Diagnostic Machine Learning Research Articles

Machine Learning Diagnostic Model for Early Stage NSTEMI: Using hs-cTnI 1/2h Changes and Multiple Cardiovascular Biomarkers.

This study demonstrates differences in the distribution of multiple cardiovascular biomarkers between non-ST-segment elevation myocardial infarction (NSTEMI) and unstable angina (UA) patients. Diagnostic machine learning predictive models measured at the time of admission and 1/2 h post-admission, achieving competitive diagnostic predictive results. This study aims to explore the diagnostic value of changes in high-sensitivity cardiac troponin I (hs-cTnI) levels in patients with suspected NSTEMI. A total of 267 patients presented with chest pain, requiring confirmation of acute coronary syndrome (ACS) subtypes (NSTEMI vs. UA). Hs-cTnI and other cardiac markers, such as creatine kinase-MB (CK-MB) and Myoglobin (Myo), were analyzed. Machine learning techniques were employed to assess the application of hs-cTnI level changes in the clinical diagnosis of NSTEMI. Levels of CK-MB, Myo, hs-cTnI measured at admission, hs-cTnI measured 1-2 h after admission, and NT-proBNP in NSTEMI patients were significantly higher than those in UA patients (p < 0.001). There was a positive correlation between hs-cTnI and CK-MB, as well as Myo (R = 0.72, R = 0.51, R = 0.60). The optimal diagnostic model, Hybiome_1/2h, demonstrated an F1-Score of 0.74, an AUROC of 0.96, and an AP of 0.89. This study confirms the significant value of hs-cTnI as a sensitive marker of myocardial injury in the diagnosis of NSTEMI. Continuous monitoring of hs-cTnI levels enhances the accuracy of distinguishing NSTEMI from UA. The models indicate that the Hybiome hs-cTnI assays perform comparably well to the Beckman assays in predicting NSTEMI. Moreover, incorporating hs-cTnI measurements taken 1-2 h post-admission significantly enhances the model's effectiveness.

Comprehensive assessment of machine learning methods for diagnosing gastrointestinal diseases through whole metagenome sequencing data

ABSTRACT The gut microbiome, linked significantly to host diseases, offers potential for disease diagnosis through machine learning (ML) pipelines. These pipelines, crucial in modeling diseases using high-dimensional microbiome data, involve selecting profile modalities, data preprocessing techniques, and classification algorithms, each impacting the model accuracy and generalizability. Despite whole metagenome shotgun sequencing (WMS) gaining popularity for human gut microbiome profiling, a consensus on the optimal methods for ML pipelines in disease diagnosis using WMS data remains elusive. Addressing this gap, we comprehensively evaluated ML methods for diagnosing Crohn’s disease and colorectal cancer, using 2,553 fecal WMS samples from 21 case-control studies. Our study uncovered crucial insights: gut-specific, species-level taxonomic features proved to be the most effective for profiling; batch correction was not consistently beneficial for model performance; compositional data transformations markedly improved the models; and while nonlinear ensemble classification algorithms typically offered superior performance, linear models with proper regularization were found to be more effective for diseases that are linearly separable based on microbiome data. An optimal ML pipeline, integrating the most effective methods, was validated for generalizability using holdout data. This research offers practical guidelines for constructing reliable disease diagnostic ML models with fecal WMS data.

Identifying necrotizing soft tissue infection using infectious fluid analysis and clinical parameters based on machine learning algorithms

