Showing papers on "Ordinal regression published in 2021"
TL;DR: This work proposes a classifier based on a deep convolutional neural network that outperforms average human interrater as well as intrarater reliability and surpasses state-of-the-art machine learning solutions for automatically grading disc degeneration.
Abstract: OBJECTIVES Although magnetic resonance imaging-based formalized grading schemes for intervertebral disc degeneration offer improved reproducibility compared with purely subjective ratings, their intrarater and interrater reliability are not nearly good enough to be able to detect small to medium effects in clinical longitudinal studies. The aim of this study thus was to develop a method that enables automatic and therefore reproducible and reliable evaluation of disc degeneration based on conventional clinical image data and Pfirrmann's grading scheme. MATERIALS AND METHODS We propose a classifier based on a deep convolutional neural network that we trained on a large, manually evaluated data set of 1599 patients (7948 intervertebral discs). To improve upon the status quo, we focused on the quality of the training data and performed extensive hyperparameter optimization. We assessed the potential benefits of optimizing loss functions beyond common cross-entropy loss, such as soft kappa loss, ordinal cross-entropy loss, or regression losses. We furthermore experimented with ways to mitigate class imbalance by pooling classes or using class-weighted loss functions. During model development and hyperparameter optimization, we used a fixed 90%/10% training/validation set split. To estimate real-world prediction performance, we performed 10-fold cross-validation. RESULTS The evaluated image data results in a Gaussian degeneration grade distribution, and thus grades 1 and 5 are slightly underrepresented in the training set. Our default cross-entropy-based classifier achieves a reliability of κ = 0.92 (Cohen κ), an average sensitivity of 90.2%, and an average precision of 92.5%. In 99.2% of validation cases, the network's prediction deviates at most 1 Pfirrmann grades from the ground truth. Framed as an ordinal regression problem, the mean absolute error between the ground truth and the prediction is 0.08 Pfirrmann grade with a correlation of r = 0.96. The results of the 10-fold cross validation confirm those performance estimates, indicating no substantial overfitting. More sophisticated loss functions, class-based loss weighting, or class pooling did not lead to improved classification performance overall. CONCLUSIONS With a reliability of κ > 0.9, our system clearly outperforms average human interrater as well as intrarater reliability. With an average sensitivity of more than 90%, our classifier also surpasses state-of-the-art machine learning solutions for automatically grading disc degeneration.
31 citations
08 Mar 2021
TL;DR: In this article, a hierarchical attention model is proposed to factor in the graded nature of increasing suicide risk levels, through soft probability distribution since not all wrong risk-levels are equally wrong.
Abstract: The rising ubiquity of social media presents a platform for individuals to express suicide ideation, instead of traditional, formal clinical settings. While neural methods for assessing suicide risk on social media have shown promise, a crippling limitation of existing solutions is that they ignore the inherent ordinal nature across fine-grain levels of suicide risk. To this end, we reformulate suicide risk assessment as an Ordinal Regression problem, over the Columbia-Suicide Severity Scale. We propose SISMO, a hierarchical attention model optimized to factor in the graded nature of increasing suicide risk levels, through soft probability distribution since not all wrong risk-levels are equally wrong. We establish the face value of SISMO for preliminary suicide risk assessment on real-world Reddit data annotated by clinical experts. We conclude by discussing the empirical, practical, and ethical considerations pertaining to SISMO in a larger picture, as a human-in-the-loop framework
22 citations
TL;DR: In this article, the authors evaluated passengers' satisfaction regarding choice A and choice B through a questionnaire survey and found that service-attributes indicated a larger positive impact on overall satisfaction with choice A as compared to ambiance.
Abstract: App-based demand-responsive transit (DRT) services are emerging where conventional public transport is unable to meet the demand. SWVL (choice A) and Airlift (choice B) are two such DRT bus services operating in Lahore, Pakistan. It is important for the policy makers and operators to evaluate the satisfaction levels of the passengers using these services. This study evaluated passengers’ satisfaction regarding choice A and choice B through a questionnaire survey. A total of 440 responses were collected from the users of the DRT services through personal interviews and a web-based approach. Factor analysis on the collected data produced two underlying factors, namely service-attributes and bus ambiance. Ordinal regression showed that the service-attributes and ambiance were significant predictors of overall satisfaction levels about choice A. Service-attributes indicated a larger positive impact on overall satisfaction with choice A as compared to ambiance. Although the ordinal model for choice B fitted the data well, the predictors were found to be insignificant. The results offer an insight into which predictors affect the overall satisfaction and how it can be improved.
