Machine Learning Classifiers Research Articles

Assessment of Surgeons' Stress Levels with Digital Sensors during Robot-Assisted Surgery: An Experimental Study.

Robot-Assisted Minimally Invasive Surgery (RAMIS) marks a paradigm shift in surgical procedures, enhancing precision and ergonomics. Concurrently it introduces complex stress dynamics and ergonomic challenges regarding the human-robot interface and interaction. This study explores the stress-related aspects of RAMIS, using the da Vinci XI Surgical System and the Sea Spikes model as a standard skill training phantom to establish a link between technological advancement and human factors in RAMIS environments. By employing different physiological and kinematic sensors for heart rate variability, hand movement tracking, and posture analysis, this research aims to develop a framework for quantifying the stress and ergonomic loads applied to surgeons. Preliminary findings reveal significant correlations between stress levels and several of the skill-related metrics measured by external sensors or the SURG-TLX questionnaire. Furthermore, early analysis of this preliminary dataset suggests the potential benefits of applying machine learning for surgeon skill classification and stress analysis. This paper presents the initial findings, identified correlations, and the lessons learned from the clinical setup, aiming to lay down the cornerstones for wider studies in the fields of clinical situation awareness and attention computing.

Abstract PO4-05-05: Immune Activation Signatures for predicting CDKi primary response in advanced breast cancer patients

Abstract Background Cyclin Dependent Kinase inhibitors (CDKi’s) in combination with Endocrine Therapy (ET) have changed the first line management of patients with the most common subtype (HR+, HER2-) of metastatic breast cancer (MBC), improving progression-free survival (PFS) and overall survival. However, in approved indications for advanced breast cancer, approximately 20% of patients have been observed to have no response to CDKi’s and ET, and to progress after starting these therapies. Unlike other targeted therapeutics, there is no companion diagnostic or other clinically reliable biomarker for clinical decision support to indicate which patients will and will not benefit from CDKi therapy. Methods We have developed a novel platform that measures the Immune Activation Signature of patient plasma samples. Via simple measurements of the total concentration of protein-incorporated amino acid residues within patient plasma samples, the platform is designed to reveal changes in the proportion of immunoglobulins and albumin, and class-switching among immunoglobulins. For that purpose, we use Bioorthogonal chemical labelling reactions to label the protein-incorporated amino acids residues within the plasma. We measured the plasma samples of N=30 HR+, HER2- MBC patients, median age 54 years old, who were baseline naïve for CDKi’s treatment (16 on palbociclib, 11 on ribociclib, and 3 on abemaciclib) and ET. Eighteen patients had metastatic visceral disease and 12 patients had bone-only MBC. All patients were prospectively followed, and response assessed according to RECIST criteria on a CT scan every 3-months. Results Non-responding patients developed objective progressive disease during the first 6 months after starting CDKis and ET. In this cohort, 4 patients (13%) were classified as non-responders. The median PFS in non-responding patients was 4.5 months (4.03-5.02) and the median PFS in the cohort of responding patients was 16.6 months (6.8 – 40.9). The Immune Activation Signatures of the Responding and Non-Responding patients are shown in table 1. We analyzed the results with classical statistics and machine learning. A multivariate analysis of variance (MANOVA) test of the null hypothesis that the Immune Activation Signatures of Non-Responders and Responders are the same gave p = 0.00388. We analyzed the measured Immune Activation Signatures using a supervised machine learning classifier and observed in a held-back validation set correct prediction of 100% of non-responding patients and 95% accurate predictions overall. Conclusions According to these results, this platform may provide a clinically helpful biomarker for clinical decision support in the advanced or aggressive breast cancer context (patients with visceral metastasis bordering on visceral crisis), or for patient stratification in new indications. Table. Immune Activation Signature measurements (mean values in Clinical Outcome) Citation Format: Luis Costa, Cong Tang, Emma Yates, Qi Shi, Wesley Sukdao, Patricia Corredeira, Gonçalo Costa, Helena Pais, Catarina Abreu, Leonor Ribeiro, Rita Teixeira de Sousa, Sofia Torres, André Mansinho, Sandra Casimiro, Ana Cavaco, Patricia Alves, Angela Rodrigues, Lisiana Szeneszi, Gonçalo Bernardes. Immune Activation Signatures for predicting CDKi primary response in advanced breast cancer patients [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO4-05-05.

