Categories Of Features Research Articles

Machine learning models for temporally precise lapse prediction in alcohol use disorder.

We developed three machine learning models that predict hour-by-hour probabilities of a future lapse back to alcohol use with increasing temporal precision (i.e., lapses in the next week, next day, and next hour). Model features were based on raw scores and longitudinal change in theoretically implicated risk factors collected through ecological momentary assessment. Participants (N = 151, 51% male, Mage = 41, 87% White, 97% non-Hispanic) in early recovery (1-8 weeks of abstinence) from alcohol use disorder provided 4 × daily ecological momentary assessment for up to 3 months. We used grouped, nested cross-validation to select the best models and evaluate the performance of those best models. Models yielded median areas under the receiver operating curves of 0.89, 0.90, and 0.93 in the 30 held-out test sets for week-, day-, and hour-level models, respectively. Some feature categories consistently emerged as being globally important to lapse prediction across our week-, day-, and hour-level models (i.e., past use, future self-efficacy). However, most of the more punctate, time-varying constructs (e.g., craving, past stressful events, arousal) appear to have a greater impact within the next-hour prediction model. This research represents an important step toward the development of a smart (machine learning guided) sensing system that can both identify periods of peak lapse risk and recommend specific supports to address factors contributing to this risk. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

CLIP-FSSC: A transferable visual model for fish and shrimp species classification based on natural language supervision

Fish and shrimp species classification is a practical need in the field of aquaculture. Traditional classification method includes extracting modal features of the single image and training them on downstream datasets. However, this method has the disadvantage of requiring manual annotated image data and significant training time. To address these issues, this paper introduces a method named CLIP-FSSC (Contrastive Language–Image Pre-training for Fish and Shrimp Species Classification) for zero-shot prediction using a pre-trained model. The proposed method aims to classify fish and shrimp species in the field of aquaculture using a multimodal pre-trained model that utilizes semantic text description as an image supervision signal for transfer learning. In the downstream fish dataset, we use natural language labels for three types of fish - grass carp, common carp, and silver carp. We extract text category features using a transformer and compare the results obtained from three different CLIP-based backbones for the image modality - vision transformer, Resnet50, and Resnet101. We compare the performance of these models with previous methods that performed well. After performing zero-shot predictions on samples of the three types of fish, we achieve similar or even better classification accuracy than models trained on downstream fish datasets. Our experiment results show an accuracy of 98.77 %, and no new training process is required. This proves that using the semantic text modality as the label for the image modality can effectively classify fish species. To demonstrate the effectiveness of this method on other species in the field of aquaculture, we collected two sets of shrimp data - prawn and cambarus. Through zero-shot prediction, we achieve the highest classification accuracy of 92.00 % for these two types of shrimp datasets. Overall, our results demonstrate that using a multimodal pre-trained model with semantic text description as an image supervision signal for transfer learning can effectively classify fish and shrimp species with high accuracy, while reducing the need for manual annotation and training time.

HQ-Net: A heatmap-based query backbone for point cloud understanding

With the advancement of sensor technology, including LiDAR and RGB-D cameras, tasks related to point cloud understanding have gained widespread application. The sampling and learning of feature points play a crucial role in point cloud understanding, representing a fundamental and essential aspect of 3D computer vision. In the realm of point cloud processing, traditional methods such as Farthest Point Sampling (FPS) are frequently employed to select feature points. These points are then used in conjunction with effective feature aggregation operators to facilitate reasoning. However, these traditional sampling methods primarily focus on the Euclidean distance between points, lacking a comprehensive consideration of point cloud features. To address this limitation, we introduce HQ-Net, a novel point cloud understanding model designed to optimize feature point sampling using a Heatmap-based Query Backbone. The Heatmap-based Query Backbone utilizes the Heatmap Module to select category-specific point features, effectively generating a heatmap tailored to the category features and automatically extracting features from the heatmap. Subsequently, the proposed Query Encoder is employed to facilitate the interaction between point features and query features, providing mutual guidance and enhancing the model’s ability to capture spatial structure and semantic information within point clouds. Additionally, we propose task-specific heads, such as classification and segmentation heads for different tasks, which can be seamlessly integrated with the backbone to achieve excellent performance. In various point cloud understanding tasks, including shape classification, part segmentation, and semantic segmentation, HQ-Net demonstrates excellent performance.