BackgroundDetermining the presence of necrotizing soft tissue infection (NSTI) poses a significant hurdle. As of late, there has been a notable increase in the application of artificial intelligence (AI) machine learning techniques in identifying diseases, a shift that can be attributed to their exceptional efficiency, unbiased nature, and high precision. MethodsInformation was gathered from a cohort of 13 patients suffering from NSTI, alongside 12 patients with cellulitis. The construction of NSTI diagnostic machine learning models utilized four different algorithms, specifically random forest, k-nearest neighbors (KNN), support vector machine (SVM), and logistic regression. These models were constructed based on 28 distinctive attributes identified through statistical examination. Following this, the diagnostic efficiency of each algorithms was evaluated. A novel random forest model, streamlined for clinical use, was later developed by focusing on 6 attributes that had the most pronounced influence on the accuracy of our initial random forest model. ResultsThe following data was noted regarding the sensitivity and specificity of the four NSTI diagnostic models:logistic regression displayed 78.2 % and 83.7 %, KNN presented 79.1 % and 87.1 %, SVM showed 83.5 % and 86.3 %, and random forest exhibited 89.6 % and 92.9 %, respectively. In comparison, lactate levels in fluid demonstrated 100 % sensitivity and 76.9 % specificity at an optimal cut-off point of 69.6 mg/dL. Among all four machine learning models, random forest outperformed the others and also showed better results than fluid lactate. A newly constructed random forest model, created using 6 of the 13 identified features, displayed promising results in diagnosing NSTI, having a sensitivity and specificity of 90.2 % and 92.2 %, respectively. ConclusionsDeveloping a diagnostic model for NSTI employing the random forest algorithm has resulted in a diagnostic technique that is more efficient, cost-effective, and expedient. This approach could provide healthcare practitioners with the tools to identify and manage NSTI with greater efficacy.

A Perspective on Crowdsourcing and Human-in-the-Loop Workflows in Precision Health.

Modern machine learning approaches have led to performant diagnostic models for a variety of health conditions. Several machine learning approaches, such as decision trees and deep neural networks, can, in principle, approximate any function. However, this power can be considered to be both a gift and a curse, as the propensity toward overfitting is magnified when the input data are heterogeneous and high dimensional and the output class is highly nonlinear. This issue can especially plague diagnostic systems that predict behavioral and psychiatric conditions that are diagnosed with subjective criteria. An emerging solution to this issue is crowdsourcing, where crowd workers are paid to annotate complex behavioral features in return for monetary compensation or a gamified experience. These labels can then be used to derive a diagnosis, either directly or by using the labels as inputs to a diagnostic machine learning model. This viewpoint describes existing work in this emerging field and discusses ongoing challenges and opportunities with crowd-powered diagnostic systems, a nascent field of study. With the correct considerations, the addition of crowdsourcing to human-in-the-loop machine learning workflows for the prediction of complex and nuanced health conditions can accelerate screening, diagnostics, and ultimately access to care.

Abstract 3896: Time to diagnosis analysis using miRNA-based ovarian cancer prediction models

Abstract Background: The recent literature proposes several diagnostic Machine Learning (ML) models for ovarian cancer based on miRNA expression profiling. These ML models are trained and validated on subjects whose miRNA sample was drawn proximate to cancer diagnosis (e.g., within one month). Whether these models remain useful remote from the time of diagnosis, when early detection or prevention would be most relevant, is unclear. Therefore, we examine the effect of time between blood draw and cancer diagnosis on miRNA-based ML model accuracy and the relevance of a cancer probability score, Pc, for estimating long-term cancer risk. Methods: The study is based on the miRNA expression profiles of 2983 total subjects, which is comprised of 1829 subjects collected as part of the Biobank at Mass General Brigham (MGB), 110 samples from the Pelvic Mass Protocol at Brigham and Women's Hospital (Cramer), and 1044 samples which were obtained from the Prostate, Lung, Colorectal, and Ovarian (PLCO) cancer screening trial. miRNA expression was measured using a pre-specified panel of 179 miRNAs optimized for serum detection using the Fireplex® circulating miRNA assay. We trained ML models using 1865 controls and 74 ovarian cancer subjects (1829 Biobank + 55 Cramer + 55 PLCO). The training cancer samples were drawn between 1 day and 14 years from diagnosis, most within 30 days. The models were then validated on 769 control and 275 ovarian cancer subjects (989 PLCO + 55 Cramer) whose time to cancer diagnosis ranges between 1 and 1814 days (up to 5 years) after blood draw. Performance was reported as Area Under the receiver operator characteristic Curve (AUC). Results: Among the validation set cases, we observe a decreasing trend in predicted cancer probability (Pc) with increasing log time (R = -0.34, p &amp;lt; 0.0001), and we see a similar trend in AUC score with time. On samples drawn within 21 days of cancer diagnosis, the ML model offers an AUC = 0.88, which decreases to AUC = 0.72 on samples drawn between 21 days and one year from diagnosis, and later plateaus at AUC ~ 0.72 up to five years from diagnosis. Out of 2928 total subjects considered in the study, 286 had multiple blood draws taken over a 5-year time period. Using this data, we analyze the change in Pc over time per subject. The results show that Pc, for the average case subject, increased by 7% per year (p = 0.02). In contrast, when we applied the same analysis to control subjects, Pc increased by only 1% per year on average (p = 0.62). Thus, monitoring changes in Pc at regular intervals could be an informative cancer diagnostic. In terms of relative risk, subjects with Pc &amp;lt; 0.5 had a relative 5-year cancer risk of 0.74, whereas subjects with Pc &amp;gt; 0.5 had relative risk 7.4 (i.e., an order of magnitude higher). Conclusion: The results indicate that miRNA-based ML models can be used to identify individuals at increased long-term risk of ovarian cancer and provide a tool for profiling at regular intervals to allow earlier diagnosis of disease. Citation Format: James Webber, Laura Wollborn, Sudhanshu Mishra, Stephanie Alimena, Bryanna Testino, Allison Vitonis, Daniel Cramer, Dipanjan Chowdhury, Kevin Elias. Time to diagnosis analysis using miRNA-based ovarian cancer prediction models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3896.