20 citations
13 Apr 2021
TL;DR: MedMNIST as discussed by the authors is a collection of 10 pre-processed medical open datasets, which can be used for educational purpose, rapid prototyping, multi-modal machine learning or AutoML in medical image analysis.
Abstract: We present MedMNIST, a collection of 10 pre-processed medical open datasets. MedMNIST is standardized to perform classification tasks on lightweight 28 $\times$ 28 images, which requires no background knowledge. Covering the primary data modalities in medical image analysis, it is diverse on data scale (from 100 to 100,000) and tasks (binary/multi-class, ordinal regression and multi-label). MedMNIST could be used for educational purpose, rapid prototyping, multi-modal machine learning or AutoML in medical image analysis. Moreover, MedMNIST Classification Decathlon is designed to benchmark AutoML algorithms on all 10 datasets; We have compared several baseline methods, including open-source or commercial AutoML tools. The datasets, evaluation code and baseline methods for MedMNIST are publicly available at https://medmnist.github.io/.
20 citations
TL;DR: An ordinal regression module for neural networks is proposed to treat Kellgren-Lawrence (KL) grading as anOrdinal regression task and performance of the model is evaluated against various notable neural networks and significant improvements on the knee OA KL grade prediction were demonstrated.
Abstract: Osteoarthritis (OA) is a common form of knee arthritis which causes significant disability and is threatening to plague patient’s quality of life. Although this chronic condition does not lead to fatality, still there exists no known cure for OA. Diagnosis of OA can be confirmed primarily based on radiographic findings. Being a progressive disease, early identification of OA is crucial for clinical interventions to curtail the OA degeneration. Kellgren-Lawrence (KL) grading system has been traditionally employed to assess the knee OA severity. Due to the recent advancements of deep learning in computer vision, more studies have employed deep neural network in automatically predicting KL grade from plain knee joint radiograph. However, these studies treat KL grading as a multi-class classification task and ignore the inherent ordinal nature within the KL grades. In this study, we propose an ordinal regression module for neural networks to treat KL grading as an ordinal regression task. Our module takes an input from neural network and produces 4 cut-points to partition the prediction space into 5 respective KL grades. The proposed model is optimized by a cumulative-link loss function. Performance of the model is evaluated against various notable neural networks and significant improvements on the knee OA KL grade prediction were demonstrated.
20 citations
TL;DR: A new multiple criteria decision aiding method to deal with sorting problems in which alternatives are evaluated on criteria structured in a hierarchical way and presenting interactions and the underlying preference model is the Choquet integral.
20 citations
TL;DR: Wang et al. as discussed by the authors proposed a convolutional OR forest (CORF) for image ordinal estimation, which can integrate OR and differentiable decision trees with a CNN for obtaining precise and stable global ordinal relationships.
Abstract: Image ordinal estimation is to predict the ordinal label of a given image, which can be categorized as an ordinal regression (OR) problem. Recent methods formulate an OR problem as a series of binary classification problems. Such methods cannot ensure that the global ordinal relationship is preserved since the relationships among different binary classifiers are neglected. We propose a novel OR approach, termed convolutional OR forest (CORF), for image ordinal estimation, which can integrate OR and differentiable decision trees with a convolutional neural network for obtaining precise and stable global ordinal relationships. The advantages of the proposed CORF are twofold. First, instead of learning a series of binary classifiers independently, the proposed method aims at learning an ordinal distribution for OR by optimizing those binary classifiers simultaneously. Second, the differentiable decision trees in the proposed CORF can be trained together with the ordinal distribution in an end-to-end manner. The effectiveness of the proposed CORF is verified on two image ordinal estimation tasks, i.e., facial age estimation and image esthetic assessment, showing significant improvements and better stability over the state-of-the-art OR methods.
16 citations
16 citations
TL;DR: In this paper, the authors investigated pedestrians' sociodemographic attributes (gender, age, qualitative components (perceived comfort) and the pedestrian streets' characteristics for LOS estimation as perceived by pedestrians.
