Area Under Curve Research Articles

The difference of functional MR imaging in evaluating outcome of patients with diffuse and compact brain arteriovenous malformation.

Microsurgical resection is an effective method to treat brain arteriovenous malformations (BAVMs). Functional magnetic resonance imaging (fMRI) can evaluate the spatial relationship of nidus and eloquent. Diffuse BAVMs are related to poor outcomes postoperatively. The role of fMRI in evaluating outcomes in patients with different nidus types remains unclear. BAVM patients received microsurgical resection were included from a prospective, multicenter cohort study. All patients underwent fMRI evaluation preoperatively and were regularly followed up postoperatively. Diffuse BAVM is radiologically identified as nidus containing normal brain tissue interspersing between malformed vessels. Lesion-to-eloquent distance (LED) was calculated based on the relationship between nidus and eloquent. The primary outcome was 180-day unfavorable neurological status postoperatively. The risk of primary outcome was investigated within different BAVM nidus types. The LED's performance to predict poor outcome was evaluated using area under curve (AUC). 346 BAVM patients were included in this study. 93 (26.9%) patients were found to have a 180-day unfavorable outcome. Multivariate logistic analysis demonstrated LED (odd ratio [OR], 0.44; 0.34-0.57; P < 0.001) and mRS at admission (OR, 2.59; 1.90-3.54; P < 0.001) as factors of unfavorable outcome. Subgroup analysis showed LED and mRS at admission as factors of unfavorable outcome for patients with compact BAVMs (all P < 0.05), but not for patients with diffuse BAVMs. Subsequent analysis showed that LED performed poorly to predict the unfavorable outcome for patients with diffuse BAVMs, compared with patients with compact BAVMs (AUC as 0.69 vs. 0.86, P < 0.05). A larger cutoff value of LED to unfavorable outcome was found in patients with diffuse BAVMs (15mm) compared with patients with compact BAVMs (4.7mm). Usage of LED to evaluate postoperative outcome of patients with diffuse BAVMs differs from its use in patients with compact BAVMs. Specific assessment strategy considering BAVM nidus types could help improve patients' outcome. MITASREAVM cohort (unique identifier: NCT02868008, https://clinicaltrials.gov/study/NCT02868008?term=NCT02868008&rank=1 ).

Profiles of cytokines in patients with antineutrophil cytoplasmic antibody-associated vasculitis.

This study aimed to identify plasma biomarkers that are significantly altered in patients with antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) and are closely associated with AAV disease activity, as well as to explore their role in the pathogenesis of AAV. Cytokines were measured using Human Immune Response Panel 80-Plex in plasma from 59 patients with AAV and 20 healthy controls (HCs). The differentially expressed cytokines between the two groups and the possible signaling pathway involved in the pathogenesis of AAV were analyzed by bioinformatics. Relationship analysis was performed between these cytokines and clinical parameters to identify the biomarkers that can effectively indicate disease activity. We identified 65 differentially expressed cytokines between the two groups. Among them, 43 cytokines significantly affected the risk of AAV. Bioinformatic analysis showed that the 43 cytokines were primarily enriched in signaling pathways such as cytokine-cytokine receptor interaction, viral protein interaction with cytokine and cytokine receptor, chemokine signaling pathway, and IL-17 signaling pathway. The levels of 25 cytokines were significantly positively correlated with Birmingham Vasculitis Activity Score (BVAS), and the levels of 2 cytokines were significantly negatively correlated with BVAS. Receiver operating characteristic analysis showed that 9 cytokines can distinguish between disease relapse and remission (PTX3: area under curve (AUC)=0.932, IL34: AUC=0.856, IL2RA: AUC=0.833, CCL23: AUC=0.826, VEGFA: AUC=0.811, TNFSF13: AUC=0.795, Granzyme A: AUC=0.788, CSF3: AUC=0.773 and IL1A: AUC=0.765). The elevated levels of these 9 cytokines suggested a risk of disease relapse. The AUC of CCL11 in disease relapse and remission was 0.811 (p=0.0116). Unlike the other 9 cytokines, a negatively association existed between CCL11 level and the risk of disease relapse. A group of cytokines that may be involved in AAV pathogenesis was identified. Increased PTX3, IL34, IL2RA, CCL23, and VEGFA levels correlate with active disease in AAV and may be used as biomarkers to identify the disease relapse of AAV.

