Top-performing Algorithm Research Articles

Prediction of Suicidal Thoughts and Suicide Attempts in People Who Gamble Based on Biological-Psychological-Social Variables: A Machine Learning Study.

Recent research has shown that people who gamble are more likely to have suicidal thoughts and attempts compared to the general population. Despite the advancements made, no study to date has predicted suicide risk factors in people who gamble using machine learning algorithms. Therefore, current study aimed to identify the most critical predictors of suicidal ideation and suicidal attempts among people who gamble using a machine learning approach. An online survey conducted a cross-sectional analysis of 741 people who gamble (mean age: 25.9 ± 5.56). To predict the risk of suicide attempts and ideation, we employed a comprehensive set of 40 biological, psychological, social, and socio-demographic variables. The predictive models were developed using Logistic Regression, Random Forest (RF), robust eXtreme Gradient Boosting (XGBoost), and ensemble machine learning algorithms. Data analysis was performed using R-Studio software. Random Forest emerged as the top-performing algorithm for predicting suicidal ideation, with an impressive AUC of 0.934, sensitivity of 0.7514, specificity of 0.9885, PPV of 0.9473, and NPV of 0.9347. Across all models, dissociation, depression, and anxiety symptoms consistently emerged as crucial predictors of suicidal ideation. However, for suicide attempt prediction, all models exhibited weaker performance. XGBoost showed the best performance in this regard, with an AUC of 0.663, sensitivity of 0.78, specificity of 0.8990, PPV of 0.34, NPV of 0.984, and accuracy of 0.8918. Depressive symptoms and rumination severity were highlighted as the most important predictors of suicide attempts according to this model. These findings have important implications for clinical practice and public health interventions. Machine learning could help detect individuals prone to suicidal ideation and suicide attempts among people who gamble, assisting in creating tailored prevention programs to address future suicide risks more effectively.

Benchmarking algorithms for single-cell multi-omics prediction and integration.

The development of single-cell multi-omics technology has greatly enhanced our understanding of biology, and in parallel, numerous algorithms have been proposed to predict the protein abundance and/or chromatin accessibility of cells from single-cell transcriptomic information and to integrate various types of single-cell multi-omics data. However, few studies have systematically compared and evaluated the performance of these algorithms. Here, we present a benchmark study of 14 protein abundance/chromatin accessibility prediction algorithms and 18 single-cell multi-omics integration algorithms using 47 single-cell multi-omics datasets. Our benchmark study showed overall totalVI and scArches outperformed the other algorithms for predicting protein abundance, and LS_Lab was the top-performing algorithm for the prediction of chromatin accessibility in most cases. Seurat, MOJITOO and scAI emerge as leading algorithms for vertical integration, whereas totalVI and UINMF excel beyond their counterparts in both horizontal and mosaic integration scenarios. Additionally, we provide a pipeline to assist researchers in selecting the optimal multi-omics prediction and integration algorithm.

Prediction of childbearing tendency in women on the verge of marriage using machine learning techniques

The declining fertility rate and increasing marriage age among girls pose challenges for policymakers, leading to issues such as population decline, higher social and economic costs, and reduced labor productivity. Using machine learning (ML) techniques to predict the desire to have children can offer a promising solution to address these challenges. Therefore, this study aimed to predict the childbearing tendency in women on the verge of marriage using ML techniques. Data from 252 participants (203 expressing a "desire to have children" and 49 indicating "reluctance to have children") in Abadan, and Khorramshahr cities (Khuzestan Province, Iran) was analyzed. Seven ML algorithms, including multilayer perceptron (MLP), support vector machine (SVM), logistic regression (LR), random forest (RF), J48 decision tree, Naive Bayes (NB), and K-nearest neighbors (KNN), were employed. The performance of these algorithms was assessed using metrics derived from the confusion matrix. The RF algorithm showed superior performance, with the highest sensitivity (99.5%), specificity (95.6%), and receiver operating characteristic curve (90.1%) values. Meanwhile, MLP emerged as the top-performing algorithm, showcasing the best overall performance in accuracy (77.75%) and precision (81.8%) compared to other algorithms. Factors such as age of marriage, place of residence, and strength of the family center with the birth of a child were the most effective predictors of a woman's desire to have children. Conversely, the number of daughters, the wife's ethnicity, and the spouse's ownership of assets such as cars and houses were among the least important factors in predicting this desire. ML algorithms exhibit excellent predictive capabilities for childbearing tendencies in women on the verge of marriage, highlighting their remarkable effectiveness. This capacity to offer accurate prognoses holds significant promise for advancing research in this field.

