Test Accuracy Research Articles

Predicting abatacept retention using machine learning

BackgroundThe incorporation of machine learning is becoming more prevalent in the clinical setting. By predicting clinical outcomes, machine learning can provide clinicians with a valuable tool for refining precision medicine approaches and improving treatment outcomes.MethodsThis was a post hoc analysis of pooled patient-level data from the global, real-world ACTION and ASCORE trials in patients with rheumatoid arthritis (RA) initiating abatacept. Patient demographic and disease characteristics were input across 10 machine learning models used to predict 12-month treatment retention. Retention was defined as treatment for > 365 days or ≤ 365 days in patients who achieved remission or major clinical response (based on European Alliance of Associations for Rheumatology response criteria). The pooled dataset was split into a training/validation cohort for model development and a test cohort for an unbiased evaluation of performance. SHapley Additive exPlanation (SHAP) values determined the level of importance and directionality for key patient features predicting abatacept retention.ResultsThe pooled ACTION and ASCORE dataset included 5320 patients with RA (mean [standard deviation] age 57.7 [12.7] years; 79% female). The 12-month abatacept retention rate was 61% (n = 3236) with a discontinuation rate of 39% (n = 2037). In the training set (n = 4218), the gradient-boosting classifier model demonstrated the best performance (testing accuracy: 62%). This model had an area under the receiver operating characteristic curve (95% confidence interval) of 0.620 (0.586, 0.653) and F1 score of 0.659 (0.625, 0.689) in the test set of patients (n = 1055). Using this model, the five most important variables predicting 12-month abatacept retention were low body mass index (BMI), low American College of Rheumatology functional status class, anti-citrullinated protein antibody (ACPA) positivity, low Patient Global Assessment, and younger age.ConclusionsThe gradient-boosting classifier model identified key patient features predictive of abatacept retention from this large, real-world study population. The SHAP values conveyed the directionality and importance of BMI, functional status, ACPA serostatus, Patient Global Assessment, and age for abatacept retention. Findings are consistent with previous observations and help validate the machine learning approach for predictive modelling in RA treatment, and may help inform clinical decision making.Trial registrationNCT02109666 (ACTION), NCT02090556 (ASCORE).

Intraoperative frozen section analysis for detection of fracture-related infection in nonunion of the upper limb. Diagnostic accuracy study.

During revision surgery for the management of patients presenting with long-bone upper extremity nonunion, it is crucial to rule out fracture-related infection (FRI). This is especially true if there are clinical signs suggestive of FRI, or if there is a history of prior FRI, open wound fracture, or surgery. This study aimed to determine the efficacy of frozen section analysis (FSA) in providing real-time diagnosis of FRI in patients with upper-limb long-bone nonunion undergoing revision surgery. We included previously surgically treated patients who underwent revision surgery for chronic/late-onset nonunion of the long bones of the upper limbs over a 9-year period. We retrospectively applied the FRI criteria to identify eligible patients and recorded their clinical features and demographic characteristics. In all included patients, samples for microbiological analysis and FSA were collected simultaneously during revision surgery. Patients were categorized according to intraoperative culture results as having aseptic or infected nonunion. The sensitivity, specificity, positive and negative predictive values, and accuracy of FSA test were calculated and compared using microbiological analysis as the reference standard test. The concordance rate between FSA and definitive histopathology was also determined. Sixty-two patients who were surgically treated for nonunion of the arm or forearm were included in this study. Septic nonunion was diagnosed intraoperatively in 9 patients based on FSA findings, while microbiological analysis confirmed septic nonunion in 8 patients. The sensitivity and specificity of FSA were 88 % (95 % confidence interval [CI] [47, 100]) and 96 % (95 % CI [87, 100]), respectively. Positive and negative predictive values were 78 % (95 % CI [47, 93]) and 98 % (95 % CI [89, 100]), respectively, with an overall test accuracy of 95 % (95 % CI [87, 99]). Cohen's Kappa coefficient between FSA and definitive histopathology was 0.74 (95 % CI [0.5, 1]). FSA showed high sensitivity and specificity for the detection of active infection during revision surgery for nonunion of the upper limb. Owing to its high negative predictive value, it can reliably rule out active infections during revision surgery. Since the results are obtained immediately during revision surgery, this approach offers the significant advantage of enabling real-time decision-making.