EXPLANA: A user-friendly workflow for EXPLoratory ANAlysis and feature selection in cross-sectional and longitudinal microbiome studies.

Longitudinal microbiome studies (LMS) are increasingly common but have analytic challenges including non-independent data requiring mixed-effects models and large amounts of data that motivate exploratory analysis to identify factors related to outcome variables. Although change analysis (i.e. calculating deltas between values at different timepoints) can be powerful, how to best conduct these analyses is not always clear. For example, observational LMS measurements show natural fluctuations, so baseline might not be a reference of primary interest; whereas, for interventional LMS, baseline is a key reference point, often indicating the start of treatment. To address these challenges, we developed a feature selection workflow for cross-sectional and LMS that supports numerical and categorical data called EXPLANA (EXPLoratory ANAlysis). Machine-learning methods were combined with different types of change calculations and downstream interpretation methods to identify statistically meaningful variables and explain their relationship to outcomes. EXPLANA generates an interactive report that textually and graphically summarizes methods and results. EXPLANA had good performance on simulated data, with an average area under the curve (AUC) of 0.91 (range: 0.79-1.0, SD = 0.05), outperformed an existing tool (AUC: 0.95 vs. 0.56), and identified novel order-dependent categorical feature changes. EXPLANA is broadly applicable and simplifies analytics for identifying features related to outcomes of interest.

Protection of Family Values in Russian Law: Questions of Theory and Practice

Introduction: at the moment, Russia is further developing the concept of traditional values, which include family values. Nevertheless, a number of regulatory and protective norms are already aimed at solving this problem, which requires doctrinal research and reflection. There are also a number of controversial issues related to the formation of a mechanism of guarantees for the protection of family values, including economic, political, as well as spiritual and moral guarantees. The purpose of the study is to reveal the features and possible directions of legal regulation for the development of the family values system in Russia. Methods: the analysis method made it possible to identify the main features of the category “family values” and its relationship with “traditional values”; the synthesis method was used to generalize the typological features of the family legal categories used in the paper; and the analogy method made it possible to identify practical features of ensuring the protection of family values in Russia. The comparative law method allowed studying the foreign experience of protecting family values, which can be used for the development of Russian family legislation. Using the formal legal method, a number of features and specifics of the statutory regulation of the protection of family values were identified. Results: despite the lack of normative consolidation, a number of political and legal acts defining the main directions of state family policy, as well as the norms of federal laws containing rules of conduct based on family values and measures of their legal support, are devoted to the protection of family values. It seems advisable to amend Article 1 of the Family Code of the Russian Federation with a new principle dedicated to the protection of family values, which will make it possible to more clearly fix this vector of state policy. The implementation of this principle will take place both through legal means and through the use of economic, political, spiritual, moral, and other ways to influence the behavior of participants in public relations. Conclusions: at the moment, the Family Code of the Russian Federation and other laws contain regulatory and protective norms dedicated to the implementation and protection of the family rights of citizens of the Russian Federation. Family values are realized in them by creating conditions for marriage, regulating property relations in the family, determining the grounds and procedure for collecting alimony for children and other family members, and protecting legally free children. The protective norms are devoted to the prohibition of family violence and propaganda of homosexuality, the adoption of measures to curb the evasion of child support payments, etc.

