Machine Learning Classification System Research Articles

Computed tomography enterography radiomics and machine learning for identification of Crohn’s disease

BackgroundCrohn’s disease is a severe chronic and relapsing inflammatory bowel disease. Although contrast-enhanced computed tomography enterography is commonly used to evaluate crohn’s disease, its imaging findings are often nonspecific and can overlap with other bowel diseases. Recent studies have explored the application of radiomics-based machine learning algorithms to aid in the diagnosis of medical images. This study aims to develop a non-invasive method for detecting bowel lesions associated with Crohn’s disease using CT enterography radiomics and machine learning algorithms.MethodsA total of 139 patients with pathologically confirmed Crohn’s disease were retrospectively enrolled in this study. Radiomics features were extracted from both arterial- and venous-phase CT enterography images, representing both bowel lesions with Crohn’s disease and segments of normal bowel. A machine learning classification system was constructed by combining six selected radiomics features with eight classification algorithms. The models were trained using leave-one-out cross-validation and evaluated for accuracy.ResultsThe classification model demonstrated robust performance and high accuracy, with an area under the curve of 0.938 and 0.961 for the arterial- and venous-phase images, respectively. The model achieved an accuracy of 0.938 for arterial-phase images and 0.961 for venous-phase images.ConclusionsThis study successfully identified a radiomics machine learning method that effectively differentiates Crohn’s disease bowel lesions from normal bowel segments. Further studies with larger sample sizes and external cohorts are needed to validate these findings.

Emotion-Aware In-Car Feedback: A Comparative Study

We investigate personalised feedback mechanisms to help drivers regulate their emotions, aiming to improve road safety. We systematically evaluate driver-preferred feedback modalities and their impact on emotional states. Using unobtrusive vision-based emotion detection and self-labeling, we captured the emotional states and feedback preferences of 21 participants in a simulated driving environment. Results show that in-car feedback systems effectively influence drivers’ emotional states, with participants reporting positive experiences and varying preferences based on their emotions. We also developed a machine learning classification system using facial marker data to demonstrate the feasibility of our approach for classifying emotional states. Our contributions include design guidelines for tailored feedback systems, a systematic analysis of user reactions across three feedback channels with variations, an emotion classification system, and a dataset with labeled face landmark annotations for future research.

PCSVD: A hybrid feature extraction technique based on principal component analysis and singular value decomposition

<p>Feature extraction plays an important role in accurate preprocessing and real-world applications. High-dimensional features in the data have a significant impact on the machine learning classification system. Relevant feature extraction is a fundamental step not only to reduce the dimensionality but also to improve the performance of the classifier. In this paper, the author proposes a hybrid dimensionality reduction technique using principal component analysis (PCA) and singular value decomposition (SVD) in a machine classification system with a support vector classifier (SVC). To evaluate the performance of PCSVD, the results are compared without using feature extraction techniques or with existing methods of independent component analysis (ICA), PCA, linear discriminant analysis (LDA), and SVD. In addition, the efficiency of the PCSVD method is measured on an increased scale of 1.54% accuracy, 2.70% sensitivity, 3.71% specificity, and 3.58% precision. In addition, reduce the 15% dimensionality and 40.60% RMSE, which are better than existing techniques found in the literature.</p>

Building machine-learning models for reducing the severity of bicyclist road traffic injuries

Predicting the severity of injuries caused by traffic accidents is an important undertaking because it may lead to establishing regulations increasing road-user safety. Bicyclists are a particularly susceptible category of road users, which is especially troubling considering the environmental, financial, and health benefits of this mode of transportation. As a result, this study aims to apply machine learning to identify risk variables that may result in serious biker injuries in the case of an accident.Machine-learning models make no assumptions about the connections between variables. Hence, it has been argued that machine-learning approaches produce better outcomes than statistical procedures. This study selects the “best” machine-learning classification system from a vast pool of similar algorithms to predict the severity of bicycling injuries. Machine learning allows the system to learn from experience and improve without being too programmed.We first use a variety of feature selection algorithms to identify a list of features related to the accident and the environment that have the greatest impact on the severity of bicyclist injuries. This feature list is then used as input data to various machine-learning algorithms that predict the class of bicyclist injury severity at one of three levels (fatal, serious, and slight). The “best” machine-learning algorithm is identified on the basis of having the highest levels of accuracy, precision, recall, and F1 score. The current models were developed and trained based on Israeli road-traffic accident data from 2009 to 2019, meaning that new models would need to be developed for other geographical locations. In addition, the models would need to be updated to take account of the changing relationships between motorists, bicyclists, and the environment. Nevertheless, the proposed methodology has universal applicability.