Independent Video Steganalysis Framework Perspective of Secure Video Processing

We have designed an algorithm of cross-domain feature set extraction for cross-domain video steganalysis of video steganography in multiple domains. For video steganography, we implemented the recently proposed steganography methods such as PMs, MVs, and IPMs. The outcomes of these techniques have fed as input to the proposed Steganalysis approach. We have designed a cross-domain technique for Steganalysis in which the global feature set is extracted according to common statistical properties. After the extraction of global features, the domain-specific features have been extracted in the local features set. Both global and local features are set to form the cross-domain steganalysis technique. The classification has been performed by using the conventional machine learning classifiers such as Support Vector Machine (SVM) and Artificial Neural Network (ANN).

Abstract PO5-26-05: Machine learning-based risk prediction for late distant recurrence in young women with estrogen receptor-positive/human epidermal growth factor 2-negative breast cancer

Abstract Background There is a clinical unmet need to predict the risk of late recurrence in premenopausal women with estrogen receptor (ER)-positive/human epidermal growth factor 2 (HER2)-negative breast cancer (BC) who completed endocrine treatment (ET). Previous studies suggest that the CTS5 model, developed using data from cohorts with postmenopausal women and predicts a likelihood of late recurrence after five years of ET, is not highly predictive in premenopausal women. We developed a machine learning-based model predicting the chance of late distant recurrence (DR) using multi-institutional cohorts of premenopausal women with ER-positive/HER2-negative BC who have had no DR for five years from the surgery. Methods The study conducted a retrospective review of patients who underwent primary surgery between 2000 and 2011 at Samsung Medical Center, Gangnam Severance Hospital, and Seoul National University Hospital. The study included premenopausal women with ER-positive/HER2-negative BC aged 45 years or younger, who were treated over two years of ET with or without ovarian function suppression and had not experienced distant recurrence for at least five years following the surgery. Patients who had received neoadjuvant chemotherapy were excluded from the study. The primary endpoint is the area under the curve (AUC) and sensitivity (Recall) of the model predicting late DR after five years from primary operation. A total of nine clinical features, including age, tumor size, the number of positive lymph nodes, nuclear grade (NG), histologic grade (HG), progesterone receptor (PR) status, chemotherapy, extension of ET, and addition of OFS, were utilized for supervised machine learning classification. Results A total of 2,555 patients were included in this study. The median age was 41 (range, 21-45) years. During a median follow-up duration of 130.3 (range, 60.0-257.6) months, 157 women (6.1%) had late DR after five years of breast cancer surgery. Age, NG, HG, tumor size, and the number of positive LNs were important clinicopathologic variables for late recurrence. The treatment variables included chemotherapy, extension of ET, and the addition of OFS. There were 1,951 patients (76.4%) who had adjuvant chemotherapy, 650 patients (25.4%) with extension of ET over 5 years, and 529 patients (20.7%) who underwent addition of OFS. To develop the model, we performed repeated 5-fold cross-validation using a Balanced Random Forest classifier. The AUC and sensitivity of the model with six features, including age, tumor size, the number of positive lymph nodes, NG, HG, and PR status, were 0.705 and 0.768, respectively. When chemotherapy was added to this model, the AUC and sensitivity of the model increased to 0.713 and 0.771, respectively. When two more treatment features, such as ET extension and addition of OFS, were added to the model, the AUC and sensitivity were 0.714 and 0.752, respectively. Conclusions Demographic and clinical characteristics of the patients in this study Table. Citation Format: Dong Seung Shin, Janghee Lee, Jong-Ho Cheun, Jun-Hee Lee, Yewon Shin, Soong June Bae, Eunhye Kang, Sunyoung Kwon, Han-Byoel Lee, Jai Min Ryu, Sung Gwe Ahn. Machine learning-based risk prediction for late distant recurrence in young women with estrogen receptor-positive/human epidermal growth factor 2-negative breast cancer [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO5-26-05.