Object-oriented multi-scale segmentation and multi-feature fusion-based method for identifying typical fruit trees in arid regions using Sentinel-1/2 satellite images

Fruit tree identification that is quick and precise lays the groundwork for scientifically evaluating orchard yields and dynamically monitoring planting areas. This study aims to evaluate the applicability of time series Sentinel-1/2 satellite data for fruit tree classification and to provide a new method for accurately extracting fruit tree species. Therefore, the study area selected is the Tarim Basin, the most important fruit-growing region in northwest China. The main focus is on identifying several major fruit tree species in this region. Time series Sentinel-1/2 satellite images acquired from the Google Earth Engine (GEE) platform are used for the study. A multi-scale segmentation approach is applied, and six categories of features including spectral, phenological, texture, polarization, vegetation index, and red edge index features are constructed. A total of forth-four features are extracted and optimized using the Vi feature importance index to determine the best time phase. Based on this, an object-oriented (OO) segmentation combined with the Random Forest (RF) method is used to identify fruit tree species. To find the best method for fruit tree identification, the results are compared with three other widely used traditional machine learning algorithms: Support Vector Machine (SVM), Gradient Boosting Decision Tree (GBDT), and Classification and Regression Tree (CART). The results show that: (1) the object-oriented segmentation method helps to improve the accuracy of fruit tree identification features, and September satellite images provide the best time window for fruit tree identification, with spectral, phenological, and texture features contributing the most to fruit tree species identification. (2) The RF model has higher accuracy in identifying fruit tree species than other machine learning models, with an overall accuracy (OA) and a kappa coefficient (KC) of 94.60% and 93.74% respectively, indicating that the combination of object-oriented segmentation and RF algorithm has great value and potential for fruit tree identification and classification. This method can be applied to large-scale fruit tree remote sensing classification and provides an effective technical means for monitoring fruit tree planting areas using medium-to-high-resolution remote sensing images.

Heterogeneity of social participation in patients three months after total knee arthroplasty: a latent profile analysis

BackgroundSocial participation is an important index of rehabilitation and social reintegration in patients after total knee arthroplasty (TKA). However, most existing studies focus on improving patients’ functioning and activities, with only a few examining the social participation among patients after TKA. Therefore, the study aims to investigate the heterogeneity of social participation in patients three months after TKA and analyze subgroup influencing factors, to promote functional exercise and postoperative follow-up in specific patients.MethodsThis cross-sectional study recruited 255 patients who underwent TKA in a Tertiary Hospital in Jinan City, China, from March to July 2022. Three months after having undergone TKA, participants’ data were collected using the Numeric Pain Rating Scale, the Chinese version of the Tampa Scale of Kinesiophobia, the 10-item Kessler Psychological Distress Scale, Hospital for Special Surgery Knee-rating Scale, and Impact on Participation and Autonomy Questionnaire. Latent profile analysis was used to identify categories of patients’ social participation. Multiple logistic regression analysis was used to analyze the influencing factors of the different subgroups.ResultsThree months after TKA, the patients were divided into three subgroups: low social participation group (17.9%), moderate social participation group (40.8%), and high social participation group (41.3%). The vast majority of patients who underwent TKA exhibited moderate-to-high level of social participation. The multiple logistic regression analysis results showed that age, degree of pain, knee function, and kinesiophobia were the influencing factors of the potential profiles of social participation in patients three months after TKA (p < 0.05).ConclusionThese results support a distinct categorical feature of social participation among patients three months after undergoing TKA. Medical staff need to provide targeted guidance according to the potential classification characteristics of social participation to improve the level of social participation and promote rehabilitation of patients.

Differential diagnosis of cystic and nodular lung diseases

Cystic and nodular lung diseases encompass abroad spectrum of diseases with different etiologies and clinicoradiological presentations. Their differentiation is crucial for patient management but can be complex due to diseases with features of both categories and overlapping radiological patterns. This study aims to describe the imaging features of cystic and nodular lung diseases in high-resolution computed tomography (CT) in detail-primarily based on their etiology-in order to allow amore accurate differential diagnosis of these diseases. Anarrative review based on current literature on the topic was conducted from aclinicoradiological perspective. This paper systematically categorizes the differential diagnosis of cystic and nodular lung disease and provides insights into their radiological patterns and etiologies. It highlights the role of CT in the diagnosis of these diseases and emphasizes the importance of multidisciplinary panels combining expertise from radiology, pulmonology, rheumatology, and pathology. Reliable differential diagnosis of cystic and nodular lung diseases, particularly based on their radiological features alone, remains difficult due to their overlapping and dynamic nature. Multidisciplinary boards should be the clinical standard for accurate work-up of these diseases, as they combine the medical history, symptoms, radiological findings, and, if necessary, histopathological examinations, thus providing amore robust framework for diagnosis and management.

