-making due to being able to accurately identify a sizable subsect of non-COVID-19 cases.Key Points • Both human experts and artificial intelligent models were used to classify the CT scans. • ROC analysis and the nonparametric approaches were used to analyze the performance of the radiologists and computer algorithms. • Unique image features or patterns may not exist for reliably distinguishing all COVID-19 from CAP; however, there may be imaging markers that can identify a sizable subset of non-COVID-19 cases. ]]> <![CDATA[A model for the assessment of bluetongue virus serotype 1 persistence in Spain]]> https://www.researchpad.co/article/elastic_article_11225 Bluetongue virus (BTV) is an arbovirus of ruminants that has been circulating in Europe continuously for more than two decades and has become endemic in some countries such as Spain. Spain is ideal for BTV epidemiological studies since BTV outbreaks from different sources and serotypes have occurred continuously there since 2000; BTV-1 has been reported there from 2007 to 2017. Here we develop a model for BTV-1 endemic scenario to estimate the risk of an area becoming endemic, as well as to identify the most influential factors for BTV-1 persistence. We created abundance maps at 1-km2 spatial resolution for the main vectors in Spain, Culicoides imicola and Obsoletus and Pulicaris complexes, by combining environmental satellite data with occurrence models and a random forest machine learning algorithm. The endemic model included vector abundance and host-related variables (farm density). The three most relevant variables in the endemic model were the abundance of C. imicola and Obsoletus complex and density of goat farms (AUC 0.86); this model suggests that BTV-1 is more likely to become endemic in central and southwestern regions of Spain. It only requires host- and vector-related variables to identify areas at greater risk of becoming endemic for bluetongue. Our results highlight the importance of suitable Culicoides spp. prediction maps for bluetongue epidemiological studies and decision-making about control and eradication measures.

]]>
<![CDATA[Using case-level context to classify cancer pathology reports]]> https://www.researchpad.co/article/elastic_article_7869 Individual electronic health records (EHRs) and clinical reports are often part of a larger sequence—for example, a single patient may generate multiple reports over the trajectory of a disease. In applications such as cancer pathology reports, it is necessary not only to extract information from individual reports, but also to capture aggregate information regarding the entire cancer case based off case-level context from all reports in the sequence. In this paper, we introduce a simple modular add-on for capturing case-level context that is designed to be compatible with most existing deep learning architectures for text classification on individual reports. We test our approach on a corpus of 431,433 cancer pathology reports, and we show that incorporating case-level context significantly boosts classification accuracy across six classification tasks—site, subsite, laterality, histology, behavior, and grade. We expect that with minimal modifications, our add-on can be applied towards a wide range of other clinical text-based tasks.

]]>
<![CDATA[Medusa: Software to build and analyze ensembles of genome-scale metabolic network reconstructions]]> https://www.researchpad.co/article/elastic_article_7734 Uncertainty in the structure and parameters of networks is ubiquitous across computational biology. In constraint-based reconstruction and analysis of metabolic networks, this uncertainty is present both during the reconstruction of networks and in simulations performed with them. Here, we present Medusa, a Python package for the generation and analysis of ensembles of genome-scale metabolic network reconstructions. Medusa builds on the COBRApy package for constraint-based reconstruction and analysis by compressing a set of models into a compact ensemble object, providing functions for the generation of ensembles using experimental data, and extending constraint-based analyses to ensemble scale. We demonstrate how Medusa can be used to generate ensembles and perform ensemble simulations, and how machine learning can be used in conjunction with Medusa to guide the curation of genome-scale metabolic network reconstructions. Medusa is available under the permissive MIT license from the Python Packaging Index (https://pypi.org) and from github (https://github.com/opencobra/Medusa), and comprehensive documentation is available at https://medusa.readthedocs.io/en/latest.

]]>
<![CDATA[FaceLift: a transparent deep learning framework to beautify urban scenes]]> https://www.researchpad.co/article/N5fd42e94-d295-4df2-a0e6-a8f34580eb5a

In the area of computer vision, deep learning techniques have recently been used to predict whether urban scenes are likely to be considered beautiful: it turns out that these techniques are able to make accurate predictions. Yet they fall short when it comes to generating actionable insights for urban design. To support urban interventions, one needs to go beyond predicting beauty, and tackle the challenge of recreating beauty. Unfortunately, deep learning techniques have not been designed with that challenge in mind. Given their ‘black-box nature’, these models cannot be directly used to explain why a particular urban scene is deemed to be beautiful. To partly fix that, we propose a deep learning framework (which we name FaceLift1) that is able to both beautify existing urban scenes (Google Street Views) and explain which urban elements make those transformed scenes beautiful. To quantitatively evaluate our framework, we cannot resort to any existing metric (as the research problem at hand has never been tackled before) and need to formulate new ones. These new metrics should ideally capture the presence (or absence) of elements that make urban spaces great. Upon a review of the urban planning literature, we identify five main metrics: walkability, green spaces, openness, landmarks and visual complexity. We find that, across all the five metrics, the beautified scenes meet the expectations set by the literature on what great spaces tend to be made of. This result is further confirmed by a 20-participant expert survey in which FaceLift has been found to be effective in promoting citizen participation. All this suggests that, in the future, as our framework’s components are further researched and become better and more sophisticated, it is not hard to imagine technologies that will be able to accurately and efficiently support architects and planners in the design of the spaces we intuitively love.

]]>