SCALED SIAMESE CNN-BASED AUTOMATIC CLASSIFICATION ALGORITHM FOR DETECTING PULMONARY EMBOLISM

The majority of patients today have vascular illness that could cause pulmonary hypertension and pulmonary emboli which can lead to serious conditions. To diagnose changes in veins and arteries, manual, semi-automatic and automatic examination of a individuals chest computed tomography image or CT Pulmonary Angiogram (CTPA) is performed. Manual CT scan analysis is more tedious, takes more time and is not standardized. Thus the separation of vascular trees in CTPA scans is now semi-automatic and automatic which enables doctors to precisely identify aberrant conditions. Recent research has focused on applying machine learning along with deep learning approaches to find and categorize pulmonary vascular disorders like pulmonary embolism and pulmonary hypertension. Here, a brand-new technique is put out for the automatically classifying pulmonary veins in order to more accurately diagnose lung illnesses related to Pulmonary Embolism (PE). The proposed method uses Scaled Siamese-based Convoutional Neural Network (CNN) for classifying Pulmonary Embolism images with an accuracy more than 90 while using two publicly available datasets FUMPE and RSNA. Finally, comparison of the proposed method with other pretrained CNN-based algorithms for detecting PE is done in terms of Area Under Curve (AUC), recall, precision, accuracy, [Formula: see text]1 score, etc. and values are tabulated and our proposed method shows good results. Comparison was also done using deep learning models like PENET, Inception [Formula: see text]3 and [Formula: see text] fold cross validation and our proposed method values are outstanding compared to other methods.

A Decision Tree Model to Help Treatment Decision-Making for Unruptured Intracranial Aneurysms: A Multi-center, Long-Term Follow-up Study in a Large Chinese Cohort.

Chinese population have a high prevalence of unruptured intracranial aneurysm (UIA). Clinical and imaging risk factors predicting UIA growth or rupture are poorly understood in the Chinese population due to the lack of large-scale longitudinal studies, and the treatment decision for UIA patients was challenging. Develop a decision tree (DT) model for UIA instability, and validate its performance in multi-center studies. Single-UIA patients from two prospective, longitudinal multicenter cohort studies were analyzed, and set as the development cohort and validation cohort. The primary endpoint was UIA instability (rupture, growth, or morphological change). A DT was established within the development cohort and validated within the validation cohort. The performance of clinicians in identifying unstable UIAs before and after the help of the DT was compared using the area under curve (AUC). The development cohort included 1270 patients with 1270 UIAs and a follow-up duration of 47.2 ± 15.5months. Aneurysm instability occurred in 187 (14.7%) patients. Multivariate Cox analysis revealed hypertension (hazard ratio [HR], 1.54; 95%CI, 1.14-2.09), aspect ratio (HR, 1.22; 95%CI, 1.17-1.28), size ratio (HR, 1.31; 95%CI, 1.23-1.41), bifurcation configuration (HR, 2.05; 95%CI, 1.52-2.78) and irregular shape (HR, 4.30; 95%CI, 3.19-5.80) as factors of instability. In the validation cohort (n = 106, 12 was unstable), the DT model incorporating these factors was highly predictive of UIA instability (AUC, 0.88 [95%CI, 0.79-0.97]), and superior to existing UIA risk scales such as PHASES and ELAPSS (AUC, 0.77 [95%CI, 0.67-0.86] and 0.76 [95%CI, 0.66-0.86], P < 0.001). Within all 1376 single-UIA patients, the use of the DT significantly improved the accuracy of junior neurosurgical clinicians to identify unstable UIAs (AUC from 0.63 to 0.82, P < 0.001). The DT incorporating hypertension, aspect ratio, size ratio, bifurcation configuration and irregular shape was able to predict UIA instability better than existing clinical scales in Chinese cohorts. CLINICAL TRIAL REGISTRATION: IARP-CP cohort were included (unique identifier: ChiCTR1900024547. Published July 15, 2019. Completed December 30, 2020), with 100-Project phase-I cohort (unique identifier: NCT04872842, Published May 5, 2021. Completed November 8, 2022) as the development cohort. The 100-Project phase-II cohort (unique identifier: NCT05608122. Published November 8, 2022) as the validation cohort.