DLRover-RM: Resource Optimization for Deep Recommendation Models Training in the Cloud

Deep learning recommendation models (DLRM) rely on large embedding tables to manage categorical sparse features. Expanding such embedding tables can significantly enhance model performance, but at the cost of increased GPU/CPU/memory usage. Meanwhile, tech companies have built extensive cloud-based services to accelerate training DLRM models at scale. In this paper, we conduct a deep investigation of the DLRM training platforms at AntGroup and reveal two critical challenges: low resource utilization due to suboptimal configurations by users and the tendency to encounter abnormalities due to an unstable cloud environment. To overcome them, we introduce DLRover, an elastic training framework for DLRMs designed to increase resource utilization and handle the instability of a cloud environment. DLRover develops a resource-performance model by considering the unique characteristics of DLRMs and a three-stage heuristic strategy to automatically allocate and dynamically adjust resources for DLRM training jobs for higher resource utilization. Further, DLRover develops multiple mechanisms to ensure efficient and reliable execution of DLRM training jobs. Our extensive evaluation shows that DLRover reduces job completion times by 31%, increases the job completion rate by 6%, enhances CPU usage by 15%, and improves memory utilization by 20%, compared to state-of-the-art resource scheduling frameworks. DLRover has been widely deployed at AntGroup and processes thousands of DLRM training jobs on a daily basis. DLRover is open-sourced and has been adopted by 10+ companies.

A novel green learning artificial intelligence model for regional electrical load prediction

Load prediction is crucial to unit commitment and scheduling in electricity systems; however, load prediction models consume considerable electricity. Most conventional prediction methods are based on deep learning (DL) models, such as long short-term memory and gated recurrent units. The nonlinear activation functions and backpropagation mechanisms used in these methods result in high computational complexity and correspondingly high energy consumption. This paper proposes a prediction model based on a green learning (GL) framework aimed at reducing energy consumption. The proposed GL model replaces the activation functions conventionally used for feature extraction with a hybrid scheme combining categorical and numerical features, such as seasonal climate features and multi-autoregression terms. The proposed GL model also eliminates the backpropagation methods used to optimize hyperparameters by employing seasonal centroids and the quantile auto-regression forest algorithm for classification/regression. In a case study, the proposed GL model significantly outperformed conventional DL models in terms of energy consumption without compromising accuracy. The energy savings of the proposed GL model are equivalent to the annual electricity consumption of 211 ∼ 565 households, increasing with model penetration. This paper demonstrates the importance of energy-saving models in mitigating the linkage effect of industrial carbon emissions. It also explores policy implications for the implementation of GL models.

Hyperspectral image classification using a deep relation network with random replacement data augmentation

ABSTRACT Achieving high classification accuracy under a limited number of training samples remains a key challenge in hyperspectral image (HSI) classification. To address this issue, we propose a random replacement data augmentation method. The proposed method randomly selects a rectangular spatial region in HSI, and the pixels in the selected region are replaced with data from different spectral dimensions randomly to generate new sample, increasing the diversity of training samples effectively. In addition, we also propose an improved deep metric learning-based relation network, which improves model’s ability to extract deep features and fully utilizes spatial information to establish relationships between feature maps and categories effectively. Experiments have shown that the proposed method can achieve more competitive results.