Current trends in blood biomarkers detection and neuroimaging for Parkinson's disease.

Parkinson's disease (PD) is a neurodegenerative disorder characterized by both motor and cognitive impairments. A significant challenge in managing PD is the variability of symptoms and disease progression rates. This variability is primarily attributed to unclear biomarkers associated with the disease and the lack of early diagnostic technologies and effective imaging methods. PD-specific biomarkers are essential for developing practical tools that facilitate accurate diagnosis, patient stratification, and monitoring of disease progression. Hence, creating valuable tools for detecting and diagnosing PD based on specific biomarkers is imperative. Blood testing, less invasive than obtaining cerebrospinal fluid through a lumbar puncture, is an ideal source for these biomarkers. Although such biomarkers were previously lacking, recent advancements in various detection techniques related to PD biomarkers and new imaging methods have emerged. However, basic research requires more detailed guidelines on effectively implementing these biomarkers in diagnostic procedures to enhance the diagnostic accuracy of PD blood testing in clinical practice. This review discusses the developmental trends of PD-related blood biomarker detection technologies, including optical analysis platforms. Despite the progress in developing various biomarkers for PD, their specificity and sensitivity remain suboptimal. Therefore, the integration of multimodal biomarkers along with optical and imaging technologies is likely to significantly improve diagnostic accuracy and facilitate the implementation of personalized medicine. This review forms valid research hypotheses for PD research and guides future empirical studies.

AI-led study of dynamic changes in milk containing hybrid nanoparticles in an electromagnetically vibrated channel subjected to thermal oscillations and rapid pressure changes: Implications for dairy industry

Oscillating electromagnetic forces generated from a vibrated Riga plate have broad implications across various scientific and engineering domains. Artificial intelligence (AI) is applied to optimize precision and energy efficiency in pasteurization and sterilization by regulating the thermal and dynamic behavior of nanoparticle-infused milk under electromagnetic heating. The technique ensures accurate temperature control, minimizes the risk of overheating, and preserves the milk's nutritional and sensory qualities. It focuses on predicting the thermal and dynamic behaviors of milk infused with silver and zinc oxide nanoparticles in an electromagnetically vibrated channel experiencing thermal oscillations and rapid pressure changes. The research integrates complex physical phenomena such as radiant heat emission and Darcy drag forces, employing Darcy's model to delve into drag within porous media. Detailed mathematical and physical descriptions of milk flow dynamics are established, with solutions efficiently derived using the Laplace Transform (LT) method. The results, encompassing shear stress (SS) and rate of heat transfer (RHT) analyses, are detailed in tables and graphs. Findings indicate enhanced milk momentum with higher modified Hartmann number and reduced momentum with wider electrode spacing. Elevated oscillation frequencies of the left channel wall stabilize milk flow in both hybrid nano-milk (HNM) and nano-milk (NM). Larger Casson parameter improves SS, while higher radiation parameter reduces RHT. An AI-driven artificial neural network (ANN) is employed for precise estimations, achieving 98.022% accuracy in SS testing, 98.99% in cross-validation, and a flawless 100% accuracy for RHT. The research findings can be implemented to precisely control the mixing of milk and its physical features at a molecular level, enabling more even heat distribution and faster, more efficient pasteurization or homogenization processes.