Comparative Analysis of Supervised Machine Learning Algorithms for Diabetes Prediction

Diabetes Mellitus is a chronic disease with far-reaching consequences, necessitating innovative diagnostic approaches. Current methods have limitations, highlighting the need for advanced techniques. This research leveraged machine learning algorithms to predict diabetes using the Pima Indian Diabetes Dataset. Six algorithms were developed and compared: Logistic Regression, Support Vector Machine, K-Nearest Neighbor, Naive Bayes, Random Forest, and Decision Tree. Standard evaluation metrics assessed their performance. The Random Forest Classifier emerged as the top-performing algorithm, achieving an impressive 91% accuracy and demonstrating exceptional reliability. This study showcases machine learning's potential to transform diabetes diagnosis and management. The developed web application provides a user-friendly interface for predicting diabetes, facilitating timely intervention and improved health outcomes. The findings contribute significantly to the development of intelligent disease prediction systems, promoting early detection and enhanced patient care. By harnessing machine learning, this research pioneers a new frontier in diabetes diagnosis, paving the way for improved patient outcomes and enhanced healthcare services.

FSDC: Flow Samples and Dimensions Compression for Efficient Detection of DNS-over-HTTPS Tunnels

This paper proposes an innovative approach capitalized on the distinctive characteristics of command and control (C&C) beacons, namely, time intervals and frequency between consecutive unique connections, to compress the network flow dataset. While previous studies on the same matter used single technique, we propose a multi-technique approach for efficient detection of DoH tunnels. We use a baseline public dataset, CIRA-CIC-DoHBrw-2020, containing over a million network flow properties and statistical features of DoH, tunnels, benign DoH and normal browsing (HTTPS) traffic. Each sample is represented by 33 features with a timestamp. Our methodology combines star graph and bar plot visualizations with supervised and unsupervised learning techniques. The approach underscores the importance of C&C beacon characteristic features in compressing a dataset and reducing a flow dimension while enabling efficient detection of DoH tunnels. Through compression, the original dataset size and dimensions are reduced by approximately 95% and 94% respectively. For supervised learning, RF emerges as the top-performing algorithm, attaining precision and recall scores of 100% each, with speed increase of ≈6796 times faster in training and ≈55 in testing. For anomaly detection models, OCSVM emerges as the most suitable choice for this purpose, with precision (88.89) and recall (100). Star graph and bar graph models also show a clear difference between normal traffic and DoH tunnels. The reduction in flow sample size and dimension, while maintaining accuracy, holds promise for edge networks with constrained resources and aids security analysts in interpreting complex ML models to identify Indicators of Compromise (IoC).

Predictive Modelling in Legal Decision-Making: Leveraging Machine Learning for Forecasting Legal Outcomes

Predictive modelling holds significant promise in enhancing legal decision-making processes, particularly within the realm of the Supreme Court of the United States (SCOTUS). This paper investigates the application of Machine Learning (ML) algorithms to forecast legal outcomes, utilizing a dataset comprising SCOTUS cases. Through rigorous preprocessing and analysis, various ML techniques including Decision Trees, Random Forest, Support Vector Machines (SVM), Naive Bayes, k-Nearest Neighbors (k-NN), and XGBoost are applied. The performance of these models is evaluated using precision, recall, F1-score, and accuracy metrics, revealing nuanced differences in their effectiveness. Notably, XGBoost emerges as the top-performing algorithm with an accuracy of 72%, showcasing its robustness in capturing intricate legal patterns. In contrast, Naive Bayes and Decision Tree algorithms exhibit lower accuracies of 61% and 52%, respectively, highlighting potential limitations in their applicability to legal datasets. The comparative analysis sheds light on the strengths and weaknesses of each algorithm, underscoring the importance of selecting appropriate techniques tailored to the complexities of legal decision-making. This study contributes to the growing body of literature on predictive modelling in legal studies, offering valuable insights into the potential applications and implications of ML in enhancing the efficiency and efficacy of legal processes.