Prediction of Employee Reassignment Using Supervised Machine Learning Algorithms and Rule-Based Experts: A Case Study in Educational Institution

Education plays a crucial role in shaping the future of a nation. To maintain the quality of education, effective human resource management is essential in educational institutions. This study addresses the challenges of employee’s placement under the Educational Institution. According data from December 2021 to May 2024, only 2,452 out of 41,722 employees were reassignment, which is significantly below the target set by regulation. This study evaluates several supervised machine learning algorithms, including Gaussian Naive Bayes, Decision Tree, Support Vector Machine, and Random Forest. Random Forest emerges as the most suitable algorithm due to its superior accuracy, precision, recall, and F1 Score. Following the evaluation of the chosen algorithm, the deployment phase includes comprehensive data preprocessing steps, such as handling missing values, data normalization, and categorical feature encoding. This system integrates with Google API for geospatial data, ensuring accurate and efficient decision-making.

Antibiotic Prescription Rates For Respiratory Tract Infections Treated In The Outpatient Sector Based On Routine Data From A German Health Insurance Company

Acute uncomplicated upper respiratory tract infections (RTIs) are common reasons for antibiotic prescriptions in the outpatient sector, although>90% are of viral origins and mostly self-limiting. Germany has a low antibiotic prescription rate compared to other European countries, but there are regional differences. Disease-specific quality indicators (QI) developed by the European Surveillance of Antimicrobial Consumption Network (ESAC-Net) address the prescription rates (target <20%) and choice of antibiotic (target>80%) in a general practitioner population. The aim of this analysis was to operationalise the ESAC-Net-QI using secondary data from a health insurance company at the level of diseases and healthcare facilities, as well as to analyse areas for improvement. The underlying database comprises the panel physician billing, hospital and prescription data in accordance with German law (§§ 295, 300 and 301 SGB V) of the Barmer health insurance company for the year 2022. Categorical features are presented as frequency and percentage. Maximum specificity was aimed, hence potentially risky disease courses were excluded. The overall RTI prescription rate of antibiotics was 25% in 2022 before applying inclusion and exclusion criteria (1,197,568/4,720,786). After applying the algorithm, the prescription rate for all RTI cases dropped to 6% (80,786/1,365,646). When specific RTI indications are considered, 35% (13,465/38,913) of acute bronchitis and acute sinusitis cases (4,971/14,051) received an antibiotic prescription each. At the facility level, 38% of practices (1,396/3,705) treating acute bronchitis and 41% of practices (539/1,300) treating acute sinusitis achieved the target of<20% antibiotic prescriptions. Only 29% of all RTI cases with antibiotic prescriptions (23,733/80,786) received a first-choice antibiotic. Our analysis indicates that the ESAC-Net-QI for RTIs can be calculated at case and facility levels using routine health insurance data. For specific RTIs, there were indications of improvement potential in the frequency and selection of prescribed antibiotics.

More than floors and domes – contexts around ovens in the late neolithic of the Central Balkans

Even though ovens are the most prominent feature in the Vinča culture houses, they have often been neglected in the archaeological publications. Usually, only the information about their location or number of floors is provided, but more detailed descriptions are missing. This is also influenced by their poor preservation in the archaeological record, as parts of their upper structure are often missing. But, although fully preserved ovens represent quite a rarity, contexts surrounding them are usually filled with different categories of artefacts and architectural features. These complex contexts that include ovens, offer plethora of information about different socio-economic phenomena in the Late Neolithic of the Central Balkans. Therefore, ovens should be perceived as more than just fire installations for food processing and house heating, as they can offer insight into household organization and symbolic aspects of the Neolithic life. Additionally, ovens located in the outdoor spaces can provide information about settlement organization and social dynamics on a larger scale. In this paper, different contexts around ovens are presented and interpreted – from functional, economic, social, and symbolic perspective.