A novel ViT-BILSTM model for physical activity intensity classification in adults using gravity-based acceleration

AimThe aim of this study is to apply a novel hybrid framework incorporating a Vision Transformer (ViT) and bidirectional long short-term memory (Bi-LSTM) model for classifying physical activity intensity (PAI) in adults using gravity-based acceleration. Additionally, it further investigates how PAI and temporal window (TW) impacts the model’ s accuracy.MethodThis research used the Capture-24 dataset, consisting of raw accelerometer data from 151 participants aged 18 to 91. Gravity-based acceleration was utilised to generate images encoding various PAIs. These images were subsequently analysed using the ViT-BiLSTM model, with results presented in confusion matrices and compared with baseline models. The model’s robustness was evaluated through temporal stability testing and examination of accuracy and loss curves.ResultThe ViT-BiLSTM model excelled in PAI classification task, achieving an overall accuracy of 98.5% ± 1.48% across five TWs-98.7% for 1s, 98.1% for 5s, 98.2% for 10s, 99% for 15s, and 98.65% for 30s of TW. The model consistently exhibited superior accuracy in predicting sedentary (98.9% ± 1%) compared to light physical activity (98.2% ± 2%) and moderate-to-vigorous physical activity (98.2% ± 3%). ANOVA showed no significant accuracy variation across PAIs (F = 2.18, p = 0.13) and TW (F = 0.52, p = 0.72). Accuracy and loss curves show the model consistently improves its performance across epochs, demonstrating its excellent robustness.ConclusionThis study demonstrates the ViT-BiLSTM model’s efficacy in classifying PAI using gravity-based acceleration, with performance remaining consistent across diverse TWs and intensities. However, PAI and TW could result in slight variations in the model’s performance. Future research should concern and investigate the impact of gravity-based acceleration on PAI thresholds, which may influence model’s robustness and reliability.

Detecting mild cognitive impairment by applying integrated random forest to finger tapping.

Early detection of dementia is essential to reduce the decline in quality of life (QoL) and the increase in medical and nursing care costs associated with dementia in an aging society. In this study, we aimed to develop a simple screening test for mild cognitive impairment (MCI), a preliminary stage of dementia, by creating an analytical method to accurately detect MCI through finger-tapping measurement. We extracted 248 characteristics from the finger-tapping waveforms of 182 MCI patients and 352 normal controls, applying five conventional classification methods along with an improved Random Forest (RF) method proposed in this study (Integrated RF). In the proposed method, the RF classification model for the MCI and normal control groups is supplementally integrated with the RF classification model for the Alzheimer's disease and normal control groups to generate a new classification model. When comparing the discrimination accuracy of each method, the proposed method achieved the highest accuracy, with an F1-score of 0.795 (recall = 0.778 and precision = 0.814). These results demonstrate the potential of finger-tapping measurement as a highly accurate screening test for MCI.

Evaluating systemic immune-inflammation indices as predictive markers for endometriosis diagnosis: A retrospective observational study.

Endometriosis is a chronic inflammatory disease for which there is currently no accurate screening test to identify or predict the probability of the disease in individuals. This can often lead to delays in diagnosis. Several systemic immune-inflammation indices are currently used as predictive or supportive markers for several inflammation-associated diseases. In this study, we investigated whether these immune-inflammation indices, such as systemic immune inflammation index (SII), systemic inflammation response index (SIRI), neutrophil-to-lymphocyte ratio (NLR), and pan-immune inflammation value (PIV) could serve as predictive or supplementary markers for diagnosing endometriosis. A total of 434 women with confirmed endometriosis and 517 controls were included in this study. Data on peripheral blood tests, including total counts of white cells, neutrophils, monocytes, lymphocytes, and platelets, were collected, and systemic immune-inflammation indices were calculated. SII, SIRI, NLR, and PIV values were significantly higher in women diagnosed with endometriosis compared to controls. Using the cut-off values of 538 for SII, 0.814 for SIRI, 2.03 for NLR, and 210 for PIV, we achieved 76 % sensitivity and 70 % specificity. When the four indices were combined, the area under the ROC curve (AUC) was 0.796 (95 % CI: 0.766, 0.827), with 76 % sensitivity and 70 % specificity. In conclusion, while ultrasonography or MRI remains the gold standard for visualizing the lesions in diagnosing endometriosis, followed by laparoscopy and histologic verification, routine peripheral blood tests in combination with clinical symptoms, could provide additional clinical diagnostic value for endometriosis as a non-invasive and cost-effective test.

Diagnostic Accuracy of Apple Watch Electrocardiogram for Atrial Fibrillation: A Systematic Review and Meta-Analysis.