Abstract: The assessment of pedestrian facilities largely depends on the Level of Service (LOS) calculation taking into account the effective width and the pedestrians’ flows. This research focuses on the investigation of pedestrians’ sociodemographic attributes (gender, age), qualitative components (perceived comfort) and the pedestrian streets’ characteristics for LOS estimation as perceived by pedestrians. Α questionnaire-based survey (including 200 interviewees) was carried out at the central pedestrianized network in the city of Ioannina, Greece. An ordinal regression model was developed based on the questionnaire’ data. According to the findings females and elderly pedestrians tend to perceive lower LOS than males and younger pedestrians respectively. Furthermore, the level of perceived comfort in the impact area of the urban central network is an important determinant for the estimation of perceived LOS. These findings can be effectively utilized in the design of sustainable mobility policy recommendations and the promotion of active transport in an urban central area.
15 citations
15 citations
TL;DR: In this paper, a coordinate descent algorithm is proposed to fit a broad class of ordinal regression models with an elastic net penalty, which can be used to shrink a non-ordinal model toward its ordinal counterpart.
Abstract: Regularization techniques such as the lasso (Tibshirani 1996) and elastic net (Zou and Hastie 2005) can be used to improve regression model coefficient estimation and prediction accuracy, as well as to perform variable selection. Ordinal regression models are widely used in applications where the use of regularization could be beneficial; however, these models are not included in many popular software packages for regularized regression. We propose a coordinate descent algorithm to fit a broad class of ordinal regression models with an elastic net penalty. Furthermore, we demonstrate that each model in this class generalizes to a more flexible form, that can be used to model either ordered or unordered categorical response data. We call this the elementwise link multinomial-ordinal class, and it includes widely used models such as multinomial logistic regression (which also has an ordinal form) and ordinal logistic regression (which also has an unordered multinomial form). We introduce an elastic net penalty class that applies to either model form, and additionally, this penalty can be used to shrink a non-ordinal model toward its ordinal counterpart. Finally, we introduce the R package ordinalNet, which implements the algorithm for this model class.
19 Sep 2021
TL;DR: A novel preprocessing method for reducing the sparseness and limited field of view provided by radar and a novel method for estimating dense depth maps from monocular 2D images and sparse radar measurements using deep learning based on the deep ordinal regression network are proposed.
Abstract: We integrate sparse radar data into a monocular depth estimation model and introduce a novel preprocessing method for reducing the sparseness and limited field of view provided by radar. We explore the intrinsic error of different radar modalities and show our proposed method results in more data points with reduced error. We further propose a novel method for estimating dense depth maps from monocular 2D images and sparse radar measurements using deep learning based on the deep ordinal regression network by Fu et al. Radar data are integrated by first converting the sparse 2D points to a height-extended 3D measurement and then including it into the network using a late fusion approach. Experiments are conducted on the nuScenes dataset. Our experiments demonstrate state-of-the-art performance in both day and night scenes.
TL;DR: In this article, the sensitivity of Rapid upper limb assessment (RULA) and identification of insensitive and sensitive posture zones is studied. But, the sensitivity analysis of RULA along with the ordinal regression analysis offer deeper insights into the methodology used for the assessment of posture.
10 Jan 2021
TL;DR: This article proposed to use several discrete data representations simultaneously to improve neural network learning compared to a single representation, which can be added as a simple extension to conventional learning methods, such as deep neural networks.
Abstract: Regression via classification (RvC) is a common method used for regression problems in deep learning, where the target variable belongs to a set of continuous values. By discretizing the target into a set of non-overlapping classes, it has been shown that training a classifier can improve neural network accuracy compared to using a standard regression approach. However, it is not clear how the set of discrete classes should be chosen and how it affects the overall solution. In this work, we propose that using several discrete data representations simultaneously can improve neural network learning compared to a single representation. Our approach is end-to-end differentiable and can be added as a simple extension to conventional learning methods, such as deep neural networks. We test our method on three challenging tasks and show that our method reduces the prediction error compared to a baseline RvC approach while maintaining a similar model complexity.
TL;DR: An ordinal regression analysis model is developed for studying the preferences of artistic goods buyers using a large set of auction data from the Art Deco furniture market and considers several different criteria that may influence buyers’ preferences.