Improving Depression Prediction Accuracy Using Fisher Score-Based Feature Selection and Dynamic Ensemble Selection Approach Based on Acoustic Features of Speech

Depression affects over 322 million people, and it is the most common source of disability worldwide. Literature in speech processing revealed that speech could be used for detecting depression. Depressed individuals exhibit varied acoustic characteristics compared to non-depressed. A four-staged machine learning classification system is developed to investigate the acoustic parameters to detect depression. Stage one uses speech recordings from a publicly available and clinically validated dataset DAIC-WOZ. The baseline acoustic feature vector, eGeMAPS, is extracted from the dataset in stage two. Adaptive synthetic (ADASYN) is performed along with data preprocessing to overcome the class imbalance. In stage three, we conducted feature selection (FS) using three techniques; Boruta FS, recursive feature elimination using support vector machine (SVM-RFE), and the fisher score-based FS. Experimentation with various machine learning base classifiers like gaussian naïve bayes (GNB), support vector machine (SVM), k-nearest neighbors (KNN), logistic regression (LR), and random forest classifier (RF) is performed in stage four. The hyperparameters of the classifiers are tuned using the GridSearchCV technique throughout the 10-fold stratified cross-validation (CV). Then we employed multiple dynamic ensemble selection of classifier algorithms (DES) with k=3 and k=5 utilizing the pool of aforementioned four base classifiers to improve the accuracy. We present a comparative study using eGeMAPS features against the base classifiers and the experimented DES classifiers. Our results on the DAIC-WOZ benchmark dataset suggested that K-Nearest Oracles Union (KNORA-U) DES with k=3 has superior accuracy using a subset of 15 features selected by fisher score-based FS than the individual base classifiers.

Hybrid feature selection-based machine learning Classification system for the prediction of injury severity in single and multiple-vehicle accidents.

To undertake a reliable analysis of injury severity in road traffic accidents, a complete understanding of important attributes is essential. As a result of the shift from traditional statistical parametric procedures to computer-aided methods, machine learning approaches have become an important aspect in predicting the severity of road traffic injuries. The paper presents a hybrid feature selection-based machine learning classification approach for detecting significant attributes and predicting injury severity in single and multiple-vehicle accidents. To begin, we employed a Random Forests (RF) classifier in conjunction with an intrinsic wrapper-based feature selection approach called the Boruta Algorithm (BA) to find the relevant important attributes that determine injury severity. The influential attributes were then fed into a set of four classifiers to accurately predict injury severity (Naive Bayes (NB), K-Nearest Neighbor (K-NN), Binary Logistic Regression (BLR), and Extreme Gradient Boosting (XGBoost)). According to BA’s experimental investigation, the vehicle type was the most influential factor, followed by the month of the year, the driver’s age, and the alignment of the road segment. The driver’s gender, the presence of a median, and the presence of a shoulder were all found to be unimportant. According to classifier performance measures, XGBoost surpasses the other classifiers in terms of prediction performance. Using the specified attributes, the accuracy, Cohen’s Kappa, F1-Measure, and AUC-ROC values of the XGBoost were 82.10%, 0.607, 0.776, and 0.880 for single vehicle accidents and 79.52%, 0.569, 0.752, and 0.86 for multiple-vehicle accidents, respectively.

Dementia ConnEEGtome: Towards multicentric harmonization of EEG connectivity in neurodegeneration

The proposal to use brain connectivity as a biomarker for dementia phenotyping can be potentiated by conducting large-scale multicentric studies using high-density electroencephalography (hd- EEG). Nevertheless, several barriers preclude the development of a systematic “ConnEEGtome” in dementia research. Here we review critical sources of variability in EEG connectivity studies, and provide general guidelines for multicentric protocol harmonization. We describe how results can be impacted by the choice for data acquisition, and signal processing workflows. The implementation of a particular processing pipeline is conditional upon assumptions made by researchers about the nature of EEG. Due to these assumptions, EEG connectivity metrics are typically applicable to restricted scenarios, e.g., to a particular neurocognitive disorder. “Ground truths” for the choice of processing workflow and connectivity analysis are impractical. Consequently, efforts should be directed to harmonizing experimental procedures, data acquisition, and the first steps of the preprocessing pipeline. Conducting multiple analyses of the same data and a proper integration of the results need to be considered in additional processing steps. Furthermore, instead of using a single connectivity measure, using a composite metric combining different connectivity measures brings a powerful strategy to scale up the replicability of multicentric EEG connectivity studies. These composite metrics can boost the predictive strength of diagnostic tools for dementia. Moreover, the implementation of multi-feature machine learning classification systems that include EEG-based connectivity analyses may help to exploit the potential of multicentric studies combining clinical-cognitive, molecular, genetics, and neuroimaging data towards a multi-dimensional characterization of the dementia.