Causality-based counterfactual explanation for classification models

Counterfactual explanation is one branch of interpretable machine learning that produces a perturbation sample to change the model’s original decision. The generated samples can act as a recommendation for end-users to achieve their desired outputs. Most of the current counterfactual explanation approaches are the gradient-based method, which can only optimize the differentiable loss functions with continuous variables. Accordingly, the gradient-free methods are proposed to handle the categorical variables, which however have several major limitations: (1) causal relationships among features are typically ignored when generating the counterfactuals, possibly resulting in impractical guidelines for decision-makers; (2) the counterfactual explanation algorithm requires a great deal of effort into parameter tuning for determining the optimal weight for each loss functions which must be conducted repeatedly for different datasets and settings. In this work, to address the above limitations, we propose a prototype-based counterfactual explanation framework (ProCE). ProCE is capable of preserving the causal relationship underlying the features of the counterfactual data. In addition, we design a novel gradient-free optimization based on the multi-objective genetic algorithm that generates the counterfactual explanations for the mixed-type of continuous and categorical features. Numerical experiments demonstrate that our method compares favorably with state-of-the-art methods and therefore is applicable to existing prediction models. All the source codes and data are available at https://github.com/tridungduong16/multiobj-scm-cf.

THE LINGUISTIC CATEGORY OF EVALUATION AND ITS CHARACTERISTICS IN THE UKRAINIAN MASS MEDIA

The article is devoted to the characterisation and interpretation of the linguistic category of evaluation, its essence and main features on the basis of modern mass media texts. The ambiguity of scientists’ interpretation of the nature of evaluation as a complex multidimensional linguistic, cultural and linguistic-pragmatic phenomenon with a dual nature — linguistic and mental — requires us to consider the peculiarities of the linguistic category of evaluation as an axiological component of word semantics. We understand evaluation as the result of the speaker’s activity in relation to the subject of speech, which is verbally fixed in the sense of heterogeneous linguistic units united by evaluative semantics, expressing a positive or negative attitude towards it in accordance with the existing ideal model of the world and aiming to evoke the speaker’s predicted reaction in the addressee of the message. Evaluative values are expressed at all levels of language: lexical, syntactic and phonetic. Evaluation fulfils cognitive, communicative, pragmatic and linguistic-cultural functions; it forms the idea of values or anti-values. The set of values reflects the hierarchical scale in which the spiritual culture of a given society is represented. It is proved that the linguistic category of evaluation is a means of realising the positive or negative significance of the results of human activity for the satisfaction of their objectively determined needs and interests. It was found that evaluation, which is explicitly or implicitly linked to comparison, always contains subjective (modal evaluation framework) and objective (descriptive, attributive) components that interact with each other.

Implementasi E-Katalog IKM Berbasis Website Pada Dinnakerind Di Kabupaten Demak

In this digital era, information technology is very important to develop small and medium industries (SMEs). The Department of Labor and Industry of Demak Regency seeks to improve the competitiveness of SMEs through a website-based e-catalog. The aim of this research is to create and implement an e-ICM catalogue that can be accessed online by the public as well as to evaluate how effective the catalogue is in promoting local products. Needs analysis, system design, implementation, and testing are the measures used in this research. This research aims to design and implement an SMEs e-catalog that can be accessed online by the community, and evaluate its effectiveness in promoting local products. The research method includes needs analysis, system design, implementation, and testing. The result is an e-catalog website that displays complete information about SMEs products in Demak Regency, including product descriptions, images, and manufacturer contacts. This website also has a search feature and product categories to facilitate users. Using the black-box testing method, the test was conducted with a usability survey of 22 respondents, which yielded a score of 86.8%. It is expected that the system will help develop small and medium-sized industries (SMEs) in Demak district, improve product visibility, and support the local economy.