Electrocardiography (ECG) is the gold standard for the diagnosis of atrial fibrillation (AF). Recently, smartwatches like the Apple Watch have emerged as a promising, user-friendly device for rapid detection and diagnosis of AF, but the reliability and diagnostic accuracy still remain controversial. The purpose of this study was to perform a systematic review and diagnostic test accuracy meta-analysis evaluating the diagnostic performance of the Apple Watch ECG in detecting AF. The literature search was conducted on PubMed, Embase, and Cochrane Library through April 2024 for studies comparing the diagnostic accuracy of Apple Watch to standard 12-lead ECG. Statistical analysis was performed using R Software version 4.4.0 and OpenMeta[Analyst]. Pooled analyses of sensitivity, specificity, and area under the receiver operating characteristic curve were determined along with their 95%CIs. The quality of studies was analyzed using the QUADAS-2 tool. The meta-analysis included 11 studies comprising 4,241 participants. Their mean age was 62.56±3.92years, and 28% of the patients were females. The pooled sensitivity and specificity of the Apple Watch for detecting AF were 94.8% (95%CI: 91.7% to 96.8%; I2=67%) and 95% (95%CI: 88.6% to 97.8%; I2=88%), respectively. The area under the receiver operating characteristic curve was 0.96 (95%CI: 0.92-0.97). The Apple Watch ECG carries high accuracy in detecting atrial fibrillation, providing a convenientdiagnostic option for patients.

Assessment of different U-Net backbones in segmenting colorectal adenocarcinoma from H&E histopathology.

Adenocarcinoma, the most prevalent type of colorectal cancer, makes up roughly 95 % of all cases and is associated with a notably high mortality rate. Owing to the various risk factors which might include personal choices and habits or genetic factors, the risk of developing the cancer for every individual might vary. However, given the statistics, the rate of acquiring the disease is pretty high. Therefore, based on the need for early detection and diagnosis of the disease, there is a pressing demand for an automated system to accurately identify adenocarcinoma in the colorectal region by utilizing the concept of binary segmentation wherein two classes are employed to indicate the presence as well as the absence of the condition. To address this, the project explored several deep learning-based segmentation methods-such as U-Net, Attention U-Net, U-Net with ResNet50 backbone, U-Net with MobileNet-v2 backbone, U-Net with EfficientNetB0 backbone, and U-Net with DenseNet121 backbone-to segment adenocarcinoma regions in histopathological images of the colon and rectum, which are essentially the various U-Net backbones. The performance of each method was then compared to identify the most effective approach, and subsequently, it was found that the U-Net with DenseNet121 backbone and U-Net with ResNet50 backbone performed better than the rest of the models in terms of accuracy with its respective training accuracy scores being 93.81 % and 93.39 % while the testing accuracy scores were 90.21 % and 89.81 %, respectively.

Distributed Backface Strain Sensing of Composite Adhesively Bonded Joints under Mode II Fatigue Loading

When investigating the fracture behaviour of adhesive joints, accurate crack tip localization is under mode II loading is particularly challenging due to the absence of crack opening. The crack front often has a curved shape, so methods that inspect the side of the specimen cannot capture the crack length throughout its width. In this study, distributed backface strain sensing using Optical Backscatter Reflectometry fibre sensors is used to monitor fatigue crack growth in adhesively bonded composite End Notched Flexure Specimens. The obtained backface strain profiles were compared with analytical and Finite Element solutions to understand their relationship with the crack length. Results show that the crack tip position can be found by a local extremum in the first derivative of the strain profile. The proposed monitoring technique was compared with more established techniques for crack length measurement, such as Visual Testing, Compliance Monitoring, and Phased Array Ultrasonic Testing, and was shown to provide good accuracy. The evaluation of these techniques not only plays a crucial role in ensuring accurate testing, but can also contribute to the development of effective strategies for inspection and monitoring of adhesive joints.