Data analysis methods in astronomic objects classification (Sloan Digital Sky Survey DR14)

In the paper Sloan Digital Sky Survey DR14 dataset was investigated. It contains statistical information about many astronomical objects. The information was obtained within the framework of the Sloan Digital Sky Survey project. There are telescopes at the Earth surface, at the Earth orbit and in the Lagrange points of some systems (Earth–Moon, Sun–Earth). The telescopes gain information in different frequency ranges. The large quantity of statistical information leads to the demand for analytical algorithms and systems capable of making classification. Such information is marked up well enough to build machine learning classification systems. The paper presents the results of a number of classifiers. The handled data contains measures of three types of astronomical objects of the Sloan Digital Sky Survey DR14 dataset (star, quasar, galaxy). The CART decision tree, logistic regression, naïve Bayes classifiers and ensembles of classifiers (random forest, gradient boosting) were implemented. Conclusions about special features of each machine learning classifier trained to solve this task are made at the end of the paper. In some cases, classifiers’ structure can be explained physically. The accuracy of the classifiers built in this research is more than 90% (metrics F1, precision and recall are implemented, because the classes are unbalanced). Taking these values into account classification task is supposed to be successfully solved. At the same time, the structure of classifiers and importance of features can be used as a physical explanation of the solution.

Understanding quality of analytics trade-offs in an end-to-end machine learning-based classification system for building information modeling

Optimizing quality trade-offs in an end-to-end big data science process is challenging, as not only do we need to deal with different types of software components, but also the domain knowledge has to be incorporated along the process. This paper focuses on methods for tackling quality trade-offs in a common data science process for classifying Building Information Modeling (BIM) elements, an important task in the architecture, engineering, and construction industry. Due to the diversity and richness of building elements, machine learning (ML) techniques have been increasingly investigated for classification tasks. However, ML-based classification faces many issues, w.r.t. vast amount of data with heterogeneous data quality, diverse underlying computing configurations, and complex integration with industrial BIM tools, in an end-to-end BIM data analysis. In this paper, we develop an end-to-end ML classification system in which quality of analytics is considered as the first-class feature across different phases, from data collection, feature processing, training to ML model serving. We present our method for studying the quality of analytics trade-offs and carry out experiments with BIM data extracted from Solibri to demonstrate the automation of several tasks in the end-to-end ML classification. Our results have demonstrated that the quality of data, data extraction techniques, and computing configurations must be carefully designed when applying ML classifications for BIM in order to balance constraints of time, cost, and prediction accuracy. Our quality of analytics methods presents generic steps and considerations for dealing with such designs, given the time, cost, and accuracy trade-offs required in specific contexts. Thus, the methods could be applied to the design of end-to-end BIM classification systems using other ML techniques and cloud services.

Consensus molecular subtyping of colorectal cancers is influenced by goblet cell content

A critical obstacle in the field of colorectal cancer (CRC) is the establishment of precise tumor subtypes to facilitate the development of targeted therapeutic regimens. While dysregulated mucin production is a histopathological feature of multiple CRC subtypes, it is not clear how well these pathologies are associated with the proportion of goblet cells in the tumor, or whether or not this proportion is variable across all CRC. This study demonstrates that consensus molecular subtype 3 (CMS3) CRC tumors and cell lines are enriched for the expression of goblet cell marker genes. Further, the proportion of goblet cells in the tumor is associated with the probability of CMS3 subtype assignment and these CMS3 subtype tumors are mutually exclusive from mucinous adenocarcinoma pathologies. This study provides proof of principle for the use of machine learning classification systems to subtype tumors based on cellular content, and provides further context regarding the features weighing CMS3 subtype assignment.