Abstract PO2-02-10: Predicting the feasibility of breast conserving surgery using pre-treatment standard of care DCE-MRI: a novel clinical decision support tool for breast cancer surgical planning

Abstract Background: The decision-making process regarding breast cancer treatment options, specifically breast conserving surgery (BCS) versus mastectomy, is a complex and sensitive matter that takes into account the patient's preferences and their eligibility for BCS. A range of anatomical factors, such as tumor size, breast size, and proximity of the tumor to the skin, can influence the surgeon's recommendation regarding BCS or mastectomy. While mastectomy may be the only viable option for patients with large tumors, those with smaller tumors often have the opportunity to collaboratively decide with their surgeon on the most appropriate approach to ensure oncologic success and minimize potential cosmetic deformity. To facilitate this decision-making process, we developed a tool capable of providing informed recommendations on BCS or mastectomy based on standard of care imaging patient data. By integrating these recommendation capabilities, we aim to enhance the decision-making process and improve patient outcomes in breast cancer treatment. Methods: To assist surgeons and patients in determining a patient’s eligibility for breast conserving surgery (BCS), we developed the BCS Feasibility Score. The BCS Feasibility Score is generated from a binary classifier machine learning model and is representative of the probability that a given patient could receive BCS. The score is based solely on the output from TumorSight Viz, a software platform that uses only pre-treatment T1-weighted DCE MRI and applies a suite of deep learning algorithms to segment the patient’s disease and the surrounding breast tissues. From the multi-tissue segmentation generated using TumorSight Viz, we extracted 25 spatial morphology features to use in the model development process. The training and testing sets were a subsample of a publicly available dataset (Saha, et al, 2018). We selected patients that received either BCS or mastectomy and had a viable T1-weighted DCE MRI. Bi-lateral cases were excluded. The total training/testing set consisted of n = 749 cases, 363 of which received mastectomy and 386 who received BCS. Before model training occurred, we split 20% of the sample to use as the test set. We then trained a random forest classifier across a range of pre-set hyper-parameters, using a total of 25 features as inputs, and using 5x cross-validation in the training set. We then assessed model performance in the testing set only. Results: In the test set, 76 cases received BCS versus 74 that received mastectomy. We successfully predicted BCS in 56 out of 76 cases (74%), and mastectomy in 46 out of 74 cases (62%). Overall, we observed an AUC = 0.76 and an F1 score = 0.66, indicating moderate to strong model performance. The most important features in the model, as measured by SHAP values, included the axis aligned tumor longest dimension (mm), the closest distance between the tumor and the nipple (mm), the volume of the tumor convex hull (mL; dilated uni-centrically), and the ratio of tumor convex hull volume to breast volume. We tested the model’s performance in a fully independent holdout validation set of 579 cases. Within this validation set, 335 patients received BCS versus 244 received mastectomy. We successfully predicted BCS in 254 out of 335 cases (76%), and mastectomy in 149 out of 244 cases (61%). Overall, we observed an AUC = 0.75 and an F1 score = 0.63. Validation set performance mirrored results observed in the test set, indicating strong generalizability. In the validation set, comparing patients who underwent BCS to those who underwent mastectomy, we observed significant differences across all four of the most predictive features. Conclusion: TumorSight Viz and the BCS Feasibility Score help empower both patients and clinicians with information and tools to facilitate surgical planning decisions. Citation Format: John Pfeiffer, Bradley Feiger, Anuja Antony, Joseph Peterson. Predicting the feasibility of breast conserving surgery using pre-treatment standard of care DCE-MRI: a novel clinical decision support tool for breast cancer surgical planning [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO2-02-10.