Validity and Intrarater Reliability of the FysioMeter—Measuring Eccentric Knee Flexor Force During the Nordic Hamstring Exercise

ABSTRACT Between limb strength, asymmetry is a leading risk factor for hamstring strain re-injury. However, few accurate testing methodologies are available in clinical settings. This study examined the validity and reliability of eccentric knee flexor torque measured with a novel Nordic Hamstring Device. Twenty-seven healthy participants were assessed in two sessions. Maximal eccentric knee flexion torque was measured by the Biodex system 4 and during the Nordic Hamstrings Exercise in the FysioMeter. A moderate correlation was found between the FysioMeter and the Biodex (rho 0.674). Limits of agreement showed a bias of −43.9 Nm, suggesting that the FysioMeter underestimated the knee flexor peak torque compared with the Biodex. The FysioMeter showed excellent test–retest reliability (ICC 0.964). In conclusion, the FysioMeter provided moderately valid measures of knee flexor peak torque and excellent test–retest reliability. The FysioMeter could be a valuable tool for guiding rehabilitation in settings where access to isokinetic dynamometry is limited.

ELW-CNN: An extremely lightweight convolutional neural network for enhancing interoperability in colon and lung cancer identification using explainable AI.

Cancer is a condition in which cells in the body grow uncontrollably, often forming tumours and potentially spreading to various areas of the body. Cancer is a hazardous medical case in medical history analysis. Every year, many people die of cancer at an early stage. Therefore, it is necessary to accurately and early identify cancer to effectively treat and save human lives. However, various machine and deep learning models are effective for cancer identification. Therefore, the effectiveness of these efforts is limited by the small dataset size, poor data quality, interclass changes between lung squamous cell carcinoma and adenocarcinoma, difficulties with mobile device deployment, and lack of image and individual-level accuracy tests. To overcome these difficulties, this study proposed an extremely lightweight model using a convolutional neural network that achieved 98.16% accuracy for a large lung and colon dataset and individually achieved 99.02% for lung cancer and 99.40% for colon cancer. The proposed lightweight model used only 70 thousand parameters, which is highly effective for real-time solutions. Explainability methods such as Grad-CAM and symmetric explanation highlight specific regions of input data that affect the decision of the proposed model, helping to identify potential challenges. The proposed models will aid medical professionals in developing an automated and accurate approach for detecting various types of colon and lung cancer.

Self-perceived knowledge, influencing factors and proposed educational interventions for diagnostic stewardship in the microbiology laboratory: The experience of Irish paediatric doctors.

Lab users should display an understanding of microbiology tests and avoid inappropriate requests which may yield negative clinical consequences. Our research addressed diagnostic stewardship education and the perspectives of paediatric hospital doctors on their (1) self-perceived knowledge, (2) motivators for test ordering, (3) proposed educational interventions and (4) preferred teaching modalities. An anonymous, mixed-methods, 34-item online questionnaire was distributed to paediatric doctors across three Irish hospital sites from March to May 2023. Descriptive statistics summarised Likert-scale responses and inductive thematic analysis was used to analyse open-ended items. Subgroup analyses examined differences between consultants and non-consultant hospital doctors (NCHDs). The analysis included 100 respondents (n = 45 consultants, n = 55 NCHDs, 24% response rate). Consultants scored higher than NCHDs in self-perceived knowledge (p < 0.001). Patient comorbidity, clinical status, local guidelines and accuracy of microbiological tests were the strongest motivators for test-ordering. Consultants were more likely to be influenced by accuracy (p = 0.03), costs (p = 0.01) and laboratory workload (p = 0.01). Only 27% reported formal teaching on diagnostic stewardship. NCHDs demonstrated increased willingness to engage in educational interventions (p = 0.024), and a comparatively higher interest in in-person teaching (p = 0.002) and gamification (p = 0.02). Respondents indicated a preference for formal guidelines on microbiology testing, in-person teaching, cost familiarisation and e-learning modules. Novel methods (peer-to-peer programmes, simulation and gamification) were less favoured. Dominant themes included; collaborative learning, increased microbiology presence, accessibility to education, barriers and administrative issues. Formal diagnostic stewardship educational programmes should be integrated into post-graduate curriculums for both NCHDs and consultants. Focuses for educational interventions include; development of local diagnostic stewardship guidelines, teaching on the cost and accuracy of microbiology tests, senior supervision and role-modelling.