Diagnostic machine learning applications on clinical populations using functional near infrared spectroscopy: a review.

Functional near-infrared spectroscopy (fNIRS) and its interaction with machine learning (ML) is a popular research topic for the diagnostic classification of clinical disorders due to the lack of robust and objective biomarkers. This review provides an overview of research on psychiatric diseases by using fNIRS and ML. Article search was carried out and 45 studies were evaluated by considering their sample sizes, used features, ML methodology, and reported accuracy. To our best knowledge, this is the first review that reports diagnostic ML applications using fNIRS. We found that there has been an increasing trend to perform ML applications on fNIRS-based biomarker research since 2010. The most studied populations are schizophrenia (n=12), attention deficit and hyperactivity disorder (n=7), and autism spectrum disorder (n=6) are the most studied populations. There is a significant negative correlation between sample size (>21) and accuracy values. Support vector machine (SVM) and deep learning (DL) approaches were the most popular classifier approaches (SVM=20) (DL=10). Eight of these studies recruited a number of participants more than 100 for classification. Concentration changes in oxy-hemoglobin (ΔHbO) based features were used more than concentration changes in deoxy-hemoglobin (ΔHb) based ones and the most popular ΔHbO-based features were mean ΔHbO (n=11) and ΔHbO-based functional connections (n=11). Using ML on fNIRS data might be a promising approach to reveal specific biomarkers for diagnostic classification.

Assessing the feasibility of applying machine learning to diagnosing non-effusive feline infectious peritonitis

Feline infectious peritonitis (FIP) is a severe feline coronavirus-associated syndrome in cats, which is invariably fatal without anti-viral treatment. In the majority of non-effusive FIP cases encountered in practice, confirmatory diagnostic testing is not undertaken and reliance is given to the interpretation of valuable, but essentially non-specific, clinical signs and laboratory markers. We hypothesised that it may be feasible to develop a machine learning (ML) approach which may be applied to the analysis of clinical data to aid in the diagnosis of disease. A dataset encompassing 1939 suspected FIP cases was scored for clinical suspicion of FIP on the basis of history, signalment, clinical signs and laboratory results, using published guidelines, comprising 683 FIP (35.2%), and 1256 non-FIP (64.8%) cases. This dataset was used to train, validate and evaluate two diagnostic machine learning ensemble models. These models, which analysed signalment and laboratory data alone, allowed the accurate discrimination of FIP and non-FIP cases in line with expert opinion. To evaluate whether these models may have value as a diagnostic tool, they were applied to a collection of 80 cases for which the FIP status had been confirmed (FIP: n = 58 (72.5%), non–FIP: n = 22 (27.5%)). Both ensemble models detected FIP with an accuracy of 97.5%, an area under the curve (AUC) of 0.969, sensitivity of 95.45% and specificity of 98.28%. This work demonstrates that, in principle, ML can be usefully applied to the diagnosis of non-effusive FIP. Further work is required before ML may be deployed in the laboratory as a diagnostic tool, such as training models on datasets of confirmed cases and accounting for inter-laboratory variation. Nevertheless, these results illustrate the potential benefit of applying ML to standardising and accelerating the interpretation of clinical pathology data, thereby improving the diagnostic utility of existing laboratory tests.