Virtual Analysis for Spinal Cord Injury Rehabilitation

Background Spinal cord injuries (SCI) are debilitating conditions affecting individuals worldwide annually, leading to physical, emotional, and cognitive challenges. Effective rehabilitation for SCI patients is crucial for restoring motor function and enhancing their overall quality of life. Advances in technology, including machine learning (ML) and computer vision, offer promising avenues for personalized SCI treatment. Aims This paper aimed to propose an automated and cost-effective system for spinal cord injury (SCI) rehabilitation using machine learning techniques, leveraging data from the Toronto Rehab Pose dataset and Mediapipe for real-time tracking. Objective The objective is to develop a system that predicts rehabilitation outcomes for upper body movements, highlighting the transformative role of ML in personalized SCI treatment and offering tailored strategies for improved outcomes. Methods The proposed system utilized data from the Toronto Rehab Pose dataset and Mediapipe for real-time tracking. Machine learning models, including Support Vector Machines (SVM), Logistic Regression, Naive Bayes, and XGBoost, were employed for outcome prediction. Features such as joint positions, angles, velocities, and accelerations were extracted from movement data to train the models. Results Statistical analysis revealed the ability of the system to accurately classify rehabilitation outcomes, with an average accuracy of 98.5%. XGBoost emerged as the top-performing algorithm, demonstrating superior accuracy and precision scores across all exercises. Conclusion This paper emphasizes the importance of continuous monitoring and adjustment of rehabilitation plans based on real-time progress data, highlighting the dynamic nature of SCI rehabilitation and the need for adaptive treatment strategies. By predicting rehabilitation outcomes with high accuracy, the system enables clinicians to devise targeted interventions, optimizing the efficacy of the rehabilitation process.

Analysis of Gradient Boosting, XGBoost, and CatBoost on Mobile Phone Classification

In the ever-evolving landscape of mobile phone technology, accurately classifying device specifications is paramount for market analysis and consumer decision-making. This research conducts a comprehensive analysis of mobile phone specification classification using three prominent machine learning algorithms: Gradient Boosting, XGBoost, and CatBoost. Through meticulous dataset acquisition and preprocessing steps, including resolution normalization and price categorization, features essential for classification analysis were standardized. Robust cross-validation techniques were employed to assess model performance effectively. The study demonstrates the significant impact of normalization techniques on improving model performance across all algorithms and fold variations. CatBoost consistently emerges as the top-performing algorithm, followed closely by XGBoost, with Gradient Boosting displaying respectable performance. Notably, CatBoost consistently achieves the highest AUC values and accuracy scores, demonstrating superior performance in accurately classifying mobile phone specifications. These findings underscore the importance of preprocessing methods and algorithm selection in achieving optimal classification results. For mobile phone manufacturers, leveraging machine learning algorithms for effective classification can inform product development strategies, optimizing offerings based on consumer preferences. Similarly, for data analysts, employing appropriate preprocessing techniques and algorithmic approaches can lead to more accurate predictions and informed decision-making. Future research avenues include exploring advanced preprocessing methods, investigating alternative algorithms, and incorporating additional features or datasets to enrich the classification process. Overall, this research contributes to understanding mobile phone specification classification through machine learning methodologies, offering actionable insights for industry practitioners and researchers to address evolving market dynamics and consumer preferences.

Predicting Fan Attendance at Mega Sports Events—A Machine Learning Approach: A Case Study of the FIFA World Cup Qatar 2022

Mega sports events generate significant media coverage and have a considerable economic impact on the host cities. Organizing such events is a complex task that requires extensive planning. The success of these events hinges on the attendees’ satisfaction. Therefore, accurately predicting the number of fans from each country is essential for the organizers to optimize planning and ensure a positive experience. This study aims to introduce a new application for machine learning in order to accurately predict the number of attendees. The model is developed using attendance data from the FIFA World Cup (FWC) Russia 2018 to forecast the FWC Qatar 2022 attendance. Stochastic gradient descent (SGD) was found to be the top-performing algorithm, achieving an R2 metric of 0.633 in an Auto-Sklearn experiment that considered a total of 2523 models. After a thorough analysis of the result, it was found that team qualification has the highest impact on attendance. Other factors such as distance, number of expatriates in the host country, and socio-geopolitical factors have a considerable influence on visitor counts. Although the model produces good results, with ML it is always recommended to have more data inputs. Therefore, using previous tournament data has the potential to increase the accuracy of the results.