Application of artificial intelligence to ultrasound imaging for benign gynecological disorders: systematic review.

Although artificial intelligence (AI) is increasingly being applied to ultrasound imaging in gynecology, efforts to synthesize the available evidence have been inadequate. The aim of this systematic review was to summarize and evaluate the literature on the role of AI applied to ultrasound imaging in benign gynecological disorders. Web of Science, PubMed and Scopus databases were searched from inception until August 2024. Inclusion criteria were studies applying AI to ultrasound imaging in the diagnosis and management of benign gynecological disorders. Studies retrieved from the literature search were imported into Rayyan software and quality assessment was performed using the Quality Assessment Tool for Artificial Intelligence-Centered Diagnostic Test Accuracy Studies (QUADAS-AI). Of the 59 studies included, 12 were on polycystic ovary syndrome (PCOS), 11 were on infertility and assisted reproductive technology, 11 were on benign ovarian pathology (i.e. ovarian cysts, ovarian torsion, premature ovarian failure), 10 were on endometrial or myometrial pathology, nine were on pelvic floor disorder and six were on endometriosis. China was the most highly represented country (22/59 (37.3%)). According to QUADAS-AI, most studies were at high risk of bias for the subject selection domain (because the sample size, source or scanner model was not specified, data were not derived from open-source datasets and/or imaging preprocessing was not performed) and the index test domain (AImodels were not validated externally), and at low risk of bias for the reference standard domain (the reference standard classified the target condition correctly) and the workflow domain (the time between the index test and the reference standard was reasonable). Most studies (40/59) developed and internally validated AI classification models for distinguishing between normal and pathological cases (i.e. presence vs absence of PCOS, pelvic endometriosis, urinary incontinence, ovarian cyst or ovarian torsion), whereas 19/59 studies aimed to automatically segment or measure ovarian follicles, ovarian volume, endometrial thickness, uterine fibroids or pelvic floor structures. The published literature on AI applied to ultrasound in benign gynecological disorders is focused mainly on creating classification models to distinguish between normal and pathological cases, and on developing models to automatically segment or measure ovarian volume or follicles. © 2025 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Smartphone-Based Applications for Atrial Fibrillation Detection: A Systematic Review and Meta-Analysis of Diagnostic Test Accuracy.

Background: Atrial fibrillation (AF) burden is strongly associated with an increased risk of stroke, which, in most cases, can be prevented through earlier detection of AF and the timely initiation of anticoagulation therapy. Smartphone devices can provide a simple, non-invasive, cost-effective early AF detection solution. Methods: PubMed, Embase, and Scopus databases were searched for studies comparing smartphone-based photoplethysmography (PPG) with standard electrocardiogram for AF detection. A bivariate random-effects model with a 95% confidence interval (CI) was applied to generate the summary receiver operating characteristic (SROC) curve. Results: Fourteen studies were included, comprising 5,090 patients with an AF prevalence of 31.6%. The pooled sensitivity and specificity were 0.96 (95% CI, 0.93-0.97) and 0.97 (95% CI, 0.95-0.98). The area under the SROC curve was 0.98 (95% CI, 0.94-0.99). The diagnostic odds ratio was 960 (95% CI, 439-2,104), with significant heterogeneity (I2 = 51%). The projected positive and negative predictive values were 66.5% and 99.7%, respectively, in the elderly population aged >65 years and 39.2% and 99.9% in the general population. Conclusion: Smartphone-based PPG demonstrated relatively high sensitivity and specificity and appears capable of ruling out AF. Patients aged >65 are more likely to benefit from AF screening.