A Diagnostic Machine Learning Algorithm for Air Brakes in Commercial Vehicles⁎

To safely introduce any level of autonomy to trucks, the health of their brake systems needs to be monitored continuously. Out-of-adjustment push rods and leakages in the air brake system are two major reasons for increased braking distances in trucks, and result in safety violations. Air leakages can occur due to small cracks or loose/improperly ft couplings which do not affect the overall braking capacity but contribute greatly to increasing the braking lag and reducing the maximum braking torque at the wheels. Similarly, an increased stroke of push rod leads to a larger delay in brake response and a smaller value of the brake torque at the wheels. Currently, the condition of an air brake system is inspected manually by measuring the push rod offset, checking the couplings and hoses of the system for air leakages. These inspections are highly labor intensive, subjective, time consuming and do not accurately quantify how adversely the braking system is affected. Having an on-board diagnostic device that can monitor the health of air brakes would be crucial in the prevention of road accidents, especially when considering any level of automation and comply with FMSCA safety requirements. The focus of this paper is to aid the development of such a diagnostic system that facilitates enforcement and pre-trip inspections and continuous on-board monitoring of trucks through the development of a model for its multi-chamber braking system; this model can be used to estimate the severity of leakage and the push rod stroke using real time brake pressure transients. A machine learning algorithm for estimating the air leakage in the air brake system and its experimental corroboration is presented in this paper.

Diagnostic suspicion bias and machine learning: Breaking the awareness deadlock for sepsis detection.

Many early warning algorithms are downstream of clinical evaluation and diagnostic testing, which means that they may not be useful when clinicians fail to suspect illness and fail to order appropriate tests. Depending on how such algorithms handle missing data, they could even indicate "low risk" simply because the testing data were never ordered. We considered predictive methodologies to identify sepsis at triage, before diagnostic tests are ordered, in a busy Emergency Department (ED). One algorithm used "bland clinical data" (data available at triage for nearly every patient). The second algorithm added three yes/no questions to be answered after the triage interview. Retrospectively, we studied adult patients from a single ED between 2014-16, separated into training (70%) and testing (30%) cohorts, and a final validation cohort of patients from four EDs between 2016-2018. Sepsis was defined per the Rhee criteria. Investigational predictors were demographics and triage vital signs (downloaded from the hospital EMR); past medical history; and the auxiliary queries (answered by chart reviewers who were blinded to all data except the triage note and initial HPI). We developed L2-regularized logistic regression models using a greedy forward feature selection. There were 1164, 499, and 784 patients in the training, testing, and validation cohorts, respectively. The bland clinical data model yielded ROC AUC's 0.78 (0.76-0.81) and 0.77 (0.73-0.81), for training and testing, respectively, and ranged from 0.74-0.79 in four hospital validation. The second model which included auxiliary queries yielded 0.84 (0.82-0.87) and 0.83 (0.79-0.86), and ranged from 0.78-0.83 in four hospital validation. The first algorithm did not require clinician input but yielded middling performance. The second showed a trend towards superior performance, though required additional user effort. These methods are alternatives to predictive algorithms downstream of clinical evaluation and diagnostic testing. For hospital early warning algorithms, consideration should be given to bias and usability of various methods.

Consolidated Reporting Guidelines for Prognostic and Diagnostic Machine Learning Modeling Studies: Development and Validation.