Sensors, Features, and Machine Learning for Oil Spill Detection and Monitoring: A Review

Remote sensing technologies and machine learning (ML) algorithms play an increasingly important role in accurate detection and monitoring of oil spill slicks, assisting scientists in forecasting their trajectories, developing clean-up plans, taking timely and urgent actions, and applying effective treatments to contain and alleviate adverse effects. Review and analysis of different sources of remotely sensed data and various components of ML classification systems for oil spill detection and monitoring are presented in this study. More than 100 publications in the field of oil spill remote sensing, published in the past 10 years, are reviewed in this paper. The first part of this review discusses the strengths and weaknesses of different sources of remotely sensed data used for oil spill detection. Necessary preprocessing and preparation of data for developing classification models are then highlighted. Feature extraction, feature selection, and widely used handcrafted features for oil spill detection are subsequently introduced and analyzed. The second part of this review explains the use and capabilities of different classical and developed state-of-the-art ML techniques for oil spill detection. Finally, an in-depth discussion on limitations, open challenges, considerations of oil spill classification systems using remote sensing, and state-of-the-art ML algorithms are highlighted along with conclusions and insights into future directions.

TargetDB: A target information aggregation tool and tractability predictor.

When trying to identify new potential therapeutic protein targets, access to data and knowledge is increasingly important. In a field where new resources and data sources become available every day, it is crucial to be able to take a step back and look at the wider picture in order to identify potential drug targets. While this task is routinely performed by bespoke literature searches, it is often time-consuming and lacks uniformity when comparing multiple targets at one time. To address this challenge, we developed TargetDB, a tool that aggregates public information available on given target(s) (links to disease, safety, 3D structures, ligandability, novelty, etc.) and assembles it in an easy to read output ready for the researcher to analyze. In addition, we developed a target scoring system based on the desirable attributes of good therapeutic targets and machine learning classification system to categorize novel targets as having promising or challenging tractrability. In this manuscript, we present the methodology used to develop TargetDB as well as test cases.

Stepping Inside the Classification Cube: An Intimate Interaction with an AI System

The Classification Cube art installation invites participants to become familiar with a machine-learning classification system. Inside a private space within the gallery, participants’ bodies are subjected to a classification process that detects their faces and estimates their age, gender, emotion and actions. Participants are also able to see how their own classification compares with how the installation classifies a series of animated figures. Rapidly changing results encourage participants to actively perform their behavior to the system and alter the way it “sees” them. The entanglement with the system raises awareness regarding the effectiveness of machine interpretation.

Toward mitigating pressure injuries: Detecting patient orientation from vertical bed reaction forces.

IntroductionProlonged bed rest without repositioning can lead to pressure injuries. However, it can be challenging for caregivers and patients to adhere to repositioning schedules. A device that alerts caregivers when a patient has remained in the same orientation for too long may reduce the incidence and/or severity of pressure injuries. This paper proposes a method to detect a person’s orientation in bed using data from load cells placed under the legs of a hospital grade bed.MethodsTwenty able-bodied individuals were positioned into one of three orientations (supine, left side-lying, or right side-lying) either with no support, a pillow, or a wedge, and the head of the bed either raised or lowered. Breathing pattern characteristics extracted from force data were used to train two machine learning classification systems (Logistic Regression and Feed Forward Neural Network) and then evaluate for their ability to identify each participant’s orientation using a leave-one-participant-out cross-validation.ResultsThe Feed Forward Neural Network yielded the highest orientation prediction accuracy at 94.2%.ConclusionsThe high accuracy of this non-invasive system’s ability to a participant’s position in bed shows potential for this algorithm to be useful in developing a pressure injury prevention tool.

Detection of Financial Opportunities in Micro-Blogging Data With a Stacked Classification System

Micro-blogging sources such as the Twitter social network provide valuable real-time data for market prediction models. Investors’ opinions in this network follow the fluctuations of the stock markets and often include educated speculations on market opportunities that may have impact on the actions of other investors. In view of this, we propose a novel system to detect positive predictions in tweets, a type of financial emotions which we term “opportunities” that are akin to “anticipation” in Plutchik’s theory. Specifically, we seek a high detection precision to present a financial operator a substantial amount of such tweets while differentiating them from the rest of financial emotions in our system. We achieve it with a three-layer stacked Machine Learning classification system with sophisticated features that result from applying Natural Language Processing techniques to extract valuable linguistic information. Experimental results on a dataset that has been manually annotated with financial emotion and ticker occurrence tags demonstrate that our system yields satisfactory and competitive performance in financial opportunity detection, with precision values up to 83%. This promising outcome endorses the usability of our system to support investors’ decision making.