TL;DR: In this article, a deep learning framework called ordinal regression and recurrent convolutional neural network (OR-RCNN) is introduced to predict protein protein interactions (PPIs) from the perspective of confidence score.
Abstract: Protein protein interactions (PPIs) are essential to most of the biological processes. The prediction of PPIs is beneficial to the understanding of protein functions and thus is helpful to pathological analysis, disease diagnosis and drug design etc. As the amount of protein data is growing fast in the post genomic era, high-throughput experimental methods are expensive and time-consuming for the prediction of PPIs. Thus, computational methods have attracted researcher’s attention in recent years. A large number of computational methods have been proposed based on different protein sequence encoders. Notably, the confidence score of a protein sequence pair could be regarded as a kind of measurement to PPIs. The higher the confidence score for one protein pair is, the more likely the protein pair interacts. Thus in this paper, a deep learning framework, called ordinal regression and recurrent convolutional neural network (OR-RCNN) method, is introduced to predict PPIs from the perspective of confidence score. It mainly contains two parts: the encoder part of protein sequence pair and the prediction part of PPIs by confidence score. In the first part, two recurrent convolutional neural networks (RCNNs) with shared parameters are applied to construct two protein sequence embedding vectors, which can automatically extract robust local features and sequential information from the protein pairs. Based on it, the two embedding vectors are encoded into one novel embedding vector by element-wise multiplication. By taking the ordinal information behind confidence score into consideration, ordinal regression is used to construct multiple sub-classifiers in the second part. The results of multiple sub-classifiers are aggregated to obtain the final confidence score. Following that, the existence of PPIs is determined by the confidence score. We set a threshold $$\theta$$
, and say the interaction exists between the protein pair if its confidence score is bigger than $$\theta$$
. We applied our method to predict PPIs on data sets S. cerevisiae and Homo sapiens. Through experimental verification, our method outperforms state-of-the-art PPI prediction models.
TL;DR: A new DL model for weakly-supervised DA with ordinal regression (WSDA-OR) that can be adapted using target domain videos with coarse labels provided on a periodic basis that can significantly improve performance over the state-of-the-art models, allowing to achieve a greater pain localization accuracy.
01 Jul 2021
TL;DR: In this article, the smooth effects-on-response penalty (SERP) is used as a connecting link between proportional and fully non-proportional odds models, assuming that parameters of the latter vary smoothly over response categories.
Abstract: The so-called proportional odds assumption is popular in cumulative, ordinal regression. In practice, however, such an assumption is sometimes too restrictive. For instance, when modeling the perception of boar taint on an individual level, it turns out that, at least for some subjects, the effects of predictors (androstenone and skatole) vary between response categories. For more flexible modeling, we consider the use of a ‘smooth-effects-on-response penalty’ (SERP) as a connecting link between proportional and fully non-proportional odds models, assuming that parameters of the latter vary smoothly over response categories. The usefulness of SERP is further demonstrated through a simulation study. Besides flexible and accurate modeling, SERP also enables fitting of parameters in cases where the pure, unpenalized non-proportional odds model fails to converge.
06 Jun 2021
TL;DR: Wang et al. as mentioned in this paper proposed a meta ordinal weighting network (MOW-Net) to explicitly align each training sample with a meta-ordinal set (MOS) containing a few samples from all classes.
Abstract: The progression of lung cancer implies the intrinsic ordinal relationship of lung nodules at different stages—from benign to unsure then to malignant. This problem can be solved by ordinal regression methods, which is between classification and regression due to its ordinal label. However, existing convolutional neural network-based ordinal regression methods only focus on modifying classification head based on a randomly sampled mini-batch of data, ignoring the ordinal relationship resided in the data itself. In this paper, we propose a Meta Ordinal Weighting Network (MOW-Net) to explicitly align each training sample with a meta ordinal set (MOS) containing a few samples from all classes. During the training process, the MOW-Net learns a mapping from samples in MOS to corresponding class-specific weight. We further propose a meta cross-entropy loss to optimize the network in a meta-learning scheme. Experimental results demonstrate that the MOW-Net achieves better accuracy than the state-of-the-art ordinal regression methods, especially for the unsure class.