The reporting of machine learning (ML) prognostic and diagnostic modeling studies is often inadequate, making it difficult to understand and replicate such studies. To address this issue, multiple consensus and expert reporting guidelines for ML studies have been published. However, these guidelines cover different parts of the analytics lifecycle, and individually, none of them provide a complete set of reporting requirements. We aimed to consolidate the ML reporting guidelines and checklists in the literature to provide reporting items for prognostic and diagnostic ML in in-silico and shadow mode studies. We conducted a literature search that identified 192 unique peer-reviewed English articles that provide guidance and checklists for reporting ML studies. The articles were screened by their title and abstract against a set of 9 inclusion and exclusion criteria. Articles that were filtered through had their quality evaluated by 2 raters using a 9-point checklist constructed from guideline development good practices. The average κ was 0.71 across all quality criteria. The resulting 17 high-quality source papers were defined as having a quality score equal to or higher than the median. The reporting items in these 17 articles were consolidated and screened against a set of 6 inclusion and exclusion criteria. The resulting reporting items were sent to an external group of 11 ML experts for review and updated accordingly. The updated checklist was used to assess the reporting in 6 recent modeling papers in JMIR AI. Feedback from the external review and initial validation efforts was used to improve the reporting items. In total, 37 reporting items were identified and grouped into 5 categories based on the stage of the ML project: defining the study details, defining and collecting the data, modeling methodology, model evaluation, and explainability. None of the 17 source articles covered all the reporting items. The study details and data description reporting items were the most common in the source literature, with explainability and methodology guidance (ie, data preparation and model training) having the least coverage. For instance, a median of 75% of the data description reporting items appeared in each of the 17 high-quality source guidelines, but only a median of 33% of the data explainability reporting items appeared. The highest-quality source articles tended to have more items on reporting study details. Other categories of reporting items were not related to the source article quality. We converted the reporting items into a checklist to support more complete reporting. Our findings supported the need for a set of consolidated reporting items, given that existing high-quality guidelines and checklists do not individually provide complete coverage. The consolidated set of reporting items is expected to improve the quality and reproducibility of ML modeling studies.

GBDTMO: as new option for early-stage breast cancer detection and classification using machine learning

Breast cancer is the second leading cause of disease death in women, after lung and bronchus cancer. According to measurements, mammography misses breast cancer in 10% to 15% of cases for women aged 50 to 69 years. In the current study, we used the Wisconsin breast cancer dataset to develop a two-stage model for breast cancer diagnosis. The main goal of this study effort is to effectively carry out feature selection and classification tasks. Gradient Boosting Decision Tree-based Mayfly Optimisation (GBDTMO), an innovative and efficient breast cancer diagnostic machine learning system, is provided. In the second stage, we employ a Mayfly search to determine which subset of traits is the best. Two more well-known datasets on breast cancer, the ICCR and the Cancer Corpus, were also compared for classification accuracy. The accuracy of the suggested GBDTMO model was higher than that of the existing GBDT and Practical Federated Gradient Boosting Decision Tree (PFGBDT), which had accuracy values of 93.25% and 94.25%, respectively. Similarly, the recall, F-measure, and ROC area values were 98.52%, 97.52%, and 96.32%, respectively. Furthermore, it demonstrated a lower RMSE of 0.98 than the existing GBDT and PFGBDT.

Applications of K-nearest neighbor algorithm in intelligent diagnosis of wind turbine blades damage

Heavy losses in the wind power generation system are incurred when the wind turbine blades are severely damaged beyond repair. Therefore, an immediate repair at an initial stage is an economical method. There are limitations and difficulties in diagnosing initial damage to wind turbine blades. For example, the measurement cannot be performed during actual operation, the diagnostic machine learning model calculations are too complex, and so on. This study measured the noise signals in the operation of wind turbines. The k-nearest neighbor (k-NN) of supervised learning was used as the diagnostic method. The calculation of generalized fractal dimensions (GFDs) was used as diagnostic feature selection. The result shows an accuracy of 98.9%. Compared to other algorithms, the k-NN algorithm is simple and easy to understand and implement. To reduce the amount of calculation of the machine learning model, the optimum numerical combination of three major parameters is found in this study. These major parameters are (1) scale index of GFDs, (2) number of neighbor points in the algorithm, and (3) range formula. According to the findings, the k-NN algorithm model can achieve high accuracy. The optimum numerical combination of three major parameters provides a rapid, convenient, and efficient diagnostic method.

Reporting and Methodological Observations on Prognostic and Diagnostic Machine Learning Studies

Common reporting and methodological patterns were observed from the peer reviews of prognostic and diagnostic machine learning modeling studies submitted to JMIR AI. In this editorial, we summarized some key observations to inform future studies and their reporting.

Prediction and diagnosis of chronic kidney disease development and progression using machine-learning: Protocol for a systematic review and meta-analysis of reporting standards and model performance.

Chronic Kidney disease (CKD) is an important yet under-recognized contributor to morbidity and mortality globally. Machine-learning (ML) based decision support tools have been developed across many aspects of CKD care. Notably, algorithms developed in the prediction and diagnosis of CKD development and progression may help to facilitate early disease prevention, assist with early planning of renal replacement therapy, and offer potential clinical and economic benefits to patients and health systems. Clinical implementation can be affected by the uncertainty surrounding the methodological rigor and performance of ML-based models. This systematic review aims to evaluate the application of prognostic and diagnostic ML tools in CKD development and progression. The protocol has been prepared using the Preferred Items for Systematic Review and Meta-analysis Protocols (PRISMA-P) guidelines. The systematic review protocol for CKD prediction and diagnosis have been registered with the International Prospective Register of Systematic Reviews (PROSPERO) (CRD42022356704, CRD42022372378). A systematic search will be undertaken of PubMed, Embase, the Cochrane Central Register of Controlled Trials (CENTRAL), the Web of Science, and the IEEE Xplore digital library. Studies in which ML has been applied to predict and diagnose CKD development and progression will be included. The primary outcome will be the comparison of the performance of ML-based models with non-ML-based models. Secondary analysis will consist of model use cases, model construct, and model reporting quality. This systematic review will offer valuable insight into the performance and reporting quality of ML-based models in CKD diagnosis and prediction. This will inform clinicians and technical specialists of the current development of ML in CKD care, as well as direct future model development and standardization.

XDecompo: Explainable Decomposition Approach in Convolutional Neural Networks for Tumour Image Classification.

Of the various tumour types, colorectal cancer and brain tumours are still considered among the most serious and deadly diseases in the world. Therefore, many researchers are interested in improving the accuracy and reliability of diagnostic medical machine learning models. In computer-aided diagnosis, self-supervised learning has been proven to be an effective solution when dealing with datasets with insufficient data annotations. However, medical image datasets often suffer from data irregularities, making the recognition task even more challenging. The class decomposition approach has provided a robust solution to such a challenging problem by simplifying the learning of class boundaries of a dataset. In this paper, we propose a robust self-supervised model, called XDecompo, to improve the transferability of features from the pretext task to the downstream task. XDecompo has been designed based on an affinity propagation-based class decomposition to effectively encourage learning of the class boundaries in the downstream task. XDecompo has an explainable component to highlight important pixels that contribute to classification and explain the effect of class decomposition on improving the speciality of extracted features. We also explore the generalisability of XDecompo in handling different medical datasets, such as histopathology for colorectal cancer and brain tumour images. The quantitative results demonstrate the robustness of XDecompo with high accuracy of 96.16% and 94.30% for CRC and brain tumour images, respectively. XDecompo has demonstrated its generalization capability and achieved high classification accuracy (both quantitatively and qualitatively) in different medical image datasets, compared with other models. Moreover, a post hoc explainable method has been used to validate the feature transferability, demonstrating highly accurate feature representations.

Accuracy of Ultrasound Diagnosis of Thyroid Nodules Based on Artificial Intelligence-Assisted Diagnostic Technology: A Systematic Review and Meta-Analysis.

