Development of a multi-fusion convolutional neural network (MF-CNN) for enhanced gastrointestinal disease diagnosis in endoscopy image analysis.

Gastrointestinal (GI) diseases are prevalent medical conditions that require accurate and timely diagnosis for effective treatment. To address this, we developed the Multi-Fusion Convolutional Neural Network (MF-CNN), a deep learning framework that strategically integrates and adapts elements from six deep learning models, enhancing feature extraction and classification of GI diseases from endoscopic images. The MF-CNN architecture leverages truncated and partially frozen layers from existing models, augmented with novel components such as Auxiliary Fusing Layers (AuxFL), Fusion Residual Block (FuRB), and Alpha Dropouts (αDO) to improve precision and robustness. This design facilitates the precise identification of conditions such as ulcerative colitis, polyps, esophagitis, and healthy colons. Our methodology involved preprocessing endoscopic images sourced from open databases, including KVASIR and ETIS-Larib Polyp DB, using adaptive histogram equalization (AHE) to enhance their quality. The MF-CNN framework supports detailed feature mapping for improved interpretability of the model's internal workings. An ablation study was conducted to validate the contribution of each component, demonstrating that the integration of AuxFL, αDO, and FuRB played a crucial part in reducing overfitting and efficiency saturation and enhancing overall model performance. The MF-CNN demonstrated outstanding performance in terms of efficacy, achieving an accuracy rate of 99.25%. It also excelled in other key performance metrics with a precision of 99.27%, a recall of 99.25%, and an F1-score of 99.25%. These metrics confirmed the model's proficiency in accurate classification and its capability to minimize false positives and negatives across all tested GI disease categories. Furthermore, the AUC values were exceptional, averaging 1.00 for both test and validation sets, indicating perfect discriminative ability. The findings of the P-R curve analysis and confusion matrix further confirmed the robust classification performance of the MF-CNN. This research introduces a technique for medical imaging that can potentially transform diagnostics in gastrointestinal healthcare facilities worldwide.

BOO-ST and CBCEC: two novel hybrid machine learning methods aim to reduce the mortality of heart failure patients.

Heart failure (HF) is a leading cause of mortality worldwide. Machine learning (ML) approaches have shown potential as an early detection tool for improving patient outcomes. Enhancing the effectiveness and clinical applicability of the ML model necessitates training an efficient classifier with a diverse set of high-quality datasets. Hence, we proposed two novel hybrid ML methods ((a) consisting of Boosting, SMOTE, and Tomek links (BOO-ST); (b) combining the best-performing conventional classifier with ensemble classifiers (CBCEC)) to serve as an efficient early warning system for HF mortality. The BOO-ST was introduced to tackle the challenge of class imbalance, while CBCEC was responsible for training the processed and selected features derived from the Feature Importance (FI) and Information Gain (IG) feature selection techniques. We also conducted an explicit and intuitive comprehension to explore the impact of potential characteristics correlating with the fatality cases of HF. The experimental results demonstrated the proposed classifier CBCEC showcases a significant accuracy of 93.67% in terms of providing the early forecasting of HF mortality. Therefore, we can reveal that our proposed aspects (BOO-ST and CBCEC) can be able to play a crucial role in preventing the death rate of HF and reducing stress in the healthcare sector.

Attribute Selection Hybrid Network Model for risk factors analysis of postpartum depression using Social media.

BACKGROUND AND OBJECTIVE: Postpartum Depression (PPD) is a frequently ignored birth-related consequence. Social network analysis can be used to address this issue because social media network serves as a platform for their users to communicate with their friends and share their opinions, photos, and videos, which reflect their moods, feelings, and sentiments. In this work, the depression of delivered mothers is identified using the PPD score and segregated into control and depressed groups. Recently, to detect depression, deep learning methods have played a vital role. However, these methods still do not clarify why some people have been identified as depressed. METHODS: We have developed Attribute Selection Hybrid Network (ASHN) to detect the postpartum depression diagnoses framework. Later analysis of the post of mothers who have been confirmed with the score calculated by the experts of the field using physiological questionnaire score. The model works on the analysis of the attributes of the negative Facebook posts for Depressed user Diagnosis, which is a large general forum. This framework explains the process of analyzing posts containing Sentiment, depressive symptoms, and reflective thinking and suggests psycho-linguistic and stylistic attributes of depression in posts. RESULTS: The experimental results show that ASHN works well and is easy to understand. Here, four attribute networks based on psychological studies were used to analyze the different parts of posts by depressed users. The results of the experiments show the extraction of psycho-linguistic markers-based attributes, the recording of assessment metrics including Precision, Recall and F1 score and visualization of those attributes were used title-wise as well as words wise and compared with daily life, depression and postpartum depressed people using Word cloud. Furthermore, a comparison to a reference with Baseline and ASHN model was carried out. CONCLUSIONS: Attribute Selection Hybrid Network (ASHN) mimics the importance of attributes in social media posts to predict depressed mothers. Those mothers were anticipated to be depressed by answering a questionnaire designed by domain experts with prior knowledge of depression. This work will help researchers look at social media posts to find useful evidence for other depressive symptoms.

Identifying predictive biomarkers for repetitive transcranial magnetic stimulation response in depression patients with explainability.

Repetitive Transcranial Magnetic Stimulation (rTMS) is an evidence-based treatment for depression. However, the patterns of response to this treatment modality are inconsistent. Whilst many people see a significant reduction in the severity of their depression following rTMS treatment, some patients do not. To support and improve patient outcomes, recent work is exploring the possibility of using Machine Learning to predict rTMS treatment outcomes. Our proposed model is the first to combine functional magnetic resonance imaging (fMRI) connectivity with deep learning techniques to predict treatment outcomes before treatment starts. Furthermore, with the use of Explainable AI (XAI) techniques, we identify potential biomarkers that may discriminate between rTMS responders and non-responders. Our experiments utilize 200 runs of repeated bootstrap sampling on two rTMS datasets. We compare performances between our proposed feedforward deep neural network against existing methods, and compare the average accuracy, balanced accuracy and F1-score on a held-out test set. The results of these experiments show that our model outperforms existing methods with an average accuracy of 0.9423, balanced accuracy of 0.9423, and F1-score of 0.9461 in a sample of 61 patients. We found that functional connectivity measures between the Subgenual Anterior Cingulate Cortex and Centeral Opercular Cortex are a key determinant of rTMS treatment response. This knowledge provides psychiatrists with further information to explore the potential mechanisms of responses to rTMS treatment. Our developed prototype is ready to be deployed across large datasets in multiple centres and different countries.

A systematic review of literature on credit card cyber fraud detection using machine and deep learning.

The increasing spread of cyberattacks and crimes makes cyber security a top priority in the banking industry. Credit card cyber fraud is a major security risk worldwide. Conventional anomaly detection and rule-based techniques are two of the most common utilized approaches for detecting cyber fraud, however, they are the most time-consuming, resource-intensive, and inaccurate. Machine learning is one of the techniques gaining popularity and playing a significant role in this field. This study examines and synthesizes previous studies on the credit card cyber fraud detection. This review focuses specifically on exploring machine learning/deep learning approaches. In our review, we identified 181 research articles, published from 2019 to 2021. For the benefit of researchers, review of machine learning/deep learning techniques and their relevance in credit card cyber fraud detection is presented. Our review provides direction for choosing the most suitable techniques. This review also discusses the major problems, gaps, and limits in detecting cyber fraud in credit card and recommend research directions for the future. This comprehensive review enables researchers and banking industry to conduct innovation projects for cyber fraud detection.

When can cancer patient treatment nonadherence be considered intentional or unintentional? A scoping review.

BACKGROUND: Treatment nonadherence in cancer patients remains high with most interventions having had limited success. Most studies omit the multi-factorial aspects of treatment adherence and refer to medication adherence. The behaviour is rarely defined as intentional or unintentional. AIM: The aim of this Scoping Review is to increase understanding of modifiable factors in treatment nonadherence through the relationships that physicians have with their patients. This knowledge can help define when treatment nonadherence is intentional or unintentional and can assist in predicting cancer patients at risk of nonadherence and in intervention design. The scoping review provides the basis for method triangulation in two subsequent qualitative studies: 1. Sentiment analysis of online cancer support groups in relation to treatment nonadherence; 2. A qualitative validation survey to refute / or validate claims from this scoping review. Thereafter, framework development for a future (cancer patient) online peer support intervention. METHODS: A Scoping Review was performed to identify peer reviewed studies that concern treatment / medication nonadherence in cancer patients-published between 2000 to 2021 (and partial 2022). The review was registered in the Prospero database CRD42020210340 and follows the PRISMA-S: an extension to the PRISMA Statement for Reporting Literature Searches in Systematic Searches. The principles of meta-ethnography are used in a synthesis of qualitative findings that preserve the context of primary data. An aim of meta-ethnography is to identify common and refuted themes across studies. This is not a mixed methods study, but due to a limited qualitativevidence base and to broaden findings, the qualitative elements (author interpretations) found within relevant quantitative studies have been included. RESULTS: Of 7510 articles identified, 240 full texts were reviewed with 35 included. These comprise 15 qualitative and 20 quantitative studies. One major theme, that embraces 6 sub themes has emerged: 'Physician factors can influence patient factors in treatment nonadherence'. The six (6) subthemes are: 1. Suboptimal Communication; 2. The concept of Information differs between Patient and Physician; 3.Inadequate time. 4. The need for Treatment Concordance is vague or missing from concepts; 5. The importance of Trust in the physician / patient relationship is understated in papers; 6. Treatment concordance as a concept is rarely defined and largely missing from studies. LINE OF ARGUMENT WAS DRAWN: Treatment (or medication) nonadherence that is intentional or unintentional is often attributed to patient factors-with far less attention to the potential influence of physician communication factors. The differentation between intentional or unintentional nonadherence is missing from most qualitative and quantitative studies. The holistic inter-dimensional / multi-factorial concept of 'treatment adherence' receives scant attention. The main focus is on medication adherence / nonadherence in the singular context. Nonadherence that is unintentional is not necessarily passive behaviour and may overlap with intentional nonadherence. The absence of treatment concordance is a barrier to treatment adherence and is rarely articulated or defined in studies. CONCLUSION: This review demonstrates how cancer patient treatment nonadherence is often a shared outcome. An equal focus on physican and patient factors can increase understanding of the two main types of nonadherence (intentional or unintentional). This differentation should help improve the fundamentals of intervention design.

Deep learning and machine learning in psychiatry: a survey of current progress in depression detection, diagnosis and treatment.

Informatics paradigms for brain and mental health research have seen significant advances in recent years. These developments can largely be attributed to the emergence of new technologies such as machine learning, deep learning, and artificial intelligence. Data-driven methods have the potential to support mental health care by providing more precise and personalised approaches to detection, diagnosis, and treatment of depression. In particular, precision psychiatry is an emerging field that utilises advanced computational techniques to achieve a more individualised approach to mental health care. This survey provides an overview of the ways in which artificial intelligence is currently being used to support precision psychiatry. Advanced algorithms are being used to support all phases of the treatment cycle. These systems have the potential to identify individuals suffering from mental health conditions, allowing them to receive the care they need and tailor treatments to individual patients who are mostly to benefit. Additionally, unsupervised learning techniques are breaking down existing discrete diagnostic categories and highlighting the vast disease heterogeneity observed within depression diagnoses. Artificial intelligence also provides the opportunity to shift towards evidence-based treatment prescription, moving away from existing methods based on group averages. However, our analysis suggests there are several limitations currently inhibiting the progress of data-driven paradigms in care. Significantly, none of the surveyed articles demonstrate empirically improved patient outcomes over existing methods. Furthermore, greater consideration needs to be given to uncertainty quantification, model validation, constructing interdisciplinary teams of researchers, improved access to diverse data and standardised definitions within the field. Empirical validation of computer algorithms via randomised control trials which demonstrate measurable improvement to patient outcomes are the next step in progressing models to clinical implementation.

Gynecological cancer prognosis using machine learning techniques: A systematic review of the last three decades (1990-2022).

OBJECTIVE: Many Computer Aided Prognostic (CAP) systems based on machine learning techniques have been proposed in the field of oncology. The objective of this systematic review was to assess and critically appraise the methodologies and approaches used in predicting the prognosis of gynecological cancers using CAPs. METHODS: Electronic databases were used to systematically search for studies utilizing machine learning methods in gynecological cancers. Study risk of bias (ROB) and applicability were assessed using the PROBAST tool. 139 studies met the inclusion criteria, of which 71 predicted outcomes for ovarian cancer patients, 41 predicted outcomes for cervical cancer patients, 28 predicted outcomes for uterine cancer patients, and 2 predicted outcomes for gynecological malignancies broadly. RESULTS: Random forest (22.30 %) and support vector machine (21.58 %) classifiers were used most commonly. Use of clinicopathological, genomic and radiomic data as predictors was observed in 48.20 %, 51.08 % and 17.27 % of studies, respectively, with some studies using multiple modalities. 21.58 % of studies were externally validated. Twenty-three individual studies compared ML and non-ML methods. Study quality was highly variable and methodologies, statistical reporting and outcome measures were inconsistent, preventing generalized commentary or meta-analysis of performance outcomes. CONCLUSION: There is significant variability in model development when prognosticating gynecological malignancies with respect to variable selection, machine learning (ML) methods and endpoint selection. This heterogeneity prevents meta-analysis and conclusions regarding the superiority of ML methods. Furthermore, PROBAST-mediated ROB and applicability analysis demonstrates concern for the translatability of existing models. This review identifies ways that this can be improved upon in future works to develop robust, clinically translatable models within this promising field.

Mobile phone enabled mental health monitoring to enhance diagnosis for severity assessment of behaviours: a review.

Mental health issues are a serious consequence of the COVID-19 pandemic, influencing about 700 million people worldwide. These physiological issues need to be consistently observed on the people through non-invasive devices such as smartphones, and fitness bands in order to remove the burden of having the conciseness of continuously being monitored. On the other hand, technological improvements have enhanced the abilities and roles of conventional mobile phones from simple communication to observations and improved accessibility in terms of size and price may reflect growing familiarity with the smartphone among a vast number of consumers. As a result of continuous monitoring, together with various embedded sensors in mobile phones, raw data can be converted into useful information about the actions and behaviors of the consumers. Thus, the aim of this comprehensive work concentrates on the literature work done so far in the prediction of mental health issues via passive monitoring data from smartphones. This study also explores the way users interact with such self-monitoring technologies and what challenges they might face. We searched several electronic databases (PubMed, IEEE Xplore, ACM Digital Libraries, Soups, APA PsycInfo, and Mendeley Data) for published studies that are relevant to focus on the topic and English language proficiency from January 2015 to December 2020. We identified 943 articles, of which 115 articles were eligible for this scoping review based on the predetermined inclusion and exclusion criteria carried out manually. These studies provided various works regarding smartphones for health monitoring such as Physical activity (26.0 percent; 30/115), Mental health analysis (27.8 percent; 32/115), Student specific monitoring (15.6 percent; 18/115) are the three analyses carried out predominantly.

Triaging Medical Referrals Based on Clinical Prioritisation Criteria Using Machine Learning Techniques.

Triaging of medical referrals can be completed using various machine learning techniques, but trained models with historical datasets may not be relevant as the clinical criteria for triaging are regularly updated and changed. This paper proposes the use of machine learning techniques coupled with the clinical prioritisation criteria (CPC) of Queensland (QLD), Australia, to deliver better triaging for referrals in accordance with the CPC's updates. The unique feature of the proposed model is its non-reliance on the past datasets for model training. Medical Natural Language Processing (NLP) was applied in the proposed approach to process the medical referrals, which are unstructured free text. The proposed multiclass classification approach achieved a Micro F1 score = 0.98. The proposed approach can help in the processing of two million referrals that the QLD health service receives annually; therefore, they can deliver better and more efficient health services.

Application of CycleGAN and transfer learning techniques for automated detection of COVID-19 using X-ray images.

Coronavirus (which is also known as COVID-19) is severely impacting the wellness and lives of many across the globe. There are several methods currently to detect and monitor the progress of the disease such as radiological image from patients' chests, measuring the symptoms and applying polymerase chain reaction (RT-PCR) test. X-ray imaging is one of the popular techniques used to visualise the impact of the virus on the lungs. Although manual detection of this disease using radiology images is more popular, it can be time-consuming, and is prone to human errors. Hence, automated detection of lung pathologies due to COVID-19 utilising deep learning (Bowles et al.) techniques can assist with yielding accurate results for huge databases. Large volumes of data are needed to achieve generalizable DL models; however, there are very few public databases available for detecting COVID-19 disease pathologies automatically. Standard data augmentation method can be used to enhance the models' generalizability. In this research, the Extensive COVID-19 X-ray and CT Chest Images Dataset has been used and generative adversarial network (GAN) coupled with trained, semi-supervised CycleGAN (SSA- CycleGAN) has been applied to augment the training dataset. Then a newly designed and finetuned Inception V3 transfer learning model has been developed to train the algorithm for detecting COVID-19 pandemic. The obtained results from the proposed Inception-CycleGAN model indicated Accuracy = 94.2%, Area under Curve = 92.2%, Mean Squared Error = 0.27, Mean Absolute Error = 0.16. The developed Inception-CycleGAN framework is ready to be tested with further COVID-19 X-Ray images of the chest.

Future IoT tools for COVID-19 contact tracing and prediction: A review of the state-of-the-science.

In 2020 the world is facing unprecedented challenges due to COVID-19. To address these challenges, many digital tools are being explored and developed to contain the spread of the disease. With the lack of availability of vaccines, there is an urgent need to avert resurgence of infections by putting some measures, such as contact tracing, in place. While digital tools, such as phone applications are advantageous, they also pose challenges and have limitations (eg, wireless coverage could be an issue in some cases). On the other hand, wearable devices, when coupled with the Internet of Things (IoT), are expected to influence lifestyle and healthcare directly, and they may be useful for health monitoring during the global pandemic and beyond. In this work, we conduct a literature review of contact tracing methods and applications. Based on the literature review, we found limitations in gathering health data, such as insufficient network coverage. To address these shortcomings, we propose a novel intelligent tool that will be useful for contact tracing and prediction of COVID-19 clusters. The solution comprises a phone application combined with a wearable device, infused with unique intelligent IoT features (complex data analysis and intelligent data visualization) embedded within the system to aid in COVID-19 analysis. Contact tracing applications must establish data collection and data interpretation. Intelligent data interpretation can assist epidemiological scientists in anticipating clusters, and can enable them to take necessary action in improving public health management. Our proposed tool could also be used to curb disease incidence in future global health crises.

Remote Patient Monitoring Using Radio Frequency Identification (RFID) Technology and Machine Learning for Early Detection of Suicidal Behaviour in Mental Health Facilities.

Remote Patient Monitoring (RPM) has gained great popularity with an aim to measure vital signs and gain patient related information in clinics. RPM can be achieved with noninvasive digital technology without hindering a patient's daily activities and can enhance the efficiency of healthcare delivery in acute clinical settings. In this study, an RPM system was built using radio frequency identification (RFID) technology for early detection of suicidal behaviour in a hospital-based mental health facility. A range of machine learning models such as Linear Regression, Decision Tree, Random Forest, and XGBoost were investigated to help determine the optimum fixed positions of RFID reader-antennas in a simulated hospital ward. Empirical experiments showed that Decision Tree had the best performance compared to Random Forest and XGBoost models. An Ensemble Learning model was also developed, took advantage of these machine learning models based on their individual performance. The research set a path to analyse dynamic moving RFID tags and builds an RPM system to help retrieve patient vital signs such as heart rate, pulse rate, respiration rate and subtle motions to make this research state-of-the-art in terms of managing acute suicidal and self-harm behaviour in a mental health ward.

Automated prediction of sepsis using temporal convolutional network.

Multiple organ failure is the trademark of sepsis. Sepsis occurs when the body's reaction to infection causes injury to its tissues and organs. As a consequence, fluid builds up in the tissues causing organ failure and leading to septic shock eventually. Some symptoms of sepsis include fever, arrhythmias, blood vessel leaks, impaired clotting, and generalised inflammation. In order to address the limitations in current diagnosis, we have proposed a cost-effective automated diagnostic tool in this study. A deep temporal convolution network has been developed for the prediction of sepsis. Septic data was fed to the model and a high accuracy and area under ROC curve (AUROC) of 98.8% and 98.0% were achieved respectively, for per time-step metrics. A relatively high accuracy and AUROC of 95.5% and 91.0% were also achieved respectively, for per-patient metrics. This is a novel study in that it has investigated per time-step metrics, compared to other studies which investigated per-patient metrics. Our model has also been evaluated by three validation methods. Thus, the recommended model is robust with high accuracy and precision and has the potential to be used as a tool for the prediction of sepsis in hospitals.

Ensemble neural network approach detecting pain intensity from facial expressions.

This paper reports on research to design an ensemble deep learning framework that integrates fine-tuned, three-stream hybrid deep neural network (i.e., Ensemble Deep Learning Model, EDLM), employing Convolutional Neural Network (CNN) to extract facial image features, detect and accurately classify the pain. To develop the approach, the VGGFace is fine-tuned and integrated with Principal Component Analysis and employed to extract features in images from the Multimodal Intensity Pain database at the early phase of the model fusion. Subsequently, a late fusion, three layers hybrid CNN and recurrent neural network algorithm is developed with their outputs merged to produce image-classified features to classify pain levels. The EDLM model is then benchmarked by means of a single-stream deep learning model including several competing models based on deep learning methods. The results obtained indicate that the proposed framework is able to outperform the competing methods, applied in a multi-level pain detection database to produce a feature classification accuracy that exceeds 89 %, with a receiver operating characteristic of 93 %. To evaluate the generalization of the proposed EDLM model, the UNBC-McMaster Shoulder Pain dataset is used as a test dataset for all of the modelling experiments, which reveals the efficacy of the proposed method for pain classification from facial images. The study concludes that the proposed EDLM model can accurately classify pain and generate multi-class pain levels for potential applications in the medical informatics area, and should therefore, be explored further in expert systems for detecting and classifying the pain intensity of patients, and automatically evaluating the patients' pain level accurately.

Financial competing interests were associated with favorable conclusions and greater author productivity in nonsystematic reviews of neuraminidase inhibitors.

OBJECTIVE: To characterize the conclusions and production of nonsystematic reviews about neuraminidase inhibitors relative to financial competing interests held by the authors. STUDY DESIGN AND SETTING: We searched for articles about neuraminidase inhibitors and influenza (January 2005 to April 2015), identifying nonsystematic reviews and grading them according to the favorable/nonfavorable presentation of evidence on safety and efficacy. We recorded financial competing interests disclosed in the reviews and from other articles written by their authors. We measured associations between competing interests, author productivity, and conclusions. RESULTS: Among 213 nonsystematic reviews, 138 (65%) presented favorable conclusions. Financial competing interests were identified for 26% (137/532) of authors; 51% (108/213) of reviews were associated with a financial competing interest. Reviews produced exclusively by authors with financial competing interests (33%; 71/213) were more likely to present favorable conclusions than reviews with no competing interests (risk ratio 1.27; 95% confidence interval 1.03-1.55). Authors with financial competing interests published more articles about neuraminidase inhibitors than their counterparts. CONCLUSION: Half of nonsystematic reviews about neuraminidase inhibitors included an author with a financial competing interest. Reviews produced exclusively by these authors were more likely to present favorable conclusions, and authors with financial competing interests published a greater number of reviews.

Using social connection information to improve opinion mining: Identifying negative sentiment about HPV vaccines on Twitter.

The manner in which people preferentially interact with others like themselves suggests that information about social connections may be useful in the surveillance of opinions for public health purposes. We examined if social connection information from tweets about human papillomavirus (HPV) vaccines could be used to train classifiers that identify anti-vaccine opinions. From 42,533 tweets posted between October 2013 and March 2014, 2,098 were sampled at random and two investigators independently identified anti-vaccine opinions. Machine learning methods were used to train classifiers using the first three months of data, including content (8,261 text fragments) and social connections (10,758 relationships). Connection-based classifiers performed similarly to content-based classifiers on the first three months of training data, and performed more consistently than content-based classifiers on test data from the subsequent three months. The most accurate classifier achieved an accuracy of 88.6% on the test data set, and used only social connection features. Information about how people are connected, rather than what they write, may be useful for improving public health surveillance methods on Twitter.

Associations Between Exposure to and Expression of Negative Opinions About Human Papillomavirus Vaccines on Social Media: An Observational Study.

BACKGROUND: Groups and individuals that seek to negatively influence public opinion about the safety and value of vaccination are active in online and social media and may influence decision making within some communities. OBJECTIVE: We sought to measure whether exposure to negative opinions about human papillomavirus (HPV) vaccines in Twitter communities is associated with the subsequent expression of negative opinions by explicitly measuring potential information exposure over the social structure of Twitter communities. METHODS: We hypothesized that prior exposure to opinions rejecting the safety or value of HPV vaccines would be associated with an increased risk of posting similar opinions and tested this hypothesis by analyzing temporal sequences of messages posted on Twitter (tweets). The study design was a retrospective analysis of tweets related to HPV vaccines and the social connections between users. Between October 2013 and April 2014, we collected 83,551 English-language tweets that included terms related to HPV vaccines and the 957,865 social connections among 30,621 users posting or reposting the tweets. Tweets were classified as expressing negative or neutral/positive opinions using a machine learning classifier previously trained on a manually labeled sample. RESULTS: During the 6-month period, 25.13% (20,994/83,551) of tweets were classified as negative; among the 30,621 users that tweeted about HPV vaccines, 9046 (29.54%) were exposed to a majority of negative tweets. The likelihood of a user posting a negative tweet after exposure to a majority of negative opinions was 37.78% (2780/7361) compared to 10.92% (1234/11,296) for users who were exposed to a majority of positive and neutral tweets corresponding to a relative risk of 3.46 (95% CI 3.25-3.67, P<.001). CONCLUSIONS: The heterogeneous community structure on Twitter appears to skew the information to which users are exposed in relation to HPV vaccines. We found that among users that tweeted about HPV vaccines, those who were more often exposed to negative opinions were more likely to subsequently post negative opinions. Although this research may be useful for identifying individuals and groups currently at risk of disproportionate exposure to misinformation about HPV vaccines, there is a clear need for studies capable of determining the factors that affect the formation and adoption of beliefs about public health interventions.

Citations alone were enough to predict favorable conclusions in reviews of neuraminidase inhibitors.

OBJECTIVES: To examine the use of supervised machine learning to identify biases in evidence selection and determine if citation information can predict favorable conclusions in reviews about neuraminidase inhibitors. STUDY DESIGN AND SETTING: Reviews of neuraminidase inhibitors published during January 2005 to May 2013 were identified by searching PubMed. In a blinded evaluation, the reviews were classified as favorable if investigators agreed that they supported the use of neuraminidase inhibitors for prophylaxis or treatment of influenza. Reference lists were used to identify all unique citations to primary articles. Three classification methods were tested for their ability to predict favorable conclusions using only citation information. RESULTS: Citations to 4,574 articles were identified in 152 reviews of neuraminidase inhibitors, and 93 (61%) of these reviews were graded as favorable. Primary articles describing drug resistance were among the citations that were underrepresented in favorable reviews. The most accurate classifier predicted favorable conclusions with 96.2% accuracy, using citations to only 24 of 4,574 articles. CONCLUSION: Favorable conclusions in reviews about neuraminidase inhibitors can be predicted using only information about the articles they cite. The approach highlights how evidence exclusion shapes conclusions in reviews and provides a method to evaluate citation practices in a corpus of reviews.