Using Machine Learning (XGBoost) to Predict Outcomes After Infrainguinal Bypass for Peripheral Artery Disease.

To develop machine learning (ML) algorithms that predict outcomes after infrainguinal bypass. Infrainguinal bypass for peripheral artery disease carries significant surgical risks; however, outcome prediction tools remain limited. The Vascular Quality Initiative database was used to identify patients who underwent infrainguinal bypass for peripheral artery disease between 2003 and 2023. We identified 97 potential predictor variables from the index hospitalization [68 preoperative (demographic/clinical), 13 intraoperative (procedural), and 16 postoperative (in-hospital course/complications)]. The primary outcome was 1-year major adverse limb event (composite of surgical revision, thrombectomy/thrombolysis, or major amputation) or death. Our data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, we trained 6 ML models using preoperative features. The primary model evaluation metric was the area under the receiver operating characteristic curve (AUROC). The top-performing algorithm was further trained using intraoperative and postoperative features. Model robustness was evaluated using calibration plots and Brier scores. Overall, 59,784 patients underwent infrainguinal bypass, and 15,942 (26.7%) developed 1-year major adverse limb event/death. The best preoperative prediction model was XGBoost, achieving an AUROC (95% CI) of 0.94 (0.93-0.95). In comparison, logistic regression had an AUROC (95% CI) of 0.61 (0.59-0.63). Our XGBoost model maintained excellent performance at the intraoperative and postoperative stages, with AUROCs (95% CI's) of 0.94 (0.93-0.95) and 0.96 (0.95-0.97), respectively. Calibration plots showed good agreement between predicted and observed event probabilities with Brier scores of 0.08 (preoperative), 0.07 (intraoperative), and 0.05 (postoperative). ML models can accurately predict outcomes after infrainguinal bypass, outperforming logistic regression.

An End-to-end Energy-efficient Approach for Intake Detection With Low Inference Time Using Wrist-worn Sensor.

Automated detection of intake gestures with wearable sensors has been a critical area of research for advancing our understanding and ability to intervene in people's eating behavior. Numerous algorithms have been developed and evaluated in terms of accuracy. However, ensuring the system is not only accurate in making predictions but also efficient in doing so is critical for real-world deployment. Despite the growing research on accurate detection of intake gestures using wearables, many of these algorithms are often energy inefficient, impeding on-device deployment for continuous and real-time monitoring of diet. This paper presents a template-based optimized multicenter classifier that enables accurate intake gesture detection while maintaining low-inference time and energy consumption using a wrist-worn accelerometer and gyroscope. We designed an Intake Gesture Counter smartphone application (CountING) and validated the practicality of our algorithm against seven state-of-the-art approaches on three public datasets (In-lab FIC, Clemson, and OREBA). Compared with other methods, we achieved optimal accuracy (81.60% F1 score) and very low inference time (15.97 msec per 2.20-sec data sample) on the Clemson dataset, and among the top performing algorithms, we achieve comparable accuracy (83.0% F1 score compared with 85.6% in the top performing algorithm) but superior inference time (13.8x faster, 33.14 msec per 2.20-sec data sample) on the In-lab FIC dataset and comparable accuracy (83.40% F1 score compared with 88.10% in the top-performing algorithm) but superior inference time (33.9x faster, 16.71 msec inference time per 2.20-sec data sample) on the OREBA dataset. On average, our approach achieved a 25-hour battery lifetime (44% to 52% improvement over state-of-the-art approaches) when tested on a commercial smartwatch for continuous real-time detection. Our approach demonstrates an effective and efficient method, enabling real-time intake gesture detection using wrist-worn devices in longitudinal studies.