Background Ultrasonography (US) is the most common method of identifying thyroid nodules, but US images require an experienced surgeon for identification. Many artificial intelligence (AI) techniques such as computer-aided diagnostic systems (CAD), deep learning (DL), and machine learning (ML) have been used to assist in the diagnosis of thyroid nodules, but whether AI techniques can improve the diagnostic accuracy of thyroid nodules still needs to be explored. Objective To clarify the accuracy of AI-based thyroid nodule US images for differentiating benign and malignant thyroid nodules. Methods A search strategy of “subject terms + key words” was used to search PubMed, Cochrane Library, Embase, Web of Science, China Biology Medicine (CBM), and China National Knowledge Infrastructure (CNKI) for studies on AI-assisted diagnosis of thyroid nodules based on US images. The summarized receiver operating characteristic (SROC) curve and the pooled sensitivity and specificity were used to assess the performance of the diagnostic tests. The quality assessment of diagnostics accuracy studies-2 (QUADAS-2) tool was used to assess the methodological quality of the included studies. The Review Manager 5.3 and Stata 15 were used to process the data. Subgroup analysis was based on the integrity of data collection. Results A total of 25 studies with 17,429 US images of thyroid nodules were included. AI-assisted diagnostic techniques had better diagnostic efficacy in the diagnosis of benign and malignant thyroid nodules: sensitivity 0.88 (95% CI: (0.85–0.90)), specificity 0.81 (95% CI: 0.74–0.86), diagnostic odds ratio (DOR) 30 (95% CI: 19–46). The SROC curve indicated that the area under the curve (AUC) was 0.92 (95% CI: 0.89–0.94). Threshold effect analysis showed a Spearman correlation coefficient: 0.17 < 0.5, suggesting no threshold effect for the included studies. After a meta-regression analysis of 4 different subgroups, the results showed a statistically significant effect of mean age ≥50 years on heterogeneity. Compared with studies with an average age of ≥50 years, AI-assisted diagnostic techniques had higher diagnostic performance in studies with an average age of <50 years (0.89 (95% CI: 0.87–0.92) vs. 0.80 (95% CI: 0.73–0.88)), (0.83 (95% CI: 0.77–0.88) vs. 0.73 (95% CI: 0.60–0.87)). Conclusions AI-assisted diagnostic techniques had good diagnostic efficacy for thyroid nodules. For the diagnosis of <50 year olds, AI-assisted diagnostic technology was more effective in diagnosis.

A low-cost pavement-rating system, based on machine learning, utilising smartphone sensors

One of the key innovations of this study is the development of a cost-efficient approach for pavement monitoring. This work aims to develop a system that evaluates roadway pavement surface conditions with improved frequency by utilising unsupervised machine learning algorithms and smartphone sensors. The evaluation of roadways utilising complex contemporary data sets is currently conducted periodically because of the high cost of collection methods. For this purpose, the study presents a data-driven framework on the use of a vehicle, a smartphone, an on-board diagnostic device and machine learning for the rating of pavement surfaces, while statistical features are considered in both time- and frequency-domain forms. The selection of features is performed utilising unsupervised classification algorithms. Moreover, the proposed system architecture has been field-tested for the detection of pavement anomalies and the classification of five rating categories. Furthermore, the proposed system may provide daily information on roadway pavement surface conditions, which can be used by agencies for automating the planning of pavement maintenance operations and for improving driving safety.

METHOD FOR EXTRACTING DIAGNOSTIC FEATURES OF THE FACILITIES TECHNICAL CONDITION IN THE SYSTEM FOR MONITORING

Technical diagnostics of facilities is an urgent problem during its operation. An integral part of the implementation of diagnostic monitoring systems is the development of a decision support system (DSS) based on the analysis of acoustic emission (AE) diagnostic data and machine learning methods. A necessary condition for the application of machine learning methods in the development of DSS is the process of extracting diagnostic features from the AE signal. In the present work, an improved method is proposed for extracting diagnostic features from time series of AE signals. This includes two successive steps. At the first step, the frequency and frequency-time characteristics are calculated in a sliding window of short duration, which describe local changes in the shape and structure of single pulses. At the second step, the resulting matrix of informative features is aggregated by calculating statistical moments of various orders, which makes it possible to effectively detect long-term trends in the AE signal changes emitted by the defect. Verification of the proposed method was carried out on a full-scale control object of the oil tank RVS No. 3 ("NTEK LLC"). Based on the results obtained, a conclusion was made about the effectiveness of the proposed method in the development of diagnostic monitoring systems based on acoustic emission data.