The application of B1 inhomogeneity-corrected variable flip angle T1 mapping for assessing liver fibrosis

PurposeThe aim of this study was to evaluate the diagnostic accuracy of the B1 inhomogeneity-corrected variable flip angle (VFA) method using native T1 values in the staging of liver fibrosis. MethodsEighty-three patients who presented for liver biopsy due to varying degrees of liver damage, underwent MR examinations and had T1-mapping images of the liver acquired using the B1 inhomogeneity-corrected VFA VIBE method. Among them, 65 patients underwent Fibroscan, and their results were used to evaluate the elasticity of liver tissue. Additionally, T1-mapping images were collected from 19 normal control patients. Independent sample t-tests were used to analyze the correlation between T1 mapping and Fibroscan. The diagnostic efficacy of T1 mapping in patients with different stages of liver fibrosis was evaluated using receiver operating characteristic (ROC) curves. ResultsThe consistency between different observer groups was intraclass correlation coefficient (ICC) =0.802. T1 mapping demonstrated significant differences between mid-stage liver fibrosis (S = 2) and late-stage liver fibrosis (S = 3), as well as moderate inflammation (G = 2) and severe inflammation (G = 3), P < 0.05. The Area Under Curve(AUC) values of T1 mapping for early liver fibrosis (S ≥ 1), significant liver fibrosis (S ≥ 2), advanced liver fibrosis (S ≥ 3), and end-stage liver fibrosis (S = 4) were 0.760, 0.709, 0.790, and 0.768, respectively. T1 mapping combined with Fibroscan had an AUC value of 0.860. ConclusionsThe B1 inhomogeneity-corrected VFA T1 mapping may be useful for the staging of liver fibrosis. It has a superior diagnostic efficiency for diagnosing advanced fibrosis (≥S3), while native T1 values combined with Fibroscan have potential value for the staging of liver fibrosis.

Differentiation of granulomatous nodules with lobulation and spiculation signs from solid lung adenocarcinomas using a CT deep learning model

BackgroundThe diagnosis of solitary pulmonary nodules has always been a difficult and important point in clinical research, especially granulomatous nodules (GNs) with lobulation and spiculation signs, which are easily misdiagnosed as malignant tumors. Therefore, in this study, we utilised a CT deep learning (DL) model to distinguish GNs with lobulation and spiculation signs from solid lung adenocarcinomas (LADCs), to improve the diagnostic accuracy of preoperative diagnosis.Methods420 patients with pathologically confirmed GNs and LADCs from three medical institutions were retrospectively enrolled. The regions of interest in non-enhanced CT (NECT) and venous contrast-enhanced CT (VECT) were identified and labeled, and self-supervised labels were constructed. Cases from institution 1 were randomly divided into a training set (TS) and an internal validation set (IVS), and cases from institutions 2 and 3 were treated as an external validation set (EVS). Training and validation were performed using self-supervised transfer learning, and the results were compared with the radiologists’ diagnoses.ResultsThe DL model achieved good performance in distinguishing GNs and LADCs, with area under curve (AUC) values of 0.917, 0.876, and 0.896 in the IVS and 0.889, 0.879, and 0.881 in the EVS for NECT, VECT, and non-enhanced with venous contrast-enhanced CT (NEVECT) images, respectively. The AUCs of radiologists 1, 2, 3, and 4 were, respectively, 0.739, 0.783, 0.883, and 0.901 in the (IVS) and 0.760, 0.760, 0.841, and 0.844 in the EVS.ConclusionsA CT DL model showed great value for preoperative differentiation of GNs with lobulation and spiculation signs from solid LADCs, and its predictive performance was higher than that of radiologists.