Genetic Folding (GF) Algorithm with Minimal Kernel Operators to Predict Stroke Patients

ABSTRACT A stroke is a medical disorder in which blood arteries in the brain rupture, causing brain damage. Symptoms may appear when the brain’s blood supply and other nutrients are cut off. According to the World Health Organization, Stroke is the leading cause of death and disability globally. Early recognition of the multiple warning signs of a stroke helps reduce the severity of the stroke. The paper presents a modified version of the Genetic Folding algorithm to predict stroke based on symptoms. Considerable Machine Learning models, including Logistic Regression, Decision Tree, Random Forest, Naïve Bayes, Support Vector Machine, and the proposed Minimal Genetic Folding, were compared to forecast the probability of having a stroke in the brain using a variety of physiological characteristics. The proposed minimal Genetic Folding approach has been developed using the open-access Stroke Prediction dataset using minimal kernel operators. The datasets generated and/or analyzed during the current study are available in the Kaggle repository. With an accuracy of 83.2%, the proposed minimal Genetic Folding approach outperformed Logistic Regression by 4.2%, Naïve Bayes by 1.2%, Decision Tree by 17.2%, and Support Vector Machine by 83.2%. The area under the curve of the proposed model is much more significant than earlier research by 7%, demonstrating that this model is more dependable and was the top-performing algorithm.

An efficient approach for optimizing the cost-effective individualized treatment rule using conditional random forest.

Evidence from observational studies has become increasingly important for supporting healthcare policy making via cost-effectiveness analyses. Similar as in comparative effectiveness studies, health economic evaluations that consider subject-level heterogeneity produce individualized treatment rules that are often more cost-effective than one-size-fits-all treatment. Thus, it is of great interest to develop statistical tools for learning such a cost-effective individualized treatment rule under the causal inference framework that allows proper handling of potential confounding and can be applied to both trials and observational studies. In this paper, we use the concept of net-monetary-benefit to assess the trade-off between health benefits and related costs. We estimate cost-effective individualized treatment rule as a function of patients' characteristics that, when implemented, optimizes the allocation of limited healthcare resources by maximizing health gains while minimizing treatment-related costs. We employ the conditional random forest approach and identify the optimal cost-effective individualized treatment rule using net-monetary-benefit-based classification algorithms, where two partitioned estimators are proposed for the subject-specific weights to effectively incorporate information from censored individuals. We conduct simulation studies to evaluate the performance of our proposals. We apply our top-performing algorithm to the NIH-funded Systolic Blood Pressure Intervention Trial to illustrate the cost-effectiveness gains of assigning customized intensive blood pressure therapy.

Machine Learning for Workflow Applications in Screening Mammography: Systematic Review and Meta-Analysis.

Background Advances in computer processing and improvements in data availability have led to the development of machine learning (ML) techniques for mammographic imaging. Purpose To evaluate the reported performance of stand-alone ML applications for screening mammography workflow. Materials and Methods Ovid Embase, Ovid Medline, Cochrane Central Register of Controlled Trials, Scopus, and Web of Science literature databases were searched for relevant studies published from January 2012 to September 2020. The study was registered with the PROSPERO International Prospective Register of Systematic Reviews (protocol no. CRD42019156016). Stand-alone technology was defined as a ML algorithm that can be used independently of a human reader. Studies were quality assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 and the Prediction Model Risk of Bias Assessment Tool, and reporting was evaluated using the Checklist for Artificial Intelligence in Medical Imaging. A primary meta-analysis included the top-performing algorithm and corresponding reader performance from which pooled summary estimates for the area under the receiver operating characteristic curve (AUC) were calculated using a bivariate model. Results Fourteen articles were included, which detailed 15 studies for stand-alone detection (n = 8) and triage (n = 7). Triage studies reported that 17%-91% of normal mammograms identified could be read by adapted screening, while "missing" an estimated 0%-7% of cancers. In total, an estimated 185 252 cases from three countries with more than 39 readers were included in the primary meta-analysis. The pooled sensitivity, specificity, and AUC was 75.4% (95% CI: 65.6, 83.2; P = .11), 90.6% (95% CI: 82.9, 95.0; P = .40), and 0.89 (95% CI: 0.84, 0.98), respectively, for algorithms, and 73.0% (95% CI: 60.7, 82.6), 88.6% (95% CI: 72.4, 95.8), and 0.85 (95% CI: 0.78, 0.97), respectively, for readers. Conclusion Machine learning (ML) algorithms that demonstrate a stand-alone application in mammographic screening workflows achieve or even exceed human reader detection performance and improve efficiency. However, this evidence is from a small number of retrospective studies. Therefore, further rigorous independent external prospective testing of ML algorithms to assess performance at preassigned thresholds is required to support these claims. ©RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Whitman and Moseley in this issue.