Application of PCA and SVM in Fault Detection and Diagnosis of Bearings with Varying Speed

Vibration analysis is widely used as an efficient condition monitoring (CM) tool for rotating machines in various industries. Fault detection and diagnosis (FDD) models play an important role in the development of any CM system. The use of artificial intelligence (AI) has since gained recognition in the development of fault detection and diagnosis systems. In this paper, a combination of principal component analysis (PCA) which is used for reducing the data dimensionality, and support vector machine (SVM) which is adopted for classification to carry out fault detection and diagnosis of faults in bearings using vibrations. The diagnostic feature design and machine learning toolbox in MATLAB are used to develop features and train the models, respectively. Real data from the Mendeley data depository is used to test and evaluate the models. Model training is carried out using data with varying speeds representing different conditions of bearing making it different from similar approaches involving SVM. The choice of data used proves that SVM can be able to classify faults with consideration of the varying operating speeds. Results have shown that the combination of PCA and SVM is effective in fault diagnosis of bearing faults under varying speeds such that a 97.4% classification accuracy was achieved. The result implies that PCA and SVM can be implemented in various industrial setups where variable speeds can occur both intentionally or nonintentionally. Furthermore, the method was able to differentiate between compounding faults and faults that occur at different times. The confusion matrix further proves the quality and accuracy of the trained model. Future work will focus on the development of models that can carry out the prognosis of faults in bearings as well as to model for other faults other than bearing faults.

Monitoring COVID-19 on Social Media: Development of an End-to-End Natural Language Processing Pipeline Using a Novel Triage and Diagnosis Approach.

BackgroundThe COVID-19 pandemic has created a pressing need for integrating information from disparate sources in order to assist decision makers. Social media is important in this respect; however, to make sense of the textual information it provides and be able to automate the processing of large amounts of data, natural language processing methods are needed. Social media posts are often noisy, yet they may provide valuable insights regarding the severity and prevalence of the disease in the population. Here, we adopt a triage and diagnosis approach to analyzing social media posts using machine learning techniques for the purpose of disease detection and surveillance. We thus obtain useful prevalence and incidence statistics to identify disease symptoms and their severities, motivated by public health concerns.ObjectiveThis study aims to develop an end-to-end natural language processing pipeline for triage and diagnosis of COVID-19 from patient-authored social media posts in order to provide researchers and public health practitioners with additional information on the symptoms, severity, and prevalence of the disease rather than to provide an actionable decision at the individual level.MethodsThe text processing pipeline first extracted COVID-19 symptoms and related concepts, such as severity, duration, negations, and body parts, from patients’ posts using conditional random fields. An unsupervised rule-based algorithm was then applied to establish relations between concepts in the next step of the pipeline. The extracted concepts and relations were subsequently used to construct 2 different vector representations of each post. These vectors were separately applied to build support vector machine learning models to triage patients into 3 categories and diagnose them for COVID-19.ResultsWe reported macro- and microaveraged F1 scores in the range of 71%-96% and 61%-87%, respectively, for the triage and diagnosis of COVID-19 when the models were trained on human-labeled data. Our experimental results indicated that similar performance can be achieved when the models are trained using predicted labels from concept extraction and rule-based classifiers, thus yielding end-to-end machine learning. In addition, we highlighted important features uncovered by our diagnostic machine learning models and compared them with the most frequent symptoms revealed in another COVID-19 data set. In particular, we found that the most important features are not always the most frequent ones.ConclusionsOur preliminary results show that it is possible to automatically triage and diagnose patients for COVID-19 from social media natural language narratives, using a machine learning pipeline in order to provide information on the severity and prevalence of the disease for use within health surveillance systems.