Performance and Agreement When Annotating Chest X-ray Text Reports-A Preliminary Step in the Development of a Deep Learning-Based Prioritization and Detection System.

A chest X-ray report is a communicative tool and can be used as data for developing artificial intelligence-based decision support systems. For both, consistent understanding and labeling is important. Our aim was to investigate how readers would comprehend and annotate 200 chest X-ray reports. Reports written between 1 January 2015 and 11 March 2022 were selected based on search words. Annotators included three board-certified radiologists, two trained radiologists (physicians), two radiographers (radiological technicians), a non-radiological physician, and a medical student. Consensus labels by two or more of the experienced radiologists were considered "gold standard". Matthew's correlation coefficient (MCC) was calculated to assess annotation performance, and descriptive statistics were used to assess agreement between individual annotators and labels. The intermediate radiologist had the best correlation to "gold standard" (MCC 0.77). This was followed by the novice radiologist and medical student (MCC 0.71 for both), the novice radiographer (MCC 0.65), non-radiological physician (MCC 0.64), and experienced radiographer (MCC 0.57). Our findings showed that for developing an artificial intelligence-based support system, if trained radiologists are not available, annotations from non-radiological annotators with basic and general knowledge may be more aligned with radiologists compared to annotations from sub-specialized medical staff, if their sub-specialization is outside of diagnostic radiology.

Inter- and Intra-Observer Agreement When Using a Diagnostic Labeling Scheme for Annotating Findings on Chest X-rays-An Early Step in the Development of a Deep Learning-Based Decision Support System.

Consistent annotation of data is a prerequisite for the successful training and testing of artificial intelligence-based decision support systems in radiology. This can be obtained by standardizing terminology when annotating diagnostic images. The purpose of this study was to evaluate the annotation consistency among radiologists when using a novel diagnostic labeling scheme for chest X-rays. Six radiologists with experience ranging from one to sixteen years, annotated a set of 100 fully anonymized chest X-rays. The blinded radiologists annotated on two separate occasions. Statistical analyses were done using Randolph's kappa and PABAK, and the proportions of specific agreements were calculated. Fair-to-excellent agreement was found for all labels among the annotators (Randolph's Kappa, 0.40-0.99). The PABAK ranged from 0.12 to 1 for the two-reader inter-rater agreement and 0.26 to 1 for the intra-rater agreement. Descriptive and broad labels achieved the highest proportion of positive agreement in both the inter- and intra-reader analyses. Annotating findings with specific, interpretive labels were found to be difficult for less experienced radiologists. Annotating images with descriptive labels may increase agreement between radiologists with different experience levels compared to annotation with interpretive labels.

Research on cross-cultural adaptation and educational management of international students in China: Case of African students at Zhejiang Normal University.

With the closer China-Africa relations and the vigorous development of the country's education for international students, the number of African students studying in China has surged, and the difficulty of education management in colleges and universities has increased accordingly. Many scholars have conducted in-depth research on the educational management model, educational management policy, educational management status and optimization strategy for international students in China. However, the existing research dimensions are relatively single and lack of comprehensiveness. This study takes the educational management practice of African international students in Zhejiang Normal University as the breakthrough point and introduces the investigation of international students' cross-cultural adaptability. A sample of African students coming to China is selected from Zhejiang Normal University (n = 475). This study designs a questionnaire based on the dimensions of enrollment management, teaching management, and daily management combined with the teaching satisfaction, cultural adaptability, and emotional attitude of African international students in China. After data analysis with SPSS 26 and AMOS 24, it can be found that enrollment management is significantly positively correlated with the time in China. Teaching management is positively correlated with age, educational background and it is positively correlated with enrollment management. Differentiated management is positively correlated with Chinese proficiency, and it is positively correlated with enrollment management and teaching management. The recognition of management staffs is positively correlated with enrollment management, teaching management and differentiated management. Teaching satisfaction is positively correlated with enrollment management, teaching management, differentiation management and recognition of management staffs. Cultural adaptation is significantly positively correlated with Chinese proficiency, enrollment management, teaching management, differentiated management, recognition of management staffs, and teaching satisfaction. Emotions and attitudes are significantly positively correlated with age and time in China, and it is positively correlated with teaching management, the recognition of management staffs, teaching satisfaction, cultural adaptation, and academic qualifications. Through in-depth analysis, it can be found that there exist problems in the current education management of African international students, such as insufficient enrollment channels, incomplete enrollment promotion, unreasonable curriculum settings, inadequate implementation of the tutorial system, and excessive differentiated management. In order to further improve the quality of educational management for African international students in China, countermeasures such as adhering to the idea of "a community with a shared future for mankind" to optimize the enrollment management; advocating the value of "harmony but different" to improve teaching management; implementing the "people-oriented" service tenet to improve daily management; deepening community value identification to promote students' cross-cultural adaptation should be implemented.

Local Anaesthetic Thoracoscopy for Pleural Effusion-A Narrative Review.

The incidence of pleural disease is increasing, and interventions are crucial in this subspecialist area of respiratory medicine. One of the cornerstones of pleural effusion investigation and management is medical, which is also known as local anaesthetic thoracoscopy. This allows fluid drainage, biopsy for diagnosis and preventative measures for further fluid potential build-up. This article summarises the evidence around this procedure through a narrative review of the available evidence.

The Added Effect of Artificial Intelligence on Physicians' Performance in Detecting Thoracic Pathologies on CT and Chest X-ray: A Systematic Review.

Our systematic review investigated the additional effect of artificial intelligence-based devices on human observers when diagnosing and/or detecting thoracic pathologies using different diagnostic imaging modalities, such as chest X-ray and CT. Peer-reviewed, original research articles from EMBASE, PubMed, Cochrane library, SCOPUS, and Web of Science were retrieved. Included articles were published within the last 20 years and used a device based on artificial intelligence (AI) technology to detect or diagnose pulmonary findings. The AI-based device had to be used in an observer test where the performance of human observers with and without addition of the device was measured as sensitivity, specificity, accuracy, AUC, or time spent on image reading. A total of 38 studies were included for final assessment. The quality assessment tool for diagnostic accuracy studies (QUADAS-2) was used for bias assessment. The average sensitivity increased from 67.8% to 74.6%; specificity from 82.2% to 85.4%; accuracy from 75.4% to 81.7%; and Area Under the ROC Curve (AUC) from 0.75 to 0.80. Generally, a faster reading time was reported when radiologists were aided by AI-based devices. Our systematic review showed that performance generally improved for the physicians when assisted by AI-based devices compared to unaided interpretation.

Modern day management of a unilateral pleural effusion.

Over 50 systemic conditions may affect the pleura and, thus, unilateral pleural effusions may present for a variety of reasons. Investigating the cause is essential to providing appropriate management. Various pleural interventions are available in current practice, but have varying diagnostic sensitivity. It is, therefore, vital to consider the intervention with the highest diagnostic yield appropriate to the particular clinical situation. The diagnostic pathway in unilateral pleural effusion is increasingly outpatient based, avoiding hospitalisation, which is particularly relevant with the recent COVID-19 pandemic.

Surgically Induced Contrast Enhancements on Intraoperative and Early Postoperative MRI Following High-Grade Glioma Surgery: A Systematic Review.

For the radiological assessment of resection of high-grade gliomas, a 72-h diagnostic window is recommended to limit surgically induced contrast enhancements. However, such enhancements may occur earlier than 72 h post-surgery. This systematic review aimed to assess the evidence on the timing of the postsurgical MRI. PubMed, Embase, Web of Science and Cochrane were searched following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Only original research articles describing surgically induced contrast enhancements on MRI after resection for high-grade gliomas were included and analysed. The frequency of different contrast enhancement patterns on intraoperative MRI (iMRI) and early postoperative MRI (epMRI) was recorded. The search resulted in 1443 studies after removing duplicates, and a total of 12 studies were chosen for final review. Surgically induced contrast enhancements were reported at all time points after surgery, including on iMRI, but their type and frequency vary. Thin linear contrast enhancements were commonly found to be surgically induced and were less frequently recorded on postoperative days 1 and 2. This suggests that the optimal time to scan may be at or before this time. However, the evidence is limited, and higher-quality studies using larger and consecutively sampled populations are needed.

miR-377 inhibition enhances the survival of trophoblast cells via upregulation of FNDC5 in gestational diabetes mellitus.

Gestational diabetes mellitus (GDM) is a metabolic dysregulation closely related to both obesity and type 2 diabetes; however, the molecular mechanism underlying GDM is still unclear. The purpose of this study was to investigate the effects of microRNA-377 (miR-377-3p) and fibronectin type III domain containing 5 (FNDC5) in regulating the cell growth of trophoblasts under high glucose (HG) conditions during the development of GDM. Serum miR-377-3p was upregulated and positively correlated with fasting blood glucose level in GDM patients. miR-377-3p downregulation increased the cell vitality and suppressed the cell apoptosis of HG-treated HTR-8/SVneo and BeWo cells. Using TargetScan prediction, luciferase assay, and western blot, it was found that miR-377-3p could target FNDC5 and suppress its expression. However, FNDC5 downregulation abolished the effect of miR-377-3p inhibitor in HTR-8/SVneo cells. Together, miR-377 is a potential target for GDM biomarker, which promotes cell growth and suppresses cell apoptosis, partly through the upregulation of FNDC5.

The Performance of Deep Learning Algorithms on Automatic Pulmonary Nodule Detection and Classification Tested on Different Datasets That Are Not Derived from LIDC-IDRI: A Systematic Review.

The aim of this study was to systematically review the performance of deep learning technology in detecting and classifying pulmonary nodules on computed tomography (CT) scans that were not from the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) database. Furthermore, we explored the difference in performance when the deep learning technology was applied to test datasets different from the training datasets. Only peer-reviewed, original research articles utilizing deep learning technology were included in this study, and only results from testing on datasets other than the LIDC-IDRI were included. We searched a total of six databases: EMBASE, PubMed, Cochrane Library, the Institute of Electrical and Electronics Engineers, Inc. (IEEE), Scopus, and Web of Science. This resulted in 1782 studies after duplicates were removed, and a total of 26 studies were included in this systematic review. Three studies explored the performance of pulmonary nodule detection only, 16 studies explored the performance of pulmonary nodule classification only, and 7 studies had reports of both pulmonary nodule detection and classification. Three different deep learning architectures were mentioned amongst the included studies: convolutional neural network (CNN), massive training artificial neural network (MTANN), and deep stacked denoising autoencoder extreme learning machine (SDAE-ELM). The studies reached a classification accuracy between 68-99.6% and a detection accuracy between 80.6-94%. Performance of deep learning technology in studies using different test and training datasets was comparable to studies using same type of test and training datasets. In conclusion, deep learning was able to achieve high levels of accuracy, sensitivity, and/or specificity in detecting and/or classifying nodules when applied to pulmonary CT scans not from the LIDC-IDRI database.

Recurrent event survival analysis predicts future risk of hospitalization in patients with paroxysmal and persistent atrial fibrillation.

BACKGROUND: In patients with paroxysmal atrial fibrillation (PAF) or persistent atrial fibrillation (PeAF) symptom burden and fear of hospital readmission are major causes of reduced quality of life. We attempted to develop a prediction model for future atrial fibrillation hospitalization (AFH) risk in PAF and PeAF patients including all previously experienced AFHs in the analysis, as opposed to time to first event. METHODS: Recurrent event survival analysis was used to model the impact of past AFHs on the risk of future AFHs. A recurrent event was defined as a hospitalization due to a new episode of AF. Death or progression to permanent AF were included as competing risks. RESULTS: We enrolled 174 patients with PAF or PeAF, mean follow up duration was 1279 days, and 325 AFHs were observed. Median patient age was 63.0 (IQR 52.2-68.0), 29% had PAF, and 71% were male. Highly significant predictors of future AFH risk were PeAF (HR 3.20, CI 2.01-5.11) and number of past AFHs observed (HR for 1 event: 2.97, CI 2.04-4.32, HR for ≥2 events: 7.54, CI 5.47-10.40). CONCLUSION: In PAF and PeAF patients, AF type and observed AFH frequency are highly significant predictors of future AFH risk. The developed model enables risk prediction in individual patients based on AFH history and baseline characteristics, utilizing all events experienced by the patient. This is the first time recurrent event survival analysis has been used in AF patients.

Females, younger patients and patients with high BMI have the highest pre-operative knee awareness measured using the Forgotten Joint Score.

PURPOSE: The Forgotten Joint Score (FJS) is a novel measurement for patients' awareness of their knee in daily life. By identifying factors that could explain pre-operative FJS levels, the clinician could better prioritize and single out patients who would benefit most from TKA. The aim of this study was to identify possible factors that may explain the variance of pre-operative FJS levels and evaluate the relationship between pre-operative FJS and pre-operative Oxford Knee Score (OKS). METHODS: Four-hundred and six individual knees undergoing primary TKA between 2014 and 2016 were included in the study. Age, gender, body mass index (BMI), pre-operative FJS and pre-operative OKS were obtained maximum 2 weeks prior to surgery. Kellgren-Lawrence (K-L) grade, alignment and joint space width (JSW) were evaluated on pre-operative radiographs. RESULTS: Mean FJS was 21.1 ± 15.6. Females, younger patients and patients with high BMI had significantly the worst pre-operative FJS (p < 0.005). Females scored 6.5 FJS points lower than males. A 0.2-point increase in FJS for every added year indicated improvement in knee awareness with age. A 0.4-point decrease in FJS points for every added BMI point indicated worse knee awareness with higher BMI. There was a strong positive correlation between pre-operative FJS and pre-operative OKS according to the Spearman's rank order test (p < 0.005). CONCLUSIONS: Females, younger patients and patients with high BMI had significantly the worst pre-operative joint awareness. FJS had a strong positive correlation to OKS in pre-operative patients for primary TKA. This information can be used for improved patient selection; clinically continuous low FJS despite weight loss and/or the passing of time may be indication for TKA. LEVEL OF EVIDENCE: Prospective cohort study, Level II.

Stromal cell-derived factor 1α (SDF-1α): A marker of disease burden in patients with atrial fibrillation.

OBJECTIVES: Stromal cell-derived factor 1a (SDF-1α), is a chemokine and is able to home hematopoietic progenitor cells to injured areas of heart tissue for structural repair. Previous studies have found increased levels of SDF-1α in several cardiac diseases, but only few studies have investigated SDF-1α in patients with atrial fibrillation (AF). We aimed to test SDF-1α in a large cohort of patients with AF and its role as a prognostic marker. DESIGN: Between January 1st 2008 to December 1st 2012, 290 patients with ECG documented AF were enrolled from the in- and outpatient clinics at the Department of Cardiology, Hvidovre Hospital, University of Copenhagen, Hvidovre, Denmark. Plasma levels of SDF-1α were measured using ELISA technique. Clinical data were registered and patient follow-up was conducted. RESULTS: Patients with permanent AF had significantly higher SDF-1α levels (2199.5 pg/ml) than the patients with paroxysmal AF (1982.0 pg/ml) and persistent AF (1906.0 pg/ml), p < 0.0005. Higher SDF-1α level was associated with longer time spent in the hospital per readmission, p < 0.05. CONCLUSION: In AF patients, a higher SDF-1α level was found in patients with a more progressive state of arrhythmia and was associated with longer hospitalizations. These findings suggest that SDF-1α could prove valuable in risk stratification and evaluating the disease burden in AF patients.

Field-testing of the ICHD-3 beta/proposed ICD-11 diagnostic criteria for migraine with aura.

INTRODUCTION: In 2013 the International Headache Society published the third International Classification of Headache Disorders beta-version, ICHD-3 beta. Its structure is identical to that of the present proposed version of the International Classification of Diseases (ICD-11), although slightly abbreviated to fulfill the needs of ICD-11. In the following, only ICHD-3 beta is mentioned, but findings regarding the validity of ICHD-3 beta categories are equally relevant to the forthcoming ICD-11. Here we field-tested the criteria for 1.2 migraine with aura (MA), 1.2.1 migraine with typical aura (MTA), 1.2.3 hemiplegic migraine, 1.2.2 migraine with brainstem aura, and the alternative criteria A1.2 MA and A1.2.1 MTA. METHODS: Clinical characteristics were systematically and prospectively collected from patients with 1.2.1 MTA, 1.2.4 familial hemiplegic migraine (FHM), 1.2.5 sporadic hemiplegic migraine (SHM) and 1.2.6 basilar-type migraine according to ICHD-2 in a cross-sectional study design. A database of 2464 patients with 1.1 migraine without aura and 1.2 migraine with non-hemiplegic aura and a database of 252 hemiplegic migraine patients (1.2.4 FHM or 1.2.5 SHM) was collected. We used SPSS 20 for Windows 8.0 for the statistical analysis. RESULTS: All ICHD-2 patients fulfilled ICHD-3 beta criteria for 1.2 MA. The ICHD-3 beta criteria for 1.2.1 MTA were more sensitive than ICHD-2 and ICHD-3 beta alternative criteria; they resulted in fewer probable MA diagnoses. Too many patients fulfilled ICHD-2 and ICHD-3 beta criteria for 1.2.2 migraine with brainstem aura. ICHD-3 beta criteria for 1.2.4 FHM and 1.2.5 SHM both comply with ICHD-2. CONCLUSION: The new criteria in ICHD-3 beta/proposed ICD-11 for 1.2 MA, 1.2.1 MTA, 1.2.3.1 FHM and 1.2.3.2 SHM have more desirable properties than ICHD-2 and the ICHD-3 beta alternative criteria. The criteria for 1.2.2 migraine with brainstem aura should be more restrictive.