Assessment of the timeliness and robustness for predicting adult sepsis

SummarySepsis is a leading cause of death among inpatients at hospitals. However, with early detection, death rate can drop substantially. In this study, we present the top-performing algorithm for Sepsis II prediction in the DII National Data Science Challenge using the Cerner Health Facts data involving more than 100,000 adult patients. This large sample size allowed us to dissect the predictability by age-groups, race, genders, and care settings and up to 192 hr of sepsis onset. This large data collection also allowed us to conclude that the last six biometric records on average are informative to the prediction of sepsis. We identified biomarkers that are common across the treatment time and novel biomarkers that are uniquely presented for early prediction. The algorithms showed meaningful signals days ahead of sepsis onset, supporting the potential of reducing death rate by focusing on high-risk populations identified from heterogeneous data integration.

Comparison of cloud detection algorithms for Sentinel-2 imagery

Accurate, automated cloud and cloud shadow detection is a key component of the processing needed to prepare optical satellite imagery for scientific analysis. Many existing cloud detection algorithms rely on temperature information to identify clouds, making detection difficult for imagers that lack a thermal band, like Sentinel-2. To get maximum benefit from Sentinel-2 products it is critical to understand which algorithms best identify clouds and their shadows in images. We examined the relative performance of five different cloud-masking algorithms (Sen2Cor, MAJA, LaSRC, Fmask and Tmask) in 6 Sentinel-2 scenes (28 total images) distributed across the Eastern Hemisphere. Expanding on these comparisons, we tested ensemble approaches to improve results. We tested three ensemble approaches to cloud and shadow classification based on the outputs of the five initial algorithms using the cloud masks in: (1) a majority prediction model; (2) a random forests model; and (3) a conditional logic model. Accuracy assessments show a trade-off between omission and commission errors in cloud detection for individual algorithms across all sites, and some algorithms are better at detecting either clouds or cloud shadows. No single algorithm outperforms the others for both clouds and shadows. Aggregating the results from multiple algorithms produces fewer undetected clouds and higher overall accuracy than any single algorithm, with as high as 2.7% improvement over the top-performing algorithm, suggesting an ensemble approach may be the most useful for processing of Sentinel-2 data.

FastClone is a probabilistic tool for deconvoluting tumor heterogeneity in bulk-sequencing samples

Dissecting tumor heterogeneity is a key to understanding the complex mechanisms underlying drug resistance in cancers. The rich literature of pioneering studies on tumor heterogeneity analysis spurred a recent community-wide benchmark study that compares diverse modeling algorithms. Here we present FastClone, a top-performing algorithm in accuracy in this benchmark. FastClone improves over existing methods by allowing the deconvolution of subclones that have independent copy number variation events within the same chromosome regions. We characterize the behavior of FastClone in identifying subclones using stage III colon cancer primary tumor samples as well as simulated data. It achieves approximately 100-fold acceleration in computation for both simulated and patient data. The efficacy of FastClone will allow its application to large-scale data and clinical data, and facilitate personalized medicine in cancers.

Identifying optimal survey-based algorithms to distinguish diabetes type among adults with diabetes

ObjectivesSurveys for U.S. diabetes surveillance do not reliably distinguish between type 1 and type 2 diabetes, potentially obscuring trends in type 1 among adults. To validate survey-based algorithms for distinguishing diabetes type, we linked survey data collected from adult patients with diabetes to a gold standard diabetes type. Research design and methodsWe collected data through a telephone survey of 771 adults with diabetes receiving care in a large healthcare system in North Carolina. We tested 34 survey classification algorithms utilizing information on respondents’ report of physician-diagnosed diabetes type, age at onset, diabetes drug use, and body mass index. Algorithms were evaluated by calculating type 1 and type 2 sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) relative to a gold standard diagnosis of diabetes type determined through analysis of EHR data and endocrinologist review of selected cases. ResultsAlgorithms based on self-reported type outperformed those based solely on other data elements. The top-performing algorithm classified as type 1 all respondents who reported type 1 and were prescribed insulin, as “other diabetes type” all respondents who reported “other,” and as type 2 the remaining respondents (type 1 sensitivity 91.6%, type 1 specificity 98.9%, type 1 PPV 82.5%, type 1 NPV 99.5%). This algorithm performed well in most demographic subpopulations. ConclusionsThe major federal health surveys should consider including self-reported diabetes type if they do not already, as the gains in the accuracy of typing are substantial compared to classifications based on other data elements. This study provides much-needed guidance on the accuracy of survey-based diabetes typing algorithms.