TL;DR: This work proposes integrating a dimensionality reduction technique, Principal Component Analysis, and Robust Ordinal Regression methods, to reduce the problem dimensionality and ensure the actual problem features are considered.
Abstract: Life Cycle Assessment quantifies the multi-dimensional impact of goods and services and can be handled by Multi-Criteria Decision Analysis. In Multi-Criteria Decision Analysis, Robust Ordinal Regression manages all the compatible preference functions at once when assessing a set of alternatives and a group of preferences on reference alternatives. Robust Ordinal Regression is thus a versatile method of reducing the cognitive effort required by decision makers for eliciting their preference structures in Life Cycle Assessment, although it does not directly operate on noisy alternatives and requires Stochastic Multicriteria Acceptability Analysis to deal with such scenarios. We propose integrating a dimensionality reduction technique, Principal Component Analysis, and Robust Ordinal Regression methods, to reduce the problem dimensionality and ensure the actual problem features are considered. A generated dataset, a dataset from literature and a Life Cycle Assessment case study are used to test the effectiveness of the proposed methods.
TL;DR: A special SVOR formulation is highlighted whose thresholds are described implicitly, so that the dual formulation is concise to apply the state-of-the-art asynchronous parallel coordinate descent algorithm, such as AsyGCD.
TL;DR: In this paper, the authors investigated the extent to which subjectively measured contextual factors contribute to PSS usefulness in smart cities and pointed out that only when PSS users can identify the critical contextual factors that are favorable and unfavorable, will the potential benefits of PSS for spatial planning be fully achieved.
Abstract: Contextual factors have been consistently argued as influencing the usefulness of planning support systems (PSS). Whereas previous studies were mostly conducted within a single planning project or based on experimental workshops, the present study looked at the application of PSS in smart city projects worldwide, and investigated the extent to which subjectively measured contextual factors contribute to PSS usefulness in smart cities. Based on a recent international questionnaire (268 respondents) designed to gather the perceptions of scholars and practitioners in the smart city realm, an ordinal regression model was fitted to assess the associations between the argued contextual factors and PSS usefulness. The results show that, in general, four contextual factors—namely the characteristics of the technology itself, user characteristics, characteristics of the planning process, and political context—have a significant influence on the usefulness of PSS, and that their impacts vary significantly. This paper emphasizes that only when PSS users can identify the critical contextual factors that are favorable and unfavorable, will the potential benefits of PSS for spatial planning be fully achieved.
TL;DR: A novel Bayesian Ordinal Regression approach for multiple criteria choice and ranking problems that employs an additive value function model to represent indirect Decision Maker’s preferences in the form of pairwise comparisons of reference alternatives to characterize the uncertainty in consequence of applying indirect preference information.
TL;DR: A regularized multi-task ordinal regression model for smaller datasets and a deep neural networks based MTOR model for large-scale datasets are developed and experiments indicate that the proposed MTOR models markedly improve the prediction performance comparing with single-task Ordinal regression models.
Abstract: Many real-world datasets are labeled with natural orders, i.e., ordinal labels. Ordinal regression is a method to predict ordinal labels that finds a wide range of applications in data-rich domains, such as natural, health and social sciences. Most existing ordinal regression approaches work well for independent and identically distributed (IID) instances via formulating a single ordinal regression task. However, for heterogeneous non-IID instances with well-defined local geometric structures, e.g., subpopulation groups, multi-task learning (MTL) provides a promising framework to encode task (subgroup) relatedness, bridge data from all tasks, and simultaneously learn multiple related tasks in efforts to improve generalization performance. Even though MTL methods have been extensively studied, there is barely existing work investigating MTL for heterogeneous data with ordinal labels. We tackle this important problem via sparse and deep multi-task approaches. Specifically, we develop a regularized multi-task ordinal regression (MTOR) model for smaller datasets and a deep neural networks based MTOR model for large-scale datasets. We evaluate the performance using three real-world healthcare datasets with applications to multi-stage disease progression diagnosis. Our experiments indicate that the proposed MTOR models markedly improve the prediction performance comparing with single-task ordinal regression models.
TL;DR: In this article, a quantitative cross-sectional study design with 633 randomly selected intercity bus passengers was conducted using a structured questionnaire in Kumasi, Ghana, where Ordinal regression was employed to determine the predictors of self-reported seatbelt use.
Abstract: Seat-belt use is effective in preventing traffic fatalities and injuries yet its use is not universal. This study sought to determine the predictors of self-reported seat-belt use among bus passengers in Ghana based on the theory of planned behaviour and health belief model. A quantitative cross-sectional study design with 633 randomly selected intercity bus passengers was conducted using a structured questionnaire in Kumasi, Ghana. The resulting data were analysed using SPSS version 23.0. Ordinal regression was employed to determine the predictors of self-reported seat-belt use. Majority of the respondents were male (61.5%) with a mean age of 32.2 (SD = 11.6). A third (33.0%) reported that they always wear their seat-belt as bus passengers. The results indicated that intention (OR = 1.49, 95% CI = 1.21–1.84, p = 0.001), subjective norm (OR = 1.57, 95% CI = 1.15–2.13, p = 0.004) and perceived behavioural control (OR = 1.53; 95% CI = 1.21–1.92, p = 0.001) variables from the theory of planned behaviour were significant independent predictors of seat-belt use. Among the health belief model variables, perceived severity (OR = 1.57, 95% CI = 1.15–2.16, p = 0.005) and perceived barriers (OR = 0.52, 95% CI = 0.39–0.67, p = 0.001) were the only significant independent predictors of self-reported seat-belt use. The findings suggest that intention, subjective norm, perceived behavioural control, perceived severity and perceived barriers play an important role in determining bus passengers’ seat-belt use behaviour. Road safety programmes to increase seat-belt use will gain from giving serious attention to these factors in the design and implementation of such programmes.
TL;DR: An ordinal approach to learning a distance, called chain maximizing ordinal metric learning, based on the maximization of ordered sequences in local neighborhoods of the data, which is able to adapt to data for which the class separations are not clear.
Abstract: The purpose of this paper is to introduce a new distance metric learning algorithm for ordinal regression. Ordinal regression addresses the problem of predicting classes for which there is a natural ordering, but the real distances between classes are unknown. Since ordinal regression walks a fine line between standard regression and classification, it is a common pitfall to either apply a regression-like numerical treatment of variables or underrate the ordinal information applying nominal classification techniques. On a different note, distance metric learning is a discipline that has proven to be very useful when improving distance-based algorithms such as the nearest neighbors classifier. In addition, an appropriate distance can enhance the explainability of this model. In our study we propose an ordinal approach to learning a distance, called chain maximizing ordinal metric learning. It is based on the maximization of ordered sequences in local neighborhoods of the data. This approach takes into account all the ordinal information in the data without making use of any of the two extremes of classification or regression, and it is able to adapt to data for which the class separations are not clear. We also show how to extend the algorithm to learn in a non-linear setup using kernel functions. We have tested our algorithm on several ordinal regression problems, showing a high performance under the usual evaluation metrics in this domain. Results are verified through Bayesian non-parametric testing. Finally, we explore the capabilities of our algorithm in terms of explainability using the case-based reasoning approach. We show these capabilities empirically on two different datasets, experiencing significant improvements over the case-based reasoning with the traditional Euclidean nearest neighbors.
TL;DR: In this article, the authors employed the ordinal regression model to predict different categories of CDHEs during July-August in northeast China (NEC) based on preceding El Nino-Southern Oscillation (or ENSO) for the period from 1952 to 2012.
TL;DR: In this paper, determining factors of the academic performance of secondary school students in the Department of Narino, Colombia were analyzed based on the results of standardized testing for secondary education (ICFES-SABER 11) conducted in 2018.
Abstract: This article seeks to illustrate the determining factors of the academic performance of secondary school students in the Department of Narino, Colombia. Two ordinal regression models were gauged (logit and probit) and arranged based on the results of standardized testing for secondary education (ICFES-SABER 11) conducted in 2018. The dependent variable used is polytomous, corresponding to the ranking from lowest to highest of the scores obtained while taking the following factors into consideration: parents' education levels, socioeconomic status, sex, access to learning technologies, the legal conditions of educational institutions, geographical location, and the number of weekly hours students dedicate to completing complementary activities. The results show that access to technological learning tools such as computers and internet connection, the highest educational level of parents, gender (male), and studying in an official urban educational institution increases the probability of obtaining better academic performance.