A Method for Sensor Reduction in a Supervised Machine Learning Classification System

Smart devices employing interconnected sensors for feedback and control are being rapidly adopted. Many useful applications for these devices are in markets that demand cost-conscious solutions. Traditional machine-learning-based control systems often rely on multiple measurements from many sensors to achieve performance targets. An alternative method is presented that leverages a time-series output produced by a single sensor. By using domain expert knowledge, the time-series output is discretized into finite intervals that correspond to the physical events occurring in the system. Statistical measures are taken across these intervals to serve as the features to the machine learning system. Additional features that decouple key physical metrics are identified, improving the performance of the system. This novel approach requires a more modest dataset and does not compromise performance. The resulting development effort is significantly more cost-effective than traditional sensor classification systems, not only due to the reduced sensor count, but also due to a significantly simplified and more robust algorithm development and testing step. Results are presented with the case study of a media-type classification system within a printing system, which was deployed to the field as a commercial product.

P53 - Artificial Intelligence can Predict Postoperative Complications Better than Traditional Statistical Testing Following Posterior Lumbar Fusion

BACKGROUND CONTEXT: Current clinical research relies on statistical models to identify independent risk factors of postoperative complications. However, complex interplay between risk factors is rarely accounted for, which can lead to inaccurate patient morbidity and mortality. Neural network is a machine learning classification system inspired by the human brain. Each network contains a large cluster of neurons which collectively but uniquely weigh the importance of input variables. Optimization of each individual neuron allows for the system to “learn” through repetitive epochs and minimizes error.

179 - Artificial Intelligence (AI) Can Predict Complications Better than Traditional Statistical Testing Following Fusion for Anterior Lumbar Fusion (ALF)

BACKGROUND CONTEXT: Current clinical research relies on statistical models to identify independent risk factors of postoperative complications. However, complex interplay between risk factors is rarely accounted for, which can lead to inaccurate patient morbidity and mortality. Neural network is a machine learning classification system inspired by the human brain. Each network contains a large cluster of neurons which collectively but uniquely weigh the importance of input variables. Optimization of each individual neuron allows for the system to “learn” through repetitive epochs and minimizes error.

A Feedback-Controlled Mandibular Positioner Identifies Individuals With Sleep Apnea Who Will Respond to Oral Appliance Therapy.

Mandibular protruding oral appliances represent a potentially important therapy for obstructive sleep apnea (OSA). However, their clinical utility is limited by a less-than-ideal efficacy rate and uncertainty regarding an efficacious mandibular position, pointing to the need for a tool to assist in delivery of the therapy. The current study assesses the ability to prospectively identify therapeutic responders and determine an efficacious mandibular position. Individuals (n = 202) with OSA participated in a blinded, 2-part investigation. A system for identifying therapeutic responders was developed in part 1 (n = 149); the predictive accuracy of this system was prospectively evaluated on a new population in part 2 (n = 53). Each participant underwent a 2-night, in-home feedback-controlled mandibular positioner (FCMP) test, followed by treatment with a custom oral appliance and an outcome study with the oral appliance in place. A machine learning classification system was trained to predict therapeutic outcome on data obtained from FCMP studies on part 1 participants. The accuracy of this trained system was then evaluated on part 2 participants by examining the agreement between prospectively predicted outcome and observed outcome. A predicted efficacious mandibular position was derived from each FCMP study. Predictive accuracy was as follows: sensitivity 85%; specificity 93%; positive predictive value 97%; and negative predictive value 72%. Of participants correctly predicted to respond to therapy, the predicted mandibular protrusive position proved efficacious in 86% of cases. An unattended, in-home FCMP test prospectively identifies individuals with OSA who will respond to oral appliance therapy and provides an efficacious mandibular position. The trial that this study reports on is registered on www.clinicaltrials.gov, ID NCT03011762, study name: Feasibility and Predictive Accuracy of an In-Home Computer Controlled Mandibular Positioner in Identifying Favourable Candidates for Oral Appliance Therapy.

Late fusion of facial dynamics for automatic expression recognition

Installment of a facial expression is associated with contractions and extensions of specific facial muscles. Noting that expression is about changes, we present a model for expression classification based on facial landmarks dynamics. Our model isolates the trajectory of facial fiducial points by wrapping them up in relevant features and discriminating among various alternatives with a machine learning classification system. The used features are geometric and temporal-based and the classification system is represented by a late fusion framework that combines several neural networks with binary responses. The proposed method is robust, being able to handle complex expression classes.