Prediction and verification of benignancy and malignancy of pulmonary nodules based on inflammatory related biological markers

ObjectiveInflammation plays an important role in the transformation of pulmonary nodules (PNs) from benign to malignant. Prediction of benignancy and malignancy of PNs is still lacking efficacy methods. Although Mayo or Brock model have been widely applied in clinical practices, their application conditions are limited. This study aims to construct a diagnostic model of PNs by machine learning using inflammation-related biological markers (IRBMs). MethodsInflammatory related genes (IRGs) were first extracted from GSE135304 chip data. Then, differentially expressed genes (DEGs) and infiltrating immune cells were screened between malignant pulmonary nodules (MN) and benign pulmonary nodule (BN). Correlation analysis was performed on DEGs and infiltrating immune cells. Molecular modules of IRGs were identified through Consistency cluster analysis. Subsequently, IRBMs in IRGs modules were filtered through Weighted gene co-expression network analysis (WGCNA). An optimal diagnostic model was established using machine learning methods. Finally, external dataset GSE108375 was used to verify this result. Results4 hub IRGs and 3 immune cells showed significantly difference between MN and BN, C1 and C2 module, namely PRTN3, ELANE, NFKB1 and CTLA4, T cells CD4 naïve, NK cells activated and Monocytes. IRBMs were screened from black module and yellowgreen module through WGCNA analysis. The Support vector machines (SVM) was identified as the optimal model with the Area Under Curve (AUC) was 0.753. A nomogram was established based on 5 hub IRBMs, namely HS.137078, KLC3, C13ORF15, STOM and KCTD13. Finally, external dataset GSE108375 verified this result, with the AUC was 0.718. ConclusionSVM model established by 5 hub IRBMs was able to effectively identify MN or BN. Accumulating inflammation and immune dysfunction were important to the transformation from BN to MN.

Effects of hydromorphone-based patient-controlled intravenous analgesia on postoperative hypoxaemia: a randomised controlled non-inferiority clinical trial

ObjectiveThis study aimed to compare the effects of patient-controlled intravenous analgesia (PCIA) with and without low-basal infusion on postoperative hypoxaemia.DesignA randomised parallel-group non-inferiority trial.SettingThe trial was conducted at a grade-A...

Diagnostic and prognostic value of independent and combined detection of computed tomography, ultrasound, and positron emission tomography - computed tomography in lymphoma

Lymphoma is a malignant tumor originating from the lymphopoietic system, which can affect all tissues and organs of the body. Lymphoma is highly heterogeneous and the therapeutic effect varies greatly. Different pathological types and stages of lymphoma differ greatly in terms of treatment intensity and prognosis. Early diagnosis of lymphoma is very important to improve the prognosis of patients. Therefore, this work explored the diagnostic value of independent and combined detection of computed tomography (CT), ultrasound, and positron emission tomography - computed tomography (PET-CT) for lymphoma. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) in the PET-CT and combination groups were greatly higher than those in the CT and ultrasound groups, showing obvious differences (P < 0.05). The area under curve (AUC) values in the CT group, ultrasound group, PET-CT group, and combination group were 0.632 (P = 0.032), 0.614 (P = 0.025), 0.793 (P = 0.002), and 0.859 (P = 0.001), respectively, exhibiting observable differences (P < 0.05). the sensitivity and specificity of PET/CT for lymphoma were higher than those of CT and ultrasound, which can clearly show the early mild results of lymphatic lymphoma. Therefore, the combined diagnosis of lymphatic lymphoma with PET/CT was of high clinical value.