Urinary DNA Methylation Test for Bladder Cancer Diagnosis

An accurate noninvasive biomarker test is needed for the early diagnosis of bladder cancer. To evaluate the performance of a urinary DNA methylation test (PENK methylation) and compare its diagnostic accuracy with that of the nuclear matrix protein 22 (NMP22) test or urine cytology test. In this prospective multicenter study at 10 sites in the Republic of Korea, individuals 40 years and older with hematuria undergoing cystoscopy within 3 months between March 11, 2022, and May 30, 2024, participated. The study participants were evaluated for bladder cancer using a urinary DNA methylation test. Urinary DNA methylation test, NMP22 test, and urine cytology test. The primary outcomes were the sensitivity and specificity of the urinary DNA methylation test for high-grade or invasive bladder cancer. Secondary objectives included the accuracy of the test for overall bladder cancer (all stages and grades) and the comparison of sensitivities and specificities for bladder cancer between the urinary DNA methylation test and the NMP22 test or urine cytology test. Among the 1099 participants, 614 (55.9%) were male; participants had a mean (SD) age of 65 (10) years. Of the 1099 participants, 219 and 176 participants had bladder cancer and high-grade or invasive bladder cancer, respectively. The urinary DNA methylation test had sensitivity and specificity for high-grade or invasive bladder cancer of 89.2% (95% CI, 84.6%-93.8%) and 87.8% (95% CI, 85.6%-89.9%), respectively. Sensitivity and specificity for overall bladder cancer were 78.1% (95% CI, 72.6%-83.6%) and 88.8% (95% CI, 86.7%-90.8%), respectively. The positive predictive value for high-grade or invasive bladder cancer was 61.3% (95% CI, 55.4%-67.3%), and the negative predictive value was 97.6% (95% CI, 96.6%-98.7%). In comparison with the NMP22 test or urine cytology test, the urinary DNA methylation test showed significantly superior sensitivity for high-grade or invasive bladder cancer and overall bladder cancer. In this prospective multicenter study of individuals with hematuria, the urinary DNA methylation test showed 89% sensitivity for detecting high-grade or invasive bladder cancer, outperforming the NMP22 test or urine cytology test with high specificity. While this test had an excellent negative predictive value, its positive predictive value was suboptimal.

Effect of Varying Frequency of Fatigue Test on Durability of AISI 1045 Carbon Steel Plate Under Spectral Loading

In many practical applications, AISI 1045 carbon steel is subjected to various external factors, one of which is vibration. This vibration can induce resonance in the material, ultimately leading to fatigue due to cyclic stress. The aim of this study is to investigate the mechanical behavior of AISI 1045 carbon steel when subjected to both static and cyclic loading conditions, and to assess how its fatigue properties change across different test frequencies, and to validate a mathematical model that predicts fatigue life through statistical analysis. To ensure accurate and standardized testing, the specimen geometry follows the ASTM E466 standard for tensile and stress-life tests, and the ASTM E290 standard for three-point bending tests. The study involves conducting a series of mechanical tests, including tensile tests, stress-life tests, and three-point bending tests, which provide critical data on the material’s fatigue life under various conditions. By analyzing the results, the researchers aim to establish a clear relationship between the applied load frequency and the material’s fatigue life. Additionally, based on this relationship, they propose a mathematical model to predict the fatigue life of AISI 1045 carbon steel across different frequencies. This model could be useful in estimating the durability and performance of components made from this steel when subjected to dynamic and fluctuating loads, helping engineers design more reliable system and predicting material performance under varying operational conditions.

A deep learning based model for diabetic retinopathy grading

Diabetic retinopathy stands as a leading cause of blindness among people. Manual examination of DR images is labor-intensive and prone to error. Existing methods to detect this disease often rely on handcrafted features which limit the adaptability and classification accuracy. Thus, the aim of this research is to develop an automated and efficient system for early detection and accurate grading of diabetic retinopathy severity with less time consumption. In our research, we have developed a deep neural network named RSG-Net (Retinopathy Severity Grading) to classify DR into 4 stages (multi-class classification) and 2 stages (binary classification). The dataset utilized in this study is Messidor-1. In preprocessing, we have used Histogram Equalization to improve image contrast and denoising techniques to remove noise and artifacts which enhanced the clarity of the fundus images. We applied data augmentation techniques to our preprocessed images in order to tackle class imbalance issues. Augmentation techniques involve flipping, rotation, zooming and adjustment of color, contrast and brightness. The proposed RSG-Net model contains convolutional layers to perform automatic feature extraction from the input images and batch normalization layers to improve training speed and performance. The model also contains max pooling, drop out and fully connected layers. Our proposed RSG-Net model achieved a testing accuracy of 99.36%, specificity of 99.79% and a sensitivity of 99.41% in classifying diabetic retinopathy into 4 grades and it achieved 99.37% accuracy, 100% sensitivity and 98.62% specificity in classifying DR into 2 grades. The performance of RSG-Net is also compared with other state-of-the-art methodologies where it outperformed these methods.