Evaluation of Combined Artificial Intelligence and Radiologist Assessment to Interpret Screening Mammograms

Mammography screening currently relies on subjective human interpretation. Artificial intelligence (AI) advances could be used to increase mammography screening accuracy by reducing missed cancers and false positives. To evaluate whether AI can overcome human mammography interpretation limitations with a rigorous, unbiased evaluation of machine learning algorithms. In this diagnostic accuracy study conducted between September 2016 and November 2017, an international, crowdsourced challenge was hosted to foster AI algorithm development focused on interpreting screening mammography. More than 1100 participants comprising 126 teams from 44 countries participated. Analysis began November 18, 2016. Algorithms used images alone (challenge 1) or combined images, previous examinations (if available), and clinical and demographic risk factor data (challenge 2) and output a score that translated to cancer yes/no within 12 months. Algorithm accuracy for breast cancer detection was evaluated using area under the curve and algorithm specificity compared with radiologists' specificity with radiologists' sensitivity set at 85.9% (United States) and 83.9% (Sweden). An ensemble method aggregating top-performing AI algorithms and radiologists' recall assessment was developed and evaluated. Overall, 144 231 screening mammograms from 85 580 US women (952 cancer positive ≤12 months from screening) were used for algorithm training and validation. A second independent validation cohort included 166 578 examinations from 68 008 Swedish women (780 cancer positive). The top-performing algorithm achieved an area under the curve of 0.858 (United States) and 0.903 (Sweden) and 66.2% (United States) and 81.2% (Sweden) specificity at the radiologists' sensitivity, lower than community-practice radiologists' specificity of 90.5% (United States) and 98.5% (Sweden). Combining top-performing algorithms and US radiologist assessments resulted in a higher area under the curve of 0.942 and achieved a significantly improved specificity (92.0%) at the same sensitivity. While no single AI algorithm outperformed radiologists, an ensemble of AI algorithms combined with radiologist assessment in a single-reader screening environment improved overall accuracy. This study underscores the potential of using machine learning methods for enhancing mammography screening interpretation.

Explainable Machine Learning Algorithms For Predicting Glass Transition Temperatures

Modern technologies demand the development of new glasses with unusual properties. Most of the previous developments occurred by slow, expensive trial-and-error approaches, which have produced a considerable amount of data over the past 100 years. By finding patterns in such types of data, Machine Learning (ML) algorithms can extract useful knowledge, providing important insights into composition-property maps. A key step in glass composition design is to identify their physical-chemical properties, such as the glass transition temperature, Tg. In this paper, we investigate how different ML algorithms can be used to predict the Tg of glasses based on their chemical composition. For such, we used a dataset of 43,240 oxide glass compositions, each one with its assigned Tg. Besides, to assess the predictive performance obtained by ML algorithms, we investigated the possible gains by tuning the hyperparameters of these algorithms. The results show that the best ML algorithm for predicting Tg is the Random Forest (RF). One of the main challenges in this task is the prediction of extreme Tg values. To do this, we assessed the predictive performance of the investigated ML algorithms in three Tg intervals. For extreme Tg values ( ≤ 450 K and ≥ 1150 K), the top-performing algorithm was the k-Nearest Neighbours, closely followed by RF. The induced RF model predicted extreme values of Tg with a Relative Deviation (RD) of 3.5% for glasses with high Tg ( ≥ 1150 K), and RD of 7.5% for glasses with very low Tg ( ≤ 450 K). Finally, we propose a new visual approach to explain what our RF model learned, highlighting the importance of each chemical element to obtain glasses with extreme Tg. This study can be easily expanded to predict other composition–property combinations and can advantageously replace empirical approaches for developing novel glasses with relevant properties and applications.