Computational kinematics of dance: distinguishing hip hop genres

Dance plays a vital role in human societies across time and culture, with different communities having invented different systems for artistic expression through movement (genres). Differences between genres can be described by experts in words and movements, but these descriptions can only be appreciated by people with certain background abilities. Existing dance notation schemes could be applied to describe genre-differences, however they fall substantially short of being able to capture the important details of movement across a wide spectrum of genres. Our knowledge and practice around dance would benefit from a general, quantitative and human-understandable method of characterizing meaningful differences between aspects of any dance style; a computational kinematics of dance. Here we introduce and apply a novel system for encoding bodily movement as 17 macroscopic, interpretable features, such as expandedness of the body or the frequency of sharp movements. We use this encoding to analyze Hip Hop Dance genres, in part by building a low-cost machine-learning classifier that distinguishes genre with high accuracy. Our study relies on an open dataset (AIST++) of pose-sequences from dancers instructed to perform one of ten Hip Hop genres, such as Breakdance, Popping, or Krump. For comparison we evaluate moderately experienced human observers at discerning these sequence’s genres from movements alone (38% where chance = 10%). The performance of a baseline, Ridge classifier model was fair (48%) and that of the model resulting from our automated machine learning pipeline was strong (76%). This indicates that the selected features represent important dimensions of movement for the expression of the attitudes, stories, and aesthetic values manifested in these dance forms. Our study offers a new window into significant relations of similarity and difference between the genres studied. Given the rich, complex, and culturally shaped nature of these genres, the interpretability of our features, and the lightweight techniques used, our approach has significant potential for generalization to other movement domains and movement-related applications.

Fast, Simple, and Accurate Time Series Analysis with Large Language Models: An Example of Mean-motion Resonances Identification

Classical machine learning has been actively utilized in astronomy to address various challenges, including predicting orbital stability, classifying asteroids, galaxies, and other objects, and analyzing images. However, the emerging trend in artificial intelligence involves the use of large language models such as GPT-4 and ChatGPT. These models are trained on a large corpus of text and can perform a wide range of natural language processing tasks, including text generation, translation, summarization, and classification. Surprisingly, these capabilities present significant potential for application in astronomy. This paper demonstrates how the new model gpt-4-vision-preview can analyze visual patterns and accurately classify asteroids as resonant or nonresonant with high accuracy. This process requires no training, fine-tuning, or coding beyond writing the appropriate prompt in natural language. Moreover, this approach can be extended to other common problems within astronomy.

The Contribution of Explainable Machine Learning Algorithms Using ROI-based Brain Surface Morphology Parameters in Distinguishing Early-onset Schizophrenia From Bipolar Disorder

Background/aimTo differentiate early-onset schizophrenia (EOS) from early-onset bipolar disorder (EBD) using surface-based morphometry measurements and brain volumes using machine learning (ML) algorithms. MethodHigh-resolution T1-weighted images were obtained to measure cortical thickness (CT), gyrification, gyrification index (GI), sulcal depth (SD), fractal dimension (FD), and brain volumes. After the feature selection step, ML classifiers were applied for each feature set and the combination of them. The SHapley Additive exPlanations (SHAP) technique was implemented to interpret the contribution of each feature. FindingsOne-hundred-forty-four adolescents (16.2±1.4 years, female=40%) with EOS (n=81) and EBD (n=63) were included. The Adaptive Boosting (AdaBoost) algorithm had the highest accuracy (82.75%) in the whole dataset that includes all variables from Destrieux atlas. The best-performing algorithms were K-Nearest Neighbors (KNN) for FD subset, Support Vector Machine (SVM) for SD subset, and AdaBoost for GI subset. The KNN algorithm had the highest accuracy (accuracy=79.31%) in the whole dataset from the Desikan-Killiany-Tourville atlas. ConclusionThis study demonstrates the use of ML in the differential diagnosis of EOS and EBD using surface-based morphometry measurements. Future studies could focus on multicenter data for the validation of these results.

Assessing Multi-source Random Forest Classification and Robustness of Predictor Variables in Flooded Areas Mapping

Flood extent delineation techniques have benefited from the increasing availability of remote sensing imagery, classification techniques and the introduction of geomorphic descriptors derived from Digital Elevation Models (DEM). On the other hand, high-performing Machine Learning (ML) methods have allowed for the development of accurate flood maps by integrating several predictor variables into supervised or unsupervised algorithms. Among others, Random Forest (RF) is a powerful and widely applied ML classifier, providing accurate predictions also with complex datasets and for varying parameters set. In the present study, the effectiveness of this algorithm for mapping flooded areas was evaluated. Various geospatial data sources were integrated, including morphological indicators, such as the Geomorphic Flood Index (GFI), Sentinel-2 bands, multispectral indices, and Sentinel-1 polarizations. The reliability of the predictor variables under different training sample sizes was evaluated and the accuracy of the RF classifier was assessed. Moreover, by exploring the algorithm ability to identify the most important variables, the predictors contributing the most to the classification were identified and their stability for varying training parameters was investigated. To gauge the adaptability and consistency of these features, we applied our analyses to different study areas around the World. The results indicate that certain predictors displayed remarkable stability across different sample sizes and remained robust under various training parameters. However, some variability in the algorithm structure and the features related to the specific complexities of each considered study case was also observed.

Leveraging network topology for credit risk assessment in P2P lending: A comparative study under the lens of machine learning

Peer-to-Peer (P2P) lending markets have witnessed remarkable growth, revolutionizing the way borrowers and lenders interact. Despite their increasing popularity, P2P lending poses significant challenges related to credit risk assessment and default prediction with meaningful implications for financial stability. Traditional credit risk models have been widely employed in the field of P2P lending; however, they may not be capable to capture latent factor information inherent to a loan network based on similarity distances. Thus, in this study we propose an enhanced two- step modeling approach for machine learning (ML) that utilises insights from network analysis and subsequently combines derived network centrality metrics with traditional credit risk factors to improve the prediction accuracy in the credit default prediction process. Through a comparative analysis of three classical ML models with varying degrees of complexity, namely Elastic Net (EN), Random Forest (RF), and Multi-Layer Perceptron (MLP), we showcase novel evidence that the systematic inclusion of network topological features in the credit scoring process can significantly improve the prediction accuracy of the scoring models. Additional robustness tests via the inclusion of randomly shuffled centrality metrics in the analysis, and a further comparison of the graph-based models against a pertinent state-of-the-art credit scoring model in form of XGBoost further confirm our results. The insights from this study bear valuable conclusions for P2P lending platforms to further improve their scoring systems with graph-enhanced metrics, thereby reducing default risk and facilitating greater access to credit.

CAPTURE: Comprehensive anti-cancer peptide predictor with a unique amino acid sequence encoder

Anticancer peptides (ACPs) key properties including bioactivity, high efficacy, low toxicity, and lack of drug resistance make them ideal candidates for cancer therapies. To deeply explore the potential of ACPs and accelerate development of cancer therapies, although 53 Artificial Intelligence supported computational predictors have been developed for ACPs and non ACPs classification but only one predictor has been developed for ACPs functional types annotations. Moreover, these predictors extract amino acids distribution patterns to transform peptides sequences into statistical vectors that are further fed to classifiers for discriminating peptides sequences and annotating peptides functional classes. Overall, these predictors remain fail in extracting diverse types of amino acids distribution patterns from peptide sequences. The paper in hand presents a unique CARE encoder that transforms peptides sequences into statistical vectors by extracting 4 different types of distribution patterns including correlation, distribution, composition, and transition. Across public benchmark dataset, proposed encoder potential is explored under two different evaluation settings namely; intrinsic and extrinsic. Extrinsic evaluation indicates that 12 different machine learning classifiers achieve superior performance with the proposed encoder as compared to 55 existing encoders. Furthermore, an intrinsic evaluation reveals that, unlike existing encoders, the proposed encoder generates more discriminative clusters for ACPs and non-ACPs classes. Across 8 public benchmark ACPs and non-ACPs classification datasets, proposed encoder and Adaboost classifier based CAPTURE predictor outperforms existing predictors with an average accuracy, recall and MCC score of 1%, 4%, and 2% respectively. In generalizeability evaluation case study, across 7 benchmark anti-microbial peptides classification datasets, CAPTURE surpasses existing predictors by an average AU-ROC of 2%. CAPTURE predictive pipeline along with label powerset method outperforms state-of-the-art ACPs functional types predictor by 5%, 5%, 5%, 6%, and 3% in terms of average accuracy, subset accuracy, precision, recall, and F1 respectively. CAPTURE web application is available at https://sds_genetic_analysis.opendfki.de/CAPTURE.