MOVING DIAGNOSIS OF LIVER FIBROSIS INTO THE COMMUNITY.

Chronic liver diseases (CLD) are a leading cause of death worldwide, with alcohol consumption and metabolic risk factors as the two reasons accounting for the majority of cases of CLD in many developed countries. Currently there is a lack of specific strategies for early diagnosis of CLD and consequently most cases are diagnosed in advanced stages of the disease, which is associated with negative consequences for disease management and prognosis. Screening for CLD is based on either detection of chronic viral hepatitis B and C, or detection of liver fibrosis in patients with steatotic liver disease related to alcohol or metabolic dysfunction. There are non-invasive tools available for detection of liver fibrosis, including serological and imaging-based tests. Clinical practice guidelines recommend screening for liver fibrosis using algorithms that combine different non-invasive tests, with widely available but low accuracy tests such as FIB-4 for a first screening step in primary care setting, and other tests with less availability but higher accuracy, such as Transient Elastography or Enhanced Liver Fibrosis Test as a second step. There are different pathways for early detection of patients with CLD from primary to specialized care, where primary care providers are key for early detection, management and referral of patients. In addition, intervention on metabolic risk factors and alcohol consumption should be carried out in collaboration between specialized therapy and primary care. This review describes liver fibrosis from the community perspective, highlighting gaps in knowledge on how to define the optimal combination of tests, target population, the ideal pathway of care for CLD, and how to increase implementation of programs for early diagnosis of liver diseases in clinical practice.

Potential Impacts of Climate Change on the Spatial Distribution Pattern of Naked Oats in China

Naked oats, a significant minor cereal crop in China popular for its nutrient richness, have experienced a surge in production in recent years, fueled by the escalating demand for wholesome healthy food. However, the dispersed and disorganized cultivation plan of naked oats poses a significant constraint on its industrial progression. Considering the dual influence of cultivation, management techniques, and global climate change on the production of naked oats, this study explores the potential impacts of climate change on the spatial distribution and yield of this cereal crop. Leveraging CMIP6 climate models (BCC-CSM2-MR, CanESM5, CNRM-ESM2-1) and an optimized MaxEnt model (RM = 0.5, FC = LQ), we simulated potential climate-suitable zones for naked oats from 1990 to 2020 and forecasted alterations under various emission scenarios from 2021 to 2100. The model achieved an average accuracy test with high value (AUC = 0.945) in predicting suitable areas; with precipitation seasonality (Coefficient of Variation) (bio15, 21.70%) and topsoil pH (H2O) (T_PH_H2O, 21.00%) as key factors, both climate and soil properties have a greater influence. Simulation results showed that the climatically suitable area for naked oats increased under all scenarios, with the largest increase in the optimal growing area under ssp126 in the 2030s. The increase was 3.93% with an area of 0.77 × 106 km2. The study also compared the data from the main producing counties of naked oats in Shanxi Province from 2020 to 2022 for statistical purposes, and found that 39 counties were in high climatic suitability zones and 39 counties were in remarkably high climatic suitability zones. The agreement rate between planting areas and climatically suitable areas was as high as 97.44%. Further, the growing area expanded westward, increasing the production intensity. This study reveals the current spatial distribution pattern of naked oats, providing a scientific rationale for addressing climate change through multi-scenario predictions. Our findings have implications for optimizing cultivation layout and identifying optimal zones, supporting sustainable agricultural development strategies in China.