A Classification Method for Incomplete Mixed Data Using Imputation and Feature Selection

Data missing is a ubiquitous problem in real-world systems that adversely affects the performance of machine learning algorithms. Although many useful imputation methods are available to address this issue, they often fail to consider the information provided by both features and labels. As a result, the performance of these methods might be constrained. Furthermore, feature selection as a data quality improvement technique has been widely used and has demonstrated its efficiency. To overcome the limitation of imputation methods, we propose a novel algorithm that combines data imputation and feature selection to tackle classification problems for mixed data. Based on the mean and standard deviation of quantitative features and the selecting probabilities of unique values of categorical features, our algorithm constructs different imputation models for quantitative and categorical features. Particle swarm optimization is used to optimize the parameters of the imputation models and select feature subsets simultaneously. Additionally, we introduce a legacy learning mechanism to enhance the optimization capability of our method. To evaluate the performance of the proposed method, seven algorithms and twelve datasets are used for comparison. The results show that our algorithm outperforms other algorithms in terms of accuracy and F1 score and has reasonable time overhead.

Examining feature extraction and classification modules in machine learning for diagnosis of low-dose computed tomographic screening-detected in vivo lesions.

Medical imaging-based machine learning (ML) for computer-aided diagnosis of in vivo lesions consists of two basic components or modules of (i)feature extraction from non-invasively acquired medical images and (ii)feature classification for prediction of malignancy of lesions detected or localized in the medical images. This study investigates their individual performances for diagnosis of low-dose computed tomography (CT) screening-detected lesions of pulmonary nodules and colorectal polyps. Three feature extraction methods were investigated. One uses the mathematical descriptor of gray-level co-occurrence image texture measure to extract the Haralick image texture features (HFs). One uses the convolutional neural network (CNN) architecture to extract deep learning (DL) image abstractive features (DFs). The third one uses the interactions between lesion tissues and X-ray energy of CT to extract tissue-energy specific characteristic features (TFs). All the above three categories of extracted features were classified by the random forest (RF) classifier with comparison to the DL-CNN method, which reads the images, extracts the DFs, and classifies the DFs in an end-to-end manner. The ML diagnosis of lesions or prediction of lesion malignancy was measured by the area under the receiver operating characteristic curve (AUC). Three lesion image datasets were used. The lesions' tissue pathological reports were used as the learning labels. Experiments on the three datasets produced AUC values of 0.724 to 0.878 for the HFs, 0.652 to 0.965 for the DFs, and 0.985 to 0.996 for the TFs, compared to the DL-CNN of 0.694 to 0.964. These experimental outcomes indicate that the RF classifier performed comparably to the DL-CNN classification module and the extraction of tissue-energy specific characteristic features dramatically improved AUC value. The feature extraction module is more important than the feature classification module. Extraction of tissue-energy specific characteristic features is more important than extraction of image abstractive and characteristic features.

Stability of Radiomic Features against Variations in Lesion Segmentations Computed on Apparent Diffusion Coefficient Maps of Breast Lesions.

Diffusion-weighted imaging (DWI) combined with radiomics can aid in the differentiation of breast lesions. Segmentation characteristics, however, might influence radiomic features. To evaluate feature stability, we implemented a standardized pipeline featuring shifts and shape variations of the underlying segmentations. A total of 103 patients were retrospectively included in this IRB-approved study after multiparametric diagnostic breast 3T MRI with a spin-echo diffusion-weighted sequence with echoplanar readout (b-values: 50, 750 and 1500 s/mm2). Lesion segmentations underwent shifts and shape variations, with >100 radiomic features extracted from apparent diffusion coefficient (ADC) maps for each variation. These features were then compared and ranked based on their stability, measured by the Overall Concordance Correlation Coefficient (OCCC) and Dynamic Range (DR). Results showed variation in feature robustness to segmentation changes. The most stable features, excluding shape-related features, were FO (Mean, Median, RootMeanSquared), GLDM (DependenceNonUniformity), GLRLM (RunLengthNonUniformity), and GLSZM (SizeZoneNonUniformity), which all had OCCC and DR > 0.95 for both shifting and resizing the segmentation. Perimeter, MajorAxisLength, MaximumDiameter, PixelSurface, MeshSurface, and MinorAxisLength were the most stable features in the Shape category with OCCC and DR > 0.95 for resizing. Considering the variability in radiomic feature stability against segmentation variations is relevant when interpreting radiomic analysis of breast DWI data.