Decoding Consumer Behavior: Predictive Analytics of Social Media Advertising Effectiveness

In the digital age, social media advertising has become an indispensable tool for marketing. This study investigates the effectiveness of social media advertisements in influencing consumer purchasing decisions. Using a dataset obtained from Kaggle, this analysis incorporates demographic and socio-economic factors—specifically, age, gender, and estimated salary—to predict consumer behavior towards product purchases after exposure to advertisements. Our dataset consists of attributes such as User ID, gender, age, salary, and purchase history. We employed several machine learning classification models including logistic regression, decision trees, random forests, SVM, Naive Bayes, KNN, and Perceptron to identify the most predictive factors and to model purchasing likelihood. Outliers were removed and missing data imputed through statistical methods to ensure data integrity. The highest predictive accuracy was achieved using a Random Forest model with 20 trees, which demonstrated a 95% accuracy rate. This model was integrated into a user-friendly interface developed with Gradio, facilitating real-time predictions. Our findings can assist companies in tailoring their marketing strategies to target demographics most likely to respond positively to advertisements, thereby enhancing the efficiency and effectiveness of their marketing campaigns on social platforms. Keywords:- Social Media Advertising, Consumer Behavior, Predictive Analytics, Machine Learning, Classification Algorithms, Random Forest, Marketing Strategy, Purchase Intentions, Data Preprocessing, Gradio Interface.

Global and local interpretability techniques of supervised machine learning black box models for numerical medical data

The most effective machine learning classification techniques, such as artificial neural networks, are not easily interpretable, which limits their usefulness in critical areas, such as medicine, where errors can have severe consequences. Researchers have been working to balance the trade-off between the model performance and interpretability. In this study, seven interpretability techniques (global surrogate, accumulated local effects, local interpretable model-agnostic explanations (LIME), Shapley additive explanations (SHAP), model agnostic post hoc local explanations (MAPLE), local rule-based explanation (LORE), and Contextual Importance and Utility (CIU)) were evaluated to interpret five medical classifiers (multilayer perceptron, support vector machines, random forests, extreme gradient boosting, and naïve bayes) using six model performance metrics and three interpretability technique metrics across six medical numerical datasets. The results confirmed the effectiveness of integrating global and local interpretability techniques, and highlighted the superior performance of global SHAP explainer and local CIU explanations. The quantitative evaluations of explanations emphasised the importance of assessing these interpretability techniques before employing them to interpret black box models.

Integrated neural network and evolutionary algorithm approach for liver fibrosis staging: Can artificial intelligence reduce patient costs?

Staging liver fibrosis is important, and liver biopsy is the gold standard diagnostic tool. We aim to design and evaluate an artificial neural network (ANN) method by taking advantage of the Teaching Learning-Based Optimization (TLBO) algorithm for the prediction of liver fibrosis stage in blood donors and hepatitis C patients. We propose a method based on a selection of machine learning classification methods including multilayer perceptron (MLP) neural network, Naive Bayesian (NB), decision tree, and deep learning. Initially, the synthetic minority oversampling technique (SMOTE) is performed to address the imbalance in the dataset. Afterward, the integration of MLP and TLBO is implemented. We propose a novel algorithm that reduces the number of required patient features to seven inputs. The accuracy of MLP using 12 features is 0.903, while that of the proposed MLP with TLBO is 0.891. Besides, the diagnostic accuracy of all methods, except the model designed with the Bayesian network, increases when the SMOTE balancer is applied. The decision tree-based deep learning methods show the highest levels of accuracy with 12 features. Interestingly, with the use of TLBO and seven features, MLP reached an accuracy rate of 0.891, which is quite satisfactory when compared with those of similar studies. The proposed model provides high diagnostic accuracy, while reducing the required number of properties from the samples. The results of our study show that the recruited algorithm of our study is more straightforward, with a smaller number of required properties and similar accuracy.

Artificial Intelligence in Ship Trajectory Prediction

Maritime traffic is increasing more and more, creating more complex navigation environments for ships. Ship trajectory prediction based on historical AIS data is a vital method of reducing navigation risks and enhancing the efficiency of maritime traffic control. At present, employing machine learning or deep learning techniques to construct predictive models based on AIS data has become a focal point in ship trajectory prediction research. This paper systematically evaluates various trajectory prediction methods, spanning classical machine learning approaches and emerging deep learning techniques, to uncover their respective merits and drawbacks. In this work, a variety of studies were investigated that applied different algorithms in ship trajectory prediction, including regression models (RMs), artificial neural networks (ANNs), Kalman filtering (KF), and random forests (RFs) in machine learning, along with deep learning such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), long short-term memory (LSTM), gate recurrent unit (GRU) networks, and sequence-to-sequence (Seq2seq) networks. The performance of predictive models based on different algorithms in trajectory prediction tasks was graded and analyzed. Among the existing studies, deep learning methods exhibit significant performance and considerable potential application value for maritime traffic systems, which can be assessed by future work on ship trajectory prediction research.

Deep Learning Radiomics Model of Contrast-Enhanced CT for Differentiating the Primary Source of Liver Metastases

Rationale and ObjectivesTo develop and validate a deep learning radiomics (DLR) model based on contrast-enhanced computed tomography (CT) to identify the primary source of liver metastases. Materials and MethodsIn total, 657 liver metastatic lesions, including breast cancer (BC), lung cancer (LC), colorectal cancer (CRC), gastric cancer (GC), and pancreatic cancer (PC), from 428 patients were collected at three clinical centers from January 2018 to October 2023 series. The lesions were randomly assigned to the training and validation sets in a 7:3 ratio. An additional 112 lesions from 61 patients at another clinical center served as an external test set. A DLR model based on contrast-enhanced CT of the liver was developed to distinguish the five pathological types of liver metastases. Stepwise classification was performed to improve the classification efficiency of the model. Lesions were first classified as digestive tract cancer (DTC) and non-digestive tract cancer (non-DTC). DTCs were divided into CRC, GC, and PC and non-DTCs were divided into LC and BC. To verify the feasibility of the DLR model, we trained classical machine learning (ML) models as comparison models. Model performance was evaluated using accuracy (ACC) and area under the receiver operating characteristic curve (AUC). ResultsThe classification model constructed by the DLR algorithm showed excellent performance in the classification task compared with ML models. Among the five categories task, highest ACC and average AUC were achieved at 0.563 and 0.796 in the validation set, respectively. In the DTC and non-DTC and the LC and BC classification tasks, AUC was achieved at 0.907 and 0.809 and ACC was achieved at 0.843 and 0.772, respectively. In the CRC, GC, and PC classification task, ACC and average AUC were the highest, at 0.714 and 0.811, respectively. ConclusionThe DLR model is an effective method for identifying the primary source of liver metastases.

Sentiment analysis of coronavirus data with ensemble and machine learning methods

The coronavirus pandemic has distanced people from social life and increased the use of social media. People's emotions can be determined with text data collected from social media applications. This is used in many fields, especially in commerce. This study aims to predict people's sentiments about the pandemic by applying sentiment analysis to Twitter tweets about the pandemic using single machine learning classifiers (Decision Tree-DT, K-Nearest Neighbor-KNN, Logistic Regression-LR, Naïve Bayes-NB, Random Forest-RF) and ensemble learning methods (Majority Voting (MV), Probabilistic Voting (PV), and Stacking (STCK)). After vectorizing the tweets using two predictive methods, Word2Vec (W2V) and Doc2Vec, and two traditional word representation methods, Term Frequency-Inverse Document Frequency (TF-IDF) and Bag of Words (BOW), classification models built using single machine learning classifiers were compared to models built using ensemble learning methods (MV, PV and STCK) by heterogeneously combining single machine classifier algorithms. Accuracy (ACC), F-measure (F), precision (P), and recall (R) were used as performance measures, with training/test separation rates of 70%-30% and 80%-20%, respectively. Among these models, the ACC of ensemble learning models ranged from 89% to 73%, while the ACC of single classifier models ranged from 60% to 80%. Among the ensemble learning methods, STCK with Doc2Vec text representation/embedding method gave the best ACC result of 89%. According to the experimental results, ensemble models built with heterogeneous machine learning classifier algorithms gave better results than single machine learning classifier algorithms.

A novel hybrid deep CNN model for breast cancer classification using Lipschitz-based image augmentation and recursive feature elimination

Breast cancer is the leading cause of death for women worldwide. Despite its difficulties, the histopathological diagnosis must be used because modern imaging techniques cannot differentiate between benign and malignant tumors. This article introduces a hybrid deep learning method with recursive feature elimination for categorizing histopathology images of breast lesions as benign or malignant, using standard datasets, to increase diagnostic accuracy. The proposed four-stage model includes several novel techniques. First, a modified Lipschitz-based image augmentation technique is utilized to ensure a proportionate representation of various sample classes. K-fold cross-validation is employed to mitigate the bias that emerges from the selection of training and testing datasets. Then, for efficient feature extraction, a hybrid convolutional neural network (CNN) approach is used by integrating three CNNs (PResNet-34, FE-VGG-16, and M-AlexNet). With the proposed universum twin support vector machine-based recursive feature elimination technique, the initial 1,000 features extracted by each deep model are further reduced to 100. Finally, the deep features are subsequently combined into a set of 300 features, which are then fed into machine learning classifiers such as decision tree, k-nearest neighborhood, linear discriminant analysis, linear regression, and support vector machine. The results demonstrate that the proposed hybrid CNN model outperforms state-of-the-art techniques in accurately classifying breast lesions as benign or malignant, with an accuracy of 99.99 %, sensitivity of 99.56 %, specificity of 99.53 %, precision of 99.55 %, and F1 score of 99.77 %.

Automatic assessment of DWI-ASPECTS for acute ischemic stroke based on deep learning.

Alberta Stroke Program Early Computed Tomography Score (ASPECTS) is a standardized semi-quantitative method for early ischemic changes in acute ischemic stroke. However, ASPECTS is still affected by expert experience and inconsistent results between readers in clinical. This study aims to propose an automatic ASPECTS scoring model based on diffusion-weighted imaging (DWI) mode to help clinicians make accurate treatment plans. Eighty-two patients with stroke were included in the study. First, we designed a new deep learning network for segmenting ASPECTS scoring brain regions. The network is improved based on U-net, which integrates multiple modules. Second, we proposed using hybrid classifiers to classify brain regions. For brain regions with larger areas, we used brain grayscale comparison algorithm to train machine learning classifiers, while using hybrid feature training for brain regions with smaller areas. The average DICE coefficient of the segmented hindbrain area can reach 0.864. With the proposed hybrid classifier, our method performs significantly on both region-level ASPECTS and dichotomous ASPECTS. The sensitivity and accuracy on the test set are 95.51% and 93.43%, respectively. For dichotomous ASPECTS, the intraclass correlation coefficient (ICC) between our automated ASPECTS score and the expert reading was 0.87. This study proposed an automated model for ASPECTS scoring of patients with acute ischemic stroke based on DWI images. Experimental results show that the method of segmentation first and then classification is feasible. Our method has the potential to assist physicians in the Alberta Stroke Program with early CT scoring and clinical stroke diagnosis.

Performance Analysis of Child Emotion Detection using Haar Cascade and CNN

A method for identifying human emotions from facial expressions is called facial emotion detection. This essay focuses on analyzing youngsters with autism's facial expressions to determine their feelings. In this research, five emotions are examined. These feelings include anger, surprise, sadness, happiness, and neutrality. Image processing and machine learning techniques are used to identify the emotions of autistic youngsters. The local binary pattern features are taken from the faces of youngsters with autism. Emotions are categorized using machine learning algorithms. Neural networks and support vector machines are two types of machine learning classifiers used in the classification process. Child age detection in film shots plays a vital role in ensuring compliance with age-restricted content regulations and safeguarding the well-being of underage actors. This abstract presents an overview of recent advancements, methodologies, and applications in using machine learning (ML) for child age detection.