Deep learning-based image quality assessment for optical coherence tomography macular scans: a multicentre study

AimsTo develop and externally test deep learning (DL) models for assessing the image quality of three-dimensional (3D) macular scans from Cirrus and Spectralis optical coherence tomography devices.MethodsWe retrospectively collected two...

Prognostic significance of negative lymph node count in microsatellite instability-high colorectal cancer

BackgroundMicrosatellite instability-high (MSI-H) tumors, with elevated tumor mutational burden and expression of neoantigens, represent a distinct immune-activated subpopulation in colorectal cancer (CRC), characterized by strong lymph node reaction, locally advanced tumor and higher total lymph nodes harvested (TLN), but less metastatic lymph nodes and fewer incidence of III-IV stage. Host immune response to tumor and lymph nodes may be an important prognostic factor. However, N stage and LNR (Lymph-Node Ratio) have limitations in predicting the prognosis of MSI-H patients. Negative lymph node count (NLC) provided a more precise representation of immune activation status and extent of tumor metastasis. The study aims to detect prognostic significance of NLC in MSI-H CRC patients, and compare it with N stage, TLN and LNR.MethodsRetrospective data of 190 consecutive MSI-H CRC patients who received curative resection were collected. Survival analyses were performed using the Kaplan–Meier method. Clinicopathological variables including NLC, N stage, TLN and LNR were studied in univariate and multivariate COX regression analyses. ROC (receiver operating characteristic curve) and concordance index were employed to compare the differences in predictive efficacy between NLC, N stage, TLN and LNR.ResultsPatients with increased NLC experienced a significantly improved 5-years DFS and OS in Kaplan–Meier analysis, univariate analysis, and multivariate analysis, independent of potential confounders examined. Increased NLC corresponded to elevated 5-years DFS rate and 5-years OS rate. AUC (area under curve) and concordance index of NLC in DFS and OS predicting were both significantly higher than N stage, TLN and LNR.ConclusionsNegative lymph node is an important independent prognostic factor for MSI-H patients. Reduced NLC is associated with tumor recurrence and poor survival, which is a stronger prognostic factor than N stage, TLN and LNR.

Association of serum resolvin D1 with the risk of major adverse cardiovascular events in hemodialysis patients.

Resolvin D1 (RvD1) inhibits inflammation, reduces oxidative stress, and forecasts the risk of cardiovascular events, but relevant evidence in hemodialysis patients is lacking. This study intended to investigate the predictive value of RvD1 for major adverse cardiovascular events (MACE) risk in hemodialysis patients. Totally, 252 patients who underwent hemodialysis were included. Serum RvD1 was measured by enzyme-linked immunosorbent assay. Patients were followed up with a median of 12.1months. MACE was recorded during the follow-up period. RvD1 was inversely correlated with diabetes history (P = 0.002), cardiac troponin T (TnT) (P = 0.029), and high sensitivity C-reactive protein (hsCRP) (P < 0.001) in hemodialysis patients. 25 hemodialysis patients experienced MACE. RvD1 was reduced in hemodialysis patients with MACE versus those without MACE (P = 0.004). RvD1 exhibited a certain value in forecasting MACE risk, with an area under curve (AUC) of 0.675 [95% confidence interval CI: 0.565-0.786]. Increased RvD1 cut by median (P = 0.043) and cut by quartile (P = 0.042) were related to decreased accumulating MACE in hemodialysis patients. Moreover, RvD1 independently predicted declined MACE risk [odds ratio (OR) = 0.644, P = 0.045], but age (OR = 1.048, P = 0.039) and TnT (OR = 1.006, P = 0.005) independently predicted ascended MACE risk in hemodialysis patients. The combination of these independent factors displayed a good value for estimating MACE risk in hemodialysis patients with an AUC of 0.744 (95% CI: 0.640-0.849). Serum RvD1 is inversely correlated with diabetes history, TnT, and hsCRP in hemodialysis patients. More importantly, it could serve as a potential marker to predict MACE risk in these patients.

Image-based machine learning and cluster analysis for urban road network: employing Orange for codeless visual programming

ABSTRACT In the pursuit of sustainable urban development, it is imperative to address strategies for optimizing and harmonizing the configuration of urban spatial layouts, especially in response to rapid urbanization. Urban road networks play a pivotal role in shaping intricate urban spatial patterns as they intricately weave throughout the cityscape, significantly influencing land utilization, spatial arrangements, and the mobility of residents. The urban environment is a complex system teeming with multifaceted human activities. With the continuous evolution of advanced data and computational technologies, it is now feasible to construct intricate models for an in-depth exploration of urban spatial structures. In this study, we constructed an analytical framework that leveraged road network image recognition, multi-scale influence factors analysis, and predictions using codeless visual programming. Within this framework, deep learning and unsupervised clustering are applied to complete the hierarchical clustering of road networks and exploit the intrinsic association between cities. Meanwhile, hierarchical clustering and natural and humanistic attributes of cities are connected to investigate the strength of different attributes on specific clusters. The research has the following results: (1) In addition to the alignment of urban road network clustering outcomes, quantitative unevenness and geographical heterogeneity are also observed. (2) Multiple attributes have a complicated impact on urban space, including city size, and socioeconomic and geographic characteristics. Among these factors, socio-economic variables emerge as the principal drivers influencing urban spatial structure. (3) The research enhances the model’s generalization capability and predictive accuracy. As attributes increase and model learning improves, both the Area Under Curve (AUC) and F-1 scores exhibit continuous enhancements, ultimately achieving error-free predictions. The study’s findings offer valuable insights into urban spatial structures and systems, supporting urban decision-makers in addressing challenges associated with urban growth.

Intelligent diagnostic model for pterygium by combining attention mechanism and MobileNetV2.

To evaluate the application of an intelligent diagnostic model for pterygium. For intelligent diagnosis of pterygium, the attention mechanisms-SENet, ECANet, CBAM, and Self-Attention-were fused with the lightweight MobileNetV2 model structure to construct a tri-classification model. The study used 1220 images of three types of anterior ocular segments of the pterygium provided by the Eye Hospital of Nanjing Medical University. Conventional classification models-VGG16, ResNet50, MobileNetV2, and EfficientNetB7-were trained on the same dataset for comparison. To evaluate model performance in terms of accuracy, Kappa value, test time, sensitivity, specificity, the area under curve (AUC), and visual heat map, 470 test images of the anterior segment of the pterygium were used. The accuracy of the MobileNetV2+Self-Attention model with 281 MB in model size was 92.77%, and the Kappa value of the model was 88.92%. The testing time using the model was 9ms/image in the server and 138ms/image in the local computer. The sensitivity, specificity, and AUC for the diagnosis of pterygium using normal anterior segment images were 99.47%, 100%, and 100%, respectively; using anterior segment images in the observation period were 88.30%, 95.32%, and 96.70%, respectively; and using the anterior segment images in the surgery period were 88.18%, 94.44%, and 97.30%, respectively. The developed model is lightweight and can be used not only for detection but also for assessing the severity of pterygium.

Factors associated with a better treatment efficacy among psoriasis patients: a study based on decision tree model and logistic regression in Shanghai, China

BackgroundMany effective therapies for psoriasis are being applied in clinical practice in recent years, however, some patients still can’t achieve satisfied effect even with biologics. Therefore, it is crucial to identify factors associated with the treatment efficacy among psoriasis patients. This study aims to explore factors influencing the treatment efficacy of psoriasis patients based on decