Terms Of Diagnostic Performance Research Articles

Comparison of immunoassay- with mass spectrometry-derived p-tau quantification for the detection of Alzheimer’s disease pathology

BackgroundAntibody-based immunoassays have enabled quantification of very low concentrations of phosphorylated tau (p-tau) protein forms in cerebrospinal fluid (CSF), aiding in the diagnosis of AD. Mass spectrometry enables absolute quantification of multiple p-tau variants within a single run. The goal of this study was to compare the performance of mass spectrometry assessments of p-tau181, p-tau217 and p-tau231 with established immunoassay techniques.MethodsWe measured p-tau181, p-tau217 and p-tau231 concentrations in CSF from 173 participants from the TRIAD cohort and 394 participants from the BioFINDER-2 cohort using both mass spectrometry and immunoassay methods. All subjects were clinically evaluated by dementia specialists and had amyloid-PET and tau-PET assessments. Bland–Altman analyses evaluated the agreement between immunoassay and mass spectrometry p-tau181, p-tau217 and p-tau231. P-tau associations with amyloid-PET and tau-PET uptake were also compared. Receiver Operating Characteristic (ROC) analyses compared the performance of mass spectrometry and immunoassays p-tau concentrations to identify amyloid-PET positivity.ResultsMass spectrometry and immunoassays of p-tau217 were highly comparable in terms of diagnostic performance, between-group effect sizes and associations with PET biomarkers. In contrast, p-tau181 and p-tau231 concentrations measured using antibody-free mass spectrometry had lower performance compared with immunoassays.ConclusionsOur results suggest that while similar overall, immunoassay-based p-tau biomarkers are slightly superior to antibody-free mass spectrometry-based p-tau biomarkers. Future work is needed to determine whether the potential to evaluate multiple biomarkers within a single run offsets the slightly lower performance of antibody-free mass spectrometry-based p-tau quantification.

Role of Renal Biopsy in the Management of Renal Cancer: Concordance between Ultrasound/CT-Guided Biopsy Results and Definitive Pathology, Adverse Events, and Complication Rate.

(1) Background: In the last decade, the number of detected renal cancer cases has increased, with the highest incidence in Western countries. Although renal biopsy is reported as a safe procedure, it is not adopted in all centres. As it is not possible to accurately distinguish benign tumours using imaging, this may lead to overtreatment. Most of the cancer detected on imaging is treated by surgery, radiofrequency ablation (RFA), or cryotherapy. (2) Methods: This was a single-centre retrospective study of 225 patients studied preoperatively with ultrasound (US)/CT-guided renal biopsy, with the aim of supporting clinical management. Decisions regarding the biopsy were based on either MDT indication or physician preference. US-guided renal biopsy was the first option for all patients; CT-guided biopsy was used when US-guided biopsy was not feasible. The efficacy of renal biopsy in terms of diagnostic performance and the concordance between biopsy results and definitive pathology were investigated. Additionally, adverse events related to the biopsy were recorded and analysed. Data collected throughout the study were analysed using binary logistic regression, Fisher's exact test, and Pearson's chi-square test to investigate possible correlations between post-procedural complications and the size of the lesion. (3) Results: Renal biopsy was not diagnostic in 23/225 (10.2%) patients. A CT-guided approach was necessary in 20/225 patients after failure of US-guided biopsy. The complication rate of renal biopsy was 4.8% overall-all Clavien grade I and without any serious sequelae. Interestingly, complications occurred in patients with very different sizes of renal cell carcinoma. No correlation between complications and anticoagulant/antiplatelet drugs was found. No seeding was reported among the patients who underwent partial/radical nephrectomy. (4) Conclusions: Renal biopsy was shown to be safe and effective, with a high concordance between biopsy results and definitive pathology and a low rate of complications. The use of a CT-guided approach whenever the US-guided approach failed improved the diagnostic performance of renal biopsy.

Meta-analysis of cardiac imaging modalities for the diagnosis of left ventricular thrombus following acute myocardial infarction

Abstract Background The formation of left ventricular thrombus (LVT) in post-acute myocardial infarction (AMI) patient carries significant risk of systemic embolism. Hence, there is a need to adopt strategies aimed at accurate and early diagnosis of LVT in order to mitigate the risk of embolic events with anti-thrombotic therapy. We aimed to conduct a systematic review and meta-analysis to appraise the current literature on the various diagnostic modalities used in post-AMI LVT detection. Methods We conducted a comprehensive search in Medline, EMBASE, and Scopus using an appropriate keyword/MeSH term search strategy up to 31 December 2021. All studies reporting a comparison between at least two imaging modalities in the diagnosis of LVT in post-AMI patients were included. Results A total of 2039 patient from 11 studies published between 2006 and 2021 were studied. Across all study populations, a majority of patients were male (82.3%) and the overall incidence of LVT was 15.2%. Using delayed-enhancement cardiac magnetic resonance imaging (DE-CMR) as the gold standard reference, transthoracic echocardiogram (TTE) with ultrasound enhancing agent (UEA) had the highest sensitivity (0.886, 95% CI 0.667-0.967) and specificity (0.980, 95% CI 0.900-0.996), while cine-CMR had the best diagnostic performance in terms of positive likelihood ratio (LR+ 437, 95% CI 37.2-1890), negative likelihood ratio (LR- 0.246, 95% CI 0.118-0.423) and diagnostic odds ratio (DOR 2050, 95% CI 134-9590). The summary receiver operating characteristic curves (SROCs) of the three studied imaging modalities were plotted and the AUC values were 0.920, 0.974 and 0.989 for TTE without and with UEA, and cine-CMR. Conclusions The diagnostic modalities used for LVT detection examined in this study had good specificity but poor sensitivity when compared to DE-CMR. Post-AMI patients at higher risk of LVT formation may benefit from further imaging studies with DE-CMR if initial TTE were negative.

Detection and localization of caries and hypomineralization on dental photographs with a vision transformer model

Caries and molar-incisor hypomineralization (MIH) are among the most prevalent diseases worldwide and need to be reliably diagnosed. The use of dental photographs and artificial intelligence (AI) methods may potentially contribute to realizing accurate and automated diagnostic visual examinations in the future. Therefore, the present study aimed to develop an AI-based algorithm that can detect, classify and localize caries and MIH. This study included an image set of 18,179 anonymous photographs. Pixelwise image labeling was achieved by trained and calibrated annotators using the Computer Vision Annotation Tool (CVAT). All annotations were made according to standard methods and were independently checked by an experienced dentist. The entire image set was divided into training (N = 16,679), validation (N = 500) and test sets (N = 1000). The AI-based algorithm was trained and finetuned over 250 epochs by using image augmentation and adapting a vision transformer network (SegFormer-B5). Statistics included the determination of the intersection over union (IoU), average precision (AP) and accuracy (ACC). The overall diagnostic performance in terms of IoU, AP and ACC were 0.959, 0.977 and 0.978 for the finetuned model, respectively. The corresponding data for the most relevant caries classes of non-cavitations (0.630, 0.813 and 0.990) and dentin cavities (0.692, 0.830, and 0.997) were found to be high. MIH-related demarcated opacity (0.672, 0.827, and 0.993) and atypical restoration (0.829, 0.902, and 0.999) showed similar results. Here, we report that the model achieves excellent precision for pixelwise detection and localization of caries and MIH. Nevertheless, the model needs to be further improved and externally validated.

Parenchymal microcalcifications in the thyroid gland: Clinical significance and management strategy.

This study is firstly, to investigate the presence of microcalcification among the patients who underwent thyroid ultrasound and biopsy and to evaluate the incidence of intrathyroid lymphatic spread and cervical lymph node metastasis of thyroid cancer with thyroid microcalcifications. Also, we compared the diagnostic performance between fine needle aspiration (FNA) and core needle biopsy (CNB) for assessing parenchymal microcalcifications in the thyroid gland. We retrospectively assessed total 66 patients with thyroid microcalcifications on ultrasound. The histopathologic characteristics of the surgical specimens considered as the gold standard for diagnosing malignancy. Patients with surgically proven malignancy were evaluated for multifocality, intrathyroid lymphatic spread in the opposite lobe, or cervical lymph node metastasis. Among the 66 confirmed patients, 53 patients had malignant lesions (80.3%) and 13 patients had benign lesions (19.7%). The pathologic results of the 44 patients who underwent total thyroidectomy. Among them, 33 patients (75%) showed multifocality, 30 patients (68.2%) showed intrathyroid lymphatic tumor spread. CNB was performed on 41 patients, and FNA was performed on 54 patients. Both CNB and FNA were performed on 29 patients. There were no statistical differences in terms of diagnostic performance between CNB and FNA. Thyroid microcalcifications demonstrate a high prevalence of malignancy. Both CNB and FNA demonstrate similar diagnostic accuracies.

Predicting response to chemoradiotherapy in rectal cancer via visual morphologicassessmentand staging on baseline MRI: a multicenter and multireader study.

Pre-treatment knowledge of the anticipated response of rectal tumors to neoadjuvant chemoradiotherapy (CRT) could help to further optimize the treatment. Van Griethuysen et al. proposed a visual 5-point confidence score to predict the likelihood of response on baseline MRI. Aim wasto evaluatethis scorein a multicenter and multireader study setting and compare it to two simplified (4-point and 2-point) adaptations in terms of diagnostic performance, interobserver agreement (IOA), and reader preference. Twenty-two radiologists from 14 countries (5 MRI-experts,17 general/abdominal radiologists) retrospectively reviewed 90 baseline MRIs to estimate if patients would likely achieve a (near-)complete response (nCR); first using the 5-point score by van Griethuysen (1=highly unlikely to 5=highly likely to achieve nCR), second using a 4-point adaptation (with 1-point each for high-risk T-stage, obvious mesorectal fascia invasion, nodal involvement, and extramural vascular invasion), and third using a 2-point score (unlikely/likely to achieve nCR). Diagnostic performance was calculated using ROC curves and IOA using Krippendorf's alpha (α). Areas under the ROC curve to predict the likelihood of a nCR were similar for the three methods (0.71-0.74). IOA was higher for the 5- and 4-point scores (α=0.55 and 0.57 versus 0.46 for the 2-point score) with best results for the MRI-experts (α=0.64-0.65). Most readers (55%) favored the 4-point score. Visual morphologic assessment and stagingmethods can predictneoadjuvant treatment response with moderate-good performance. Compared to a previously published confidence-based scoring system, study readers preferred a simplified 4-point risk score based on high-risk T-stage, MRF involvement, nodal involvement, and EMVI.

Role of tomosynthesis and ultrasound in the assessment of asymmetric breast densities: a comparative prospective study

BackgroundTomosynthesis is a recent advancement of full-field digital mammography involves transforming two-dimensional (2D) breast images into three dimensions (3D) images. It reduces the adverse effect of tissue superimposition on conventional 2D- mammography, therefore having high potential enhancing identification and assessment of asymmetric breast densities. The aim of the study was to assess and compare the diagnostic performance of breast ultrasound and 3D digital breast tomosynthesis in the assessment of asymmetric breast densities.ResultsIn the current study, 80 patients with 80 mammographically and/or tomosynthesized breast asymmetries were included. The patients' ages ranged from 30 to 70 years old, with a mean age of 47.2 ± 9.2 SD. Breast ultrasound outperformed digital breast tomosynthesis in terms of diagnostic performance. Tomosynthesis had a sensitivity of 86.4%, specificity of 93.1%, positive predictive value of 82.6%, negative predictive value of 94.7%, and accuracy of 91.3% compared to ultrasounds' sensitivity of 100.00%, specificity of 93.1%, positive predictive value of 84.6%, negative predictive value of 100.00%, and accuracy of 95%.ConclusionsIncorporating ultrasonography in the assessment of asymmetric breast densities outperformed tomosynthesis and shown to be more precise in characterisation of lesions underlying asymmetric breast density.

Uterine mesenchymal tumors: development and preliminary results of a magnetic resonance imaging (MRI) diagnostic algorithm.

The aim of our study is to propose a diagnostic algorithm to guide MRI findings interpretation and malignancy risk stratification of uterine mesenchymal masses with a multiparametric step-by-step approach. A non-interventional retrospective multicenter study was performed: Preoperative MRI of 54 uterine masses was retrospectively evaluated. Firstly, the performance of MRI with monoparametric and multiparametric approach was assessed. Reference standard for final diagnosis was surgical pathologic result (n = 53 patients) or at least 1-year MR imaging follow-up (n = 1 patient). Subsequently, a diagnostic algorithm was developed for MR interpretation, resulting in a Likert score from 1 to 5 predicting risk of malignancy of the uterine lesion. The accuracy and reproducibility of the MRI scoring system were then tested: 26 preoperative pelvic MRI were double-blind evaluated by a senior (SR) and junior radiologist (JR). Diagnostic performances and the agreement between the two readers with and without the application of the proposed algorithm were compared, using histological results as standard reference. Multiparametric approach showed the best diagnostic performance in terms of accuracy (94.44%,) and specificity (97.56%). DWI was confirmed as the most sensible parameter with a relative high specificity: low ADC values (mean 0.66) significantly correlated to uterine sarcomas diagnosis (p < 0.01). Proposed algorithm allowed to improve both JR and SR performance (algorithm-aided accuracy 88.46% and 96%, respectively) and determined a significant increase in inter-observer agreement, helping even the less-experienced radiologist in this difficult differential diagnosis. Uterine leiomyomas and sarcomas often show an overlap of clinical and imaging features. The application of a diagnostic algorithm can help radiologists to standardize their approach to a complex myometrial mass and to easily identify suspicious MRI features favoring malignancy.

Gearbox Fault Diagnosis Based on Refined Time-Shift Multiscale Reverse Dispersion Entropy and Optimised Support Vector Machine

The fault diagnosis of a gearbox is crucial to ensure its safe operation. Entropy has become a common tool for measuring the complexity of time series. However, entropy bias may occur when the data are not long enough or the scale becomes larger. This paper proposes a gearbox fault diagnosis method based on Refined Time-Shifted Multiscale Reverse Dispersion Entropy (RTSMRDE), t-distributed Stochastic Neighbour Embedding (t-SNE), and the Sparrow Search Algorithm Support Vector Machine (SSA-SVM). First, the proposed RTSMRDE was used to calculate the multiscale fault features. By incorporating the refined time-shift method into Multiscale Reverse Dispersion Entropy (MRDE), errors that arose during the processing of complex time series could be effectively reduced. Second, the t-SNE algorithm was utilized to extract sensitive features from the multiscale, high-dimensional fault features. Finally, the low-dimensional feature matrix was input into SSA-SVM for fault diagnosis. Two gearbox experiments showed that the diagnostic model proposed in this paper had an accuracy rate of 100%, and the proposed model performed better than other methods in terms of diagnostic performance.

Accuracy of a Standalone Atrial Fibrillation Detection Algorithm Added to a Popular Wristband and Smartwatch: Prospective Diagnostic Accuracy Study

Silent paroxysmal atrial fibrillation (AF) may be difficult to diagnose, and AF burden is hard to establish. In contrast to conventional diagnostic devices, photoplethysmography (PPG)-driven smartwatches or wristbands allow for long-term continuous heart rhythm assessment. However, most smartwatches lack an integrated PPG-AF algorithm. Adding a standalone PPG-AF algorithm to these wrist devices might open new possibilities for AF screening and burden assessment. The aim of this study was to assess the accuracy of a well-known standalone PPG-AF detection algorithm added to a popular wristband and smartwatch, with regard to discriminating AF and sinus rhythm, in a group of patients with AF before and after cardioversion (CV). Consecutive consenting patients with AF admitted for CV in a large academic hospital in Amsterdam, the Netherlands, were asked to wear a Biostrap wristband or Fitbit Ionic smartwatch with Fibricheck algorithm add-on surrounding the procedure. A set of 1-min PPG measurements and 12-lead reference electrocardiograms was obtained before and after CV. Rhythm assessment by the PPG device-software combination was compared with the 12-lead electrocardiogram. A total of 78 patients were included in the Biostrap-Fibricheck cohort (156 measurement sets) and 73 patients in the Fitbit-Fibricheck cohort (143 measurement sets). Of the measurement sets, 19/156 (12%) and 7/143 (5%), respectively, were not classifiable by the PPG algorithm due to bad quality. The diagnostic performance in terms of sensitivity, specificity, positive predictive value, negative predictive value, and accuracy was 98%, 96%, 96%, 99%, 97%, and 97%, 100%, 100%, 97%, and 99%, respectively, at an AF prevalence of ~50%. This study demonstrates that the addition of a well-known standalone PPG-AF detection algorithm to a popular PPG smartwatch and wristband without integrated algorithm yields a high accuracy for the detection of AF, with an acceptable unclassifiable rate, in a semicontrolled environment.

Integration of Artificial Intelligence Decision Aids to Reduce Workload and Enhance Efficiency in Thyroid Nodule Management

To optimize the integration of artificial intelligence (AI) decision aids and reduce workload in thyroid nodule management, it is critical to incorporate personalized AI into the decision-making processes of radiologists with varying levels of expertise. To develop an optimized integration of AI decision aids for reducing radiologists' workload while maintaining diagnostic performance compared with traditional AI-assisted strategy. In this diagnostic study, a retrospective set of 1754 ultrasonographic images of 1048 patients with 1754 thyroid nodules from July 1, 2018, to July 31, 2019, was used to build an optimized strategy based on how 16 junior and senior radiologists incorporated AI-assisted diagnosis results with different image features. In the prospective set of this diagnostic study, 300 ultrasonographic images of 268 patients with 300 thyroid nodules from May 1 to December 31, 2021, were used to compare the optimized strategy with the traditional all-AI strategy in terms of diagnostic performance and workload reduction. Data analyses were completed in September 2022. The retrospective set of images was used to develop an optimized integration of AI decision aids for junior and senior radiologists based on the selection of AI-assisted significant or nonsignificant features. In the prospective set of images, the diagnostic performance, time-based cost, and assisted diagnosis were compared between the optimized strategy and the traditional all-AI strategy. The retrospective set included 1754 ultrasonographic images from 1048 patients (mean [SD] age, 42.1 [13.2] years; 749 women [71.5%]) with 1754 thyroid nodules (mean [SD] size, 16.4 [10.6] mm); 748 nodules (42.6%) were benign, and 1006 (57.4%) were malignant. The prospective set included 300 ultrasonographic images from 268 patients (mean [SD] age, 41.7 [14.1] years; 194 women [72.4%]) with 300 thyroid nodules (mean [SD] size, 17.2 [6.8] mm); 125 nodules (41.7%) were benign, and 175 (58.3%) were malignant. For junior radiologists, the ultrasonographic features that were not improved by AI assistance included cystic or almost completely cystic nodules, anechoic nodules, spongiform nodules, and nodules smaller than 5 mm, whereas for senior radiologists the features that were not improved by AI assistance were cystic or almost completely cystic nodules, anechoic nodules, spongiform nodules, very hypoechoic nodules, nodules taller than wide, lobulated or irregular nodules, and extrathyroidal extension. Compared with the traditional all-AI strategy, the optimized strategy was associated with increased mean task completion times for junior radiologists (reader 11, from 15.2 seconds [95% CI, 13.2-17.2 seconds] to 19.4 seconds [95% CI, 15.6-23.3 seconds]; reader 12, from 12.7 seconds [95% CI, 11.4-13.9 seconds] to 15.6 seconds [95% CI, 13.6-17.7 seconds]), but shorter times for senior radiologists (reader 14, from 19.4 seconds [95% CI, 18.1-20.7 seconds] to 16.8 seconds [95% CI, 15.3-18.3 seconds]; reader 16, from 12.5 seconds [95% CI, 12.1-12.9 seconds] to 10.0 seconds [95% CI, 9.5-10.5 seconds]). There was no significant difference in sensitivity (range, 91%-100%) or specificity (range, 94%-98%) between the 2 strategies for readers 11 to 16. This diagnostic study suggests that an optimized AI strategy in thyroid nodule management may reduce diagnostic time-based costs without sacrificing diagnostic accuracy for senior radiologists, while the traditional all-AI strategy may still be more beneficial for junior radiologists.

Association between swallowing muscle mass and dysphagia in older adults: A case-control study.

Swallowing function is formed by the synergistic, coordinated, and complex movements of approximately 30 muscles. However, There is no study on which one(s) of these muscles can predict the possibility of dyspfagia. The purpose of this study was to investigate whether there is a relation between dysphagia and ultrasonographic measurements of swallowing muscles mass, and to determine which muscle mass would predict the possibility of dysphagia. This prospective and controlled study was performed on 87 older people (aged >65 years). Clinical (Eating assessment tool-10 and The Gugging Swallowing Screen test), endoscopic, electrophysiological (dysphagia limit and swallowing intervals), and ultrasonographic (masseter [MM], genioglossus [GGM], geniohyoid [GHM], myohyoid [MH] and anterior digastric muscles [ADM]) evaluations were performed on all patients. The relationship between ultrasound measurements and demographic/swallowing characteristics and presence of dysphagia were investigated. The muscle mass measurement which was associated with dysphagia was contracted masseter muscle. In terms of diagnostic performance, the sensitivity was the highest in GHM (92.3%) and the specificity was the highest in contracted MM thickness (90.3%). The probability of dysphagia increases from 54% to 80-89% with a decrease in contracted MM thickness, and the probability of dysphagia increased from 48% to 72% with a decrease in contracted GGM thickness. In addition, the highest AUC level was detected in contracted MM and contracted GGM muscles. Moreover, it was determined that the risk of dysphagia increased from 9.6 to 14.1 times with one unit decrease in muscle thickness in contracted MM muscles, and the risk of dysphagia increased 9.1 times with one unit decrease in contracted GGM muscles. Decreased swallowing muscle mass in older adults are independent risk factors for dysphagia. Especially the decrease in the thickness of the contracted MM increases the possibility of dysphagia at high rate.

Evaluation of an automated rapid urine culture method for urinary tract infection: Comparison with gold standard conventional culture method.

Urinary tract infection is one of the most prevalent disease affecting people from all age groups. For its diagnosis, conventional culture and antibiotic susceptibility is the gold standard. However, its major limitation is that the results take minimum of 24​h and antibiotic susceptibility is available after 48​h. Automated culture methods having comparable sensitivity and specificity as compared to conventional culture should be evaluated for routine diagnostics. With this aim we evaluated the diagnostic accuracy of automated urine culture method HB&L uroquattro by comparing with the gold standard conventional culture method. A total of 1220 urine samples were included in the study. Semi-quantitative urine culture was performed using standard methods on cysteine lactose electrolyte deficient medium. For the automated culture, HB&L Uroquattro (Alifax, Polverara, PD, Italy), standard guidelines given in the manual of the instrument were followed. Diagnostic performance in terms of sensitivity, specificity, positive and negative likelihood ratios, positive and negative predictive values were calculated. Based on the final interpretation of conventional culture for the total 1220 samples, 26 samples (2.1%) showed major non-concordance as they were identified as sterile by HB&L but had significant growth by conventional culture and 19.9% showed minor non-concordance. At 100-999 colony forming unit/ml, HB&L has high negative predictive value i.e. 96.6% with 95% CI (95.2%-97.6%) and sensitivity i.e. 92.66% with 95% CI of (89.42%-95.15%). The HB&L Uroquattro seems to be a reliable instrument to obtain urine microbiological results in a timely manner. This technique can give presumptive report to the clinician within 5​h only for initiation of empirical antibiotics in cases of positive results.

Diagnostic performance of cryobiopsy guided by radial-probe EBUS with a guide sheath for peripheral pulmonary lesions.

Transbronchial lung cryobiopsy (TBCB) has developed rapidly and has become one of the research hotspots of lung biopsy technology. The present study sought to evaluate the efficacy of TBCB guided by radial-probe EBUS (RP-EBUS) and a guide sheath (GS) without fluoroscopy for peripheral pulmonary lesions. In this retrospective study, McNemar's test was used in order to compare TBCB and transbronchial forceps biopsy (TBFB) in terms of diagnostic performance. A multivariate logistic regression model was designed to explore the association between predictive variables and the diagnostic yield of TBCB. A total of 168 patients underwent GS-guided RP-EBUS. Of those, 157 had lesions that were visible and 11 had lesions that were not. Of those 157 patients, 24 were excluded because of missing data or an unclear final diagnosis. Therefore, 133 patients underwent RP-EBUS-GS-guided TBFB and TBCB. The pooled diagnostic yield of RP-EBUS-GS-guided TBCB without fluoroscopy was 71.5% (103/144). In 133 patients, the diagnostic yield of TBCB was significantly higher than that of TBFB (77.4% vs. 59.4%; p < 0.05). Multivariate analysis indicated that lesion size and site were independently associated with the diagnostic yield of TBCB (OR = 2.8, p = 0.03 and OR = 4.1, p = 0.01, respectively), although cryoprobe size was not. There was no significant difference between the 1.1-mm cryoprobe and the 1.9-mm cryoprobe in terms of diagnostic performance (78.4% vs. 76.8%; p > 0.05). GS-guided RP-EBUS is regarded as a practical option for guiding cryobiopsy, although it may not be able to replace fluoroscopy. Peripheral pulmonary lesions not located in the upper lobes or larger than 30 mm are significantly associated with a higher diagnostic yield of cryobiopsy.

Performance of the revised 2017 McDonald criteria

A rapid and reliable diagnosis of multiple sclerosis (MS) is crucial to initiate adapted disease-modifying treatment. The 2017 McDonald criteria were revised with the aim of further improving the diagnostic performance. In this article the published studies comparing the use of the 2017 and 2010 McDonald criteria were reviewed and analyzed in terms of diagnostic performance. A total of 20studies and 1review article with atotal of 3006 evaluated patients were identified by means of aliterature search in the PubMed database (search term: McDonald criteria 2010 and McDonald criteria 2017). Using the 2017 McDonald criteria, adiagnosis of MS was made in more patients (2277/3006patients, 76%) and in an earlier stage (3-10months) compared with the 2010 revision (1562/3006patients, 52%). Of the additional MS diagnoses, 193/715 were due to the adjustment of the imaging criteria of temporal dissemination and 536/715 were due to the introduction of oligoclonal bands as adiagnostic criterion. The revised McDonald criteria of 2017 have achieved their goal and enable the diagnosis of MS in ahigher proportion of patients at the first clinical event.

Thy-Wise: An interpretable machine learning model for the evaluation of thyroid nodules.

Current risk stratification systems for thyroid nodules suffer from low specificity and high biopsy rates. Recently, machine learning (ML) is introduced to assist thyroid nodule diagnosis but lacks interpretability. Here, we developed and validated ML models on 3965 thyroid nodules, as compared to the American College of Radiology Thyroid Imaging, Reporting and Data System (ACR TI-RADS). Subsequently, a SHapley Additive exPlanation (SHAP) algorithm was leveraged to interpret the results of the best-performing ML model. Clinical characteristics including thyroid-function tests were collected from medical records. Five ACR TI-RADS ultrasonography (US) categories plus nodule size were assessed by experienced radiologists. Random forest (RF), support vector machine (SVM) and extreme gradient boosting (XGBoost) were used to build US-only and US-clinical ML models. The ML models and ACR TI-RADS were compared in terms of diagnostic performance and unnecessary biopsy rate. Among the ML models, the US-only RF model (hereafter, Thy-Wise) achieved the optimal performance. Compared to ACR TI-RADS, Thy-Wise showed higher accuracy (82.4% vs 74.8% for the internal validation; 82.1% vs 73.4% for external validation) and specificity (78.7% vs 68.3% for internal validation; 78.5% vs 66.9% for external validation) while maintaining sensitivity (91.7% vs 91.2% for internal validation; 91.9% vs 91.1% for external validation), as well as reduced unnecessary biopsies (15.3% vs 32.3% for internal validation; 15.7% vs 47.3% for external validation). The SHAP-based interpretation of Thy-Wise enables clinicians to better understand the reasoning behind the diagnosis, which may facilitate the clinical translation of this model.

Rolling bearing fault diagnosis method based on multi-sensor two-stage fusion

Despite the great achievements of deep learning methods based on a single sensor in fault diagnosis, learning useful information from multi-sensor data is still a challenge. In order to make full use of multi-sensor information and improve the performance of rolling bearing fault diagnosis, a novel multi-sensor information fusion framework is proposed in this paper. First, a multi-sensor-based multi-frequency information fusion method is proposed. The multi-frequency information of each sensor is segmented first to enhance the datasets, and then a weighted fusion rule based on fuzzy entropy is constructed to fuse the information of different frequency components for multi-sensors. Second, a multi-kernel attention convolutional neural network is designed to realize multi-frequency feature capture, fusion, and fault classification of multi-sensors. Finally, two different rolling bearing datasets are used to implement fault diagnosis experiments. Experimental results show that the proposed method outperforms the comparative methods in terms of diagnostic performance and robustness.

The diagnostic utility of IL-10, IL-17, and PCT in patients with sepsis infection.

ObjectiveThe purpose of this study is to determine the diagnostic value and net clinical benefit of interleukin-10 (IL-10), interleukin-17 (IL-17), procalcitonin (PCT), and combination tests in patients with sepsis, which will serve as a standard for sepsis early detection.Patients and methodsAn investigation of 84 sepsis patients and 81 patients with local inflammatory diseases admitted to the ICU of Tongji University Hospital in 2021. In addition to comparing inter-group variability, indicators relevant to sepsis diagnosis and therapy were screened.ResultsLASSO regression was used to examine PCT, WBC, CRP, IL-10, IFN-, IL-12, and IL-17. Multivariate logistic regression linked IL-10, IL-17, and PCT to sepsis risk. The AUC values of IL-10, IL-17, PCT, and the combination of the three tests were much higher than those of standard laboratory infection indicators. The combined AUC was greater than the sum of IL-10, IL-17, and PCT (P < 0.05). A clinical decision curve analysis of IL-10, IL-17, PCT, and the three combined tests found that the three combined tests outperformed the individual tests in terms of total clinical benefit rate. To predict the risk of sepsis using IL-10, IL-17, and PCT had an AUC of 0.951, and the model's predicted probability was well matched. An examination of the nomogram model's clinical value demonstrated a considerable net therapeutic benefit between 3 and 87%.ConclusionThe IL-10, IL-17, and PCT tests all have a high diagnostic value for patients with sepsis, and the combination of the three tests outperforms the individual tests in terms of diagnostic performance, while the combined tests have a higher overall clinical benefit rate.

A rolling bearing status monitoring method based on subband spectral fuzzy description

Vibration signals provided by rotating machinery are informative signals about their operating states. By nature, the vibration signal behavior is non-stationary. To this end, the extraction of discriminating and fault-sensitive parameters is a major challenge in the field of monitoring rotating machines. Conventional fault diagnosis methods based on signal processing use statistical feature parameters in time domain, frequency domain and time-frequency representation. In this article, a new method is proposed for the detection and classification of bearing defects based on spectral subband using frequency membership functions. Statistical parameters including subband energy, Center frequency, root variance frequency and Shannon entropy are considered. Compared to the common features, the extracted parameters can provide discriminating information. These feature parameters are finally fed into a generalized RBF neural network system trained with the Resilient Backpropagation (Rprop) algorithm to classify seven pre-established fault types in ball bearings operating under multiple shaft speeds and load conditions. The results suggest that the proposed system can significantly improve the diagnostic performance in terms of accuracy and estimation of the bearing fault level.

The role of laboratory parameters in predicting severity of COVID-19 disease in pregnant patients

We aimed to examine the relationship between laboratory markers and the severity of the disease in pregnant women diagnosed with coronavirus disease 2019 (COVID-19). Clinical records were retrospectively reviewed for 112 pregnant women. Patients diagnosed with COVID-19 were divided into two groups as mild/moderate and severe. The relationship between predicting the severity of the disease and laboratory parameters was investigated. Neutrophil lymphocyte ratio, C-reactive protein (CRP), ferritin and aspartate aminotransferase levels were significantly higher in severe COVID-19 cases than mild/moderate cases (p = .048, p = .003, p = .015 and p = .035, respectively). CRP was found to be the most useful marker in terms of diagnostic performance with a cut off value of 10.8 (sensitivity 80%, specificity 56.1%, NPV 88.5% and PPV 40.0%). The best diagnostic performance was obtained using CRP and ferritin combined with cut-offs of 10.8 mg/L for CRP and 26.5 µg/L for ferritin. Combined CRP and ferritin showed sensitivity, specificity, negative predictive value and positive predictive value of 94.7%, 52.8%, 96.6% and 41.9%, respectively, in predicting severe COVID-19. The combination of CRP and ferritin parameters may be useful in estimating the severity of the disease in pregnant patients who were initially diagnosed with COVID-19. Impact Statement What is already known about this subject? Coronavirus disease 2019 (COVID-19) can rapidly develop into acute respiratory distress syndrome (ARDS) and result in serious complications in some pregnant patients. Therefore, timely diagnosis of patients is crucial. Most previous reports of COVID-19 laboratory results are based on data from the general population and limited information is available regarding pregnancy status. Although laboratory medicine makes an important contribution to clinical decision making in many infectious diseases, including COVID-19, studies to predict the severity of the disease with laboratory markers are limited and the results are contradictory. What do the results of this study add? Our study shows that C-reactive protein (CRP), neutrophil lymphocyte ratio (NLR), ferritin and aspartate aminotransferase (AST) are associated with severe disease in pregnant women diagnosed with COVID-19. In addition, the use of combined CRP and ferritin appears to have higher sensitivity and negative predictive value than using other tests alone. Furthermore, this study shows that coagulation markers are not useful in predicting disease severity in pregnancy. What are the implications of these findings for clinical practice and/or further research? Predicting the severity of COVID-19 disease in pregnancy can prevent unnecessary hospitalisations and allow the implementation of the necessary clinical approach. Further studies can focus on the clinical usefulness of these parameters in predicting severe COVID-19 in pregnancy.