Search

This presentation explores how n-grams can be used to automatically classify and learn about music. An overall discussion is provided of various ways in which n-grams can be adapted for use with digital scores, and of how musically meaningful features can be extracted from them. The jSymbolic 3.0 alpha prototype feature extractor is then used in three sets of music classification experiments investigating how n-gram features perform relative to and combined with other types of features extracted from symbolic music files., Funded by the FRQSC and SSHRC.

We present a Web application that facilitates the deep visual search of image collections using contemporary machine learning. We discuss image retrieval as a combined computer vision/human-computer interaction problem, and propose that the standardization of feature extraction is one of the main problems that digital art history faces today.

In this study, we are looking for early mentions of basic principles of sustainability in literary texts using machine learning. With a trained CRF-classifier we automatically annotated a corpus of about 70 novels from German romanticism and singled out passages with a high frequency of annotations, which we analysed in terms of a concept of sustainability. We thus test whether the application of machine learning used to operationalise high-frequency phenomena can also help uncover less frequent and highly interpretive aspects of narrative texts.

Collective sentiment plays a crucial role in financial markets. In this paper, we present our workflow for creating an index that captures the sentiment at the Berlin Stock Exchange from 1872 to 1930. Based on a historical newspaper, we train three models that provide daily aspect-level sentiment data.

The project develops an open-source machine learning platform for East Asian Studies, the first one in the United States.Our platform will be so versatile that humanists will not have to devote time to learning various digital tools. With our sample codes and tutorials, researchers can conduct computer-aided research.

What role does color play in how we imagine our own world and the world(s) of others? Using machine learning and methods from Explainable AI, we study and explain how color use shaped the geographic imaginaries of the Orient and Occident in two different media: photochromes and autochromes

This contribution deals with Dynamic Topic Modelling of Austrian historical newspapers (found in the ANNO Corpus), focusing on the time period from the 18th to the mid-20th Century. The aim of the investigation is to uncover the development of the topic surrounding migration from and via Southeast Europe to Austria.

A decade ago, Johanna Drucker asked: "Is There a 'Digital' Art History?" In this paper, we attempt to reconsider Drucker's question in the light of recent developments in computer vision by investigating the epistemic implications and methodological affordances of large-scale, transformer-based vision models.

This paper explores representations of holiness in a collection of 63.000 lantern slides. Using the multimodal model CLIP, it sheds light on the transition between the real and the imaginary, the holy and the profane. It demonstrates how CLIP can help scholars deal with complex visual concepts such as holiness.

The poster develops and discusses a new technique of modeling prototypicality in genre concepts based on the c@1-score as an uncertainty-based accuracy score and perspectival modeling within the framework of supervised machine learning. The technique is intended to provide better understanding of loosely ordered genres.

We present a modeling of document dates such that it can allow scholars to express uncertainty when annotating data, act as a differentiable loss function for training models and allow for unbiased interpretable performance evaluation under uncertainty.

Kao is a stylized signature used in East Asian cultures. Its location and size are important elements of document manners. This study explores information about their location and size in images through numerical summarization and visualization, and provides quantitative results that support existing tacit knowledge.

In large digital libraries, such as the HathiTrust, metadata is insufficient to identify items of interest. Metadata records are often incomplete and challenging for fiction, where metadata categories, when present, are too broad. This project constructs a machine learning pipeline for fiction classification using the HTRC Extracted Features Dataset, and based on previous work from Underwood et al. We will detail the methodology, early results, and planned future work in generating this dataset.

We consider the problem of estimating a signal from measurements obtained via a generalized linear model. We focus on estimators based on approximate message passing (AMP), a family of iterative algorithms with many appealing features: the performance of AMP in the high-dimensional limit can be succinctly characterized under suitable model assumptions; AMP can also be tailored to the empirical distribution of the signal entries, and for a wide class of estimation problems, AMP is conjectured to be optimal among all polynomial-time algorithms. However, a major issue of AMP is that in many models (such as phase retrieval), it requires an initialization correlated with the ground-truth signal and independent from the measurement matrix. Assuming that such an initialization is available is typically not realistic. In this paper, we solve this problem by proposing an AMP algorithm initialized with a spectral estimator. With such an initialization, the standard AMP analysis fails since the spectral estimator depends in a complicated way on the design matrix. Our main contribution is a rigorous characterization of the performance of AMP with spectral initialization in the high-dimensional limit. The key technical idea is to define and analyze a two-phase artificial AMP algorithm that first produces the spectral estimator, and then closely approximates the iterates of the true AMP. We also provide numerical results that demonstrate the validity of the proposed approach.

We consider the problem of estimating a signal from measurements obtained via a generalized linear model. We focus on estimators based on approximate message passing (AMP), a family of iterative algorithms with many appealing features: the performance of AMP in the high-dimensional limit can be succinctly characterized under suitable model assumptions; AMP can also be tailored to the empirical distribution of the signal entries, and for a wide class of estimation problems, AMP is conjectured to be optimal among all polynomial-time algorithms. However, a major issue of AMP is that in many models (such as phase retrieval), it requires an initialization correlated with the ground-truth signal and independent from the measurement matrix. Assuming that such an initialization is available is typically not realistic. In this paper, we solve this problem by proposing an AMP algorithm initialized with a spectral estimator. With such an initialization, the standard AMP analysis fails since the spectral estimator depends in a complicated way on the design matrix. Our main contribution is a rigorous characterization of the performance of AMP with spectral initialization in the high-dimensional limit. The key technical idea is to define and analyze a two-phase artificial AMP algorithm that first produces the spectral estimator, and then closely approximates the iterates of the true AMP. We also provide numerical results that demonstrate the validity of the proposed approach., Comment: 38 pages, 5 figures, AISTATS 2021

This article reports a simple and inexpensive leak-proof paper pad with an initial selection of a paper substrate on the grounds of surface morphology and fluid absorption time. Herein, a drying method is used for glucose detection on a paper pad through colorimetric analysis, and the spot detection of glucose is analyzed by optimizing the HRP concentration and volume to obtain accurate results. The rapid colorimetric method for the detection of glucose on the paper pad was developed with a limit of detection (LOD) of 2.92 mmol L−1. Furthermore, the effects of the detection conditions were investigated and discussed comprehensively with the help of chemometric methods. Paper pads were developed for glucose detection with a range of 0.5–20 mM (apropos to the normal glucose level in the human body) and 0.1–0.5 M (to test the excessive intake of glucose). The developed concept has huge potential in the healthcare sector, and its extension could be envisioned to develop the reported paper pad as a point-of-care testing device for the initial screening of a variety of diseases.

Microfluidic paper-based analytical devices (μPADs) have been widely used in point-of-care testing owing to their simple operation, low volume of the sample required, and the lack of the need for an external force. To obtain accurate semi-quantitative or quantitative results, μPADs need to respond to the challenges posed by differences in reaction conditions. In this paper, multi-layer μPADs are fabricated by the imprinting method for the colorimetric detection of C-reactive protein (CRP). Different lighting conditions and shooting angles of scenes are simulated in image acquisition, and the detection-related performance of μPADs is improved by using a machine learning algorithm. The You Only Look Once (YOLO) model is used to identify the areas of reaction in μPADs. This model can observe an image only once to predict the objects present in it and their locations. The YOLO model trained in this study was able to identify all the reaction areas quickly without incurring any error. These reaction areas were categorized by classification algorithms to determine the risk level of CRP concentration. Multi-layer perceptron, convolutional neural network, and residual network algorithms were used for the classification tasks, where the latter yielded the highest accuracy of 96%. It has a promising application prospect in fast recognition and analysis of μPADs.

We demonstrate the use of a convolutional neural network to perform neutron-gamma pulse shape discrimination, where the only inputs to the network are the raw digitised SiPM signals from a dual scintillator detector element made of 6LiF:ZnS(Ag) scintillator and PVT plastic. A realistic labelled dataset was created to train the network by exposing the detector to an AmBe source, and a data-driven method utilsing a separate PMT was used to assign labels to the recorded signals. This approach is compared to the charge integration and continuous wavelet transform methods and is found to provide superior levels of discrimination, achieving an AUC of 0.995 +/- 0.003. We find that the neural network is capable of extracting interpretable features directly from the raw data. In addition, by visualising the high-dimensional representations of the network with the t-SNE algorithm, we discover that not only is this method robust to minor mislabeling of the training dataset but that it is possible to identify an underlying substructure within the signals that goes beyond the original labelling. This technique could be utilised to explore and cluster complex, raw detector data in a novel way that may reveal more insights than standard analysis methods., 13 pages, 8 figures, submitted to Nucl. Instr. Meth. A

Significance: Diffuse reflectance spectroscopy (DRS) allows discrimination of tissue type. Its application is limited by the inability to mark the scanned tissue and the lack of real-time measurements. Aim: This study aimed to develop a real-time tracking system to enable localization of a DRS probe to aid the classification of tumor and non-tumor tissue. Approach: A green-colored marker attached to the DRS probe was detected using hue-saturation-value (HSV) segmentation. A live, augmented view of tracked optical biopsy sites was recorded in real time. Supervised classifiers were evaluated in terms of sensitivity, specificity, and overall accuracy. A developed software was used for data collection, processing, and statistical analysis. Results: The measured root mean square error (RMSE) of DRS probe tip tracking was 1.18±0.58 mm and 1.05±0.28 mm for the x and y dimensions, respectively. The diagnostic accuracy of the system to classify tumor and non-tumor tissue in real time was 94% for stomach and 96% for the esophagus. Conclusions: We have successfully developed a real-time tracking and classification system for a DRS probe. When used on stomach and esophageal tissue for tumor detection, the accuracy derived demonstrates the strength and clinical value of the technique to aid margin assessment in cancer resection surgery.

In this work, the capability of restricted Boltzmann machines (RBMs) to find solutions for the Kitaev honeycomb model with periodic boundary conditions is investigated. The measured groundstate (GS) energy of the system is compared and, for small lattice sizes (e.g. $3 \times 3$ with $18$ spinors), shown to agree with the analytically derived value of the energy up to a deviation of $0.09\%$. Moreover, the wave-functions we find have $99.89\%$ overlap with the exact ground state wave-functions. Furthermore, the possibility of realizing anyons in the RBM is discussed and an algorithm is given to build these anyonic excitations and braid them for possible future applications in quantum computation. Using the correspondence between topological field theories in (2+1)d and 2d CFTs, we propose an identification between our RBM states with the Moore-Read state and conformal blocks of the $2$d Ising model., Comment: (15+5) pages, 11 figures, 5 tables: Minor journal revisions and additions. + Journal ref

Funder: Royal Society; doi: http://dx.doi.org/10.13039/501100000288, Funder: Trinity College, University of Cambridge; doi: http://dx.doi.org/10.13039/501100000727, Funder: John Fell Fund, University of Oxford; doi: http://dx.doi.org/10.13039/501100004789, We use a deep neural network (NN) to detect and place region-of-interest (ROI) boxes around ultracold atom clouds in absorption and fluorescence images—with the ability to identify and bound multiple clouds within a single image. The NN also outputs segmentation masks that identify the size, shape and orientation of each cloud from which we extract the clouds’ Gaussian parameters. This allows 2D Gaussian fits to be reliably seeded thereby enabling fully automatic image processing. The method developed performs significantly better than a more conventional method based on a standardized image analysis library (Scikit-image) both for identifying ROI and extracting Gaussian parameters.

In the present study, iodide-mediated 3,3′,5,5′-tetramethylbenzidine (TMB)-H2O2 reaction system was applied to a microfluidic paper-based analytical device (μPAD) for non-enzymatic colorimetric determination of H2O2. The proposed system is portable and incorporates a μPAD with a machine learning-based smartphone app. A smartphone app called “Hi-perox Sens” capable of image capture, cropping and processing was developed to make the system simple and user-friendly. Briefly, circular μPADs were designed and tested with varying concentrations of H2O2. Following the color change, the images of the μPADs were taken with four different smartphones under seven different illumination conditions. In order to make the system more robust and adaptive against illumination variation and camera optics, the images were first processed for feature extraction and then used to train machine learning classifiers. According to the results, TMB + KI showed the highest classification accuracy (97.8%) with inter-phone repeatability at t = 30 s under versatile illumination and maintained its accuracy for 10 minutes. In addition, the performance of the system was also comparable to two different commercially available H2O2 kits in real samples.

Significance Prostate cancer is the most common cancer among men. An accurate diagnosis of its severity at detection plays a major role in improving their survival. Recently, machine learning models using biomarkers identified from Raman micro-spectroscopy discriminated intraductal carcinoma of the prostate (IDC-P) from cancer tissue with a ≥85 % detection accuracy and differentiated high-grade prostatic intraepithelial neoplasia (HGPIN) from IDC-P with a ≥97.8 % accuracy. Aim To improve the classification performance of machine learning models identifying different types of prostate cancer tissue using a new dimensional reduction technique. Approach A radial basis function (RBF) kernel support vector machine (SVM) model was trained on Raman spectra of prostate tissue from a 272-patient cohort (Centre hospitalier de l'Universite de Montreal, CHUM) and tested on two independent cohorts of 76 patients [University Health Network (UHN)] and 135 patients (Centre hospitalier universitaire de Quebec-Universite Laval, CHUQc-UL). Two types of engineered features were used. Individual intensity features, i.e., Raman signal intensity measured at particular wavelengths and novel Raman spectra fitted peak features consisting of peak heights and widths. Results Combining engineered features improved classification performance for the three aforementioned classification tasks. The improvements for IDC-P/cancer classification for the UHN and CHUQc-UL testing sets in accuracy, sensitivity, specificity, and area under the curve (AUC) are (numbers in parenthesis are associated with the CHUQc-UL testing set): +4 % (+8 % ), +7 % (+9 % ), +2 % (6%), +9 (+9) with respect to the current best models. Discrimination between HGPIN and IDC-P was also improved in both testing cohorts: +2.2 % (+1.7 % ), +4.5 % (+3.6 % ), +0 % (+0 % ), +2.3 (+0). While no global improvements were obtained for the normal versus cancer classification task [+0 % (-2 % ), +0 % (-3 % ), +2 % (-2 % ), +4 (+3)], the AUC was improved in both testing sets. Conclusions Combining individual intensity features and novel Raman fitted peak features, improved the classification performance on two independent and multicenter testing sets in comparison to using only individual intensity features.

Significance: Early detection of oral cancer is vital for high-risk patients, and machine learning-based automatic classification is ideal for disease screening. However, current datasets collected from high-risk populations are unbalanced and often have detrimental effects on the performance of classification. Aim: To reduce the class bias caused by data imbalance. Approach: We collected 3851 polarized white light cheek mucosa images using our customized oral cancer screening device. We use weight balancing, data augmentation, undersampling, focal loss, and ensemble methods to improve the neural network performance of oral cancer image classification with the imbalanced multi-class datasets captured from high-risk populations during oral cancer screening in low-resource settings. Results: By applying both data-level and algorithm-level approaches to the deep learning training process, the performance of the minority classes, which were difficult to distinguish at the beginning, has been improved. The accuracy of “premalignancy” class is also increased, which is ideal for screening applications. Conclusions: Experimental results show that the class bias induced by imbalanced oral cancer image datasets could be reduced using both data- and algorithm-level methods. Our study may provide an important basis for helping understand the influence of unbalanced datasets on oral cancer deep learning classifiers and how to mitigate.

Significance: Sub-diffuse optical properties may serve as useful cancer biomarkers, and wide-field heatmaps of these properties could aid physicians in identifying cancerous tissue. Sub-diffuse spatial frequency domain imaging (sd-SFDI) can reveal such wide-field maps, but the current time cost of experimentally validated methods for rendering these heatmaps precludes this technology from potential real-time applications. Aim: Our study renders heatmaps of sub-diffuse optical properties from experimental sd-SFDI images in real time and reports these properties for cancerous and normal skin tissue subtypes. Approach: A phase function sampling method was used to simulate sd-SFDI spectra over a wide range of optical properties. A machine learning model trained on these simulations and tested on tissue phantoms was used to render sub-diffuse optical property heatmaps from sd-SFDI images of cancerous and normal skin tissue. Results: The model accurately rendered heatmaps from experimental sd-SFDI images in real time. In addition, heatmaps of a small number of tissue samples are presented to inform hypotheses on sub-diffuse optical property differences across skin tissue subtypes. Conclusion: These results bring the overall process of sd-SFDI a fundamental step closer to real-time speeds and set a foundation for future real-time medical applications of sd-SFDI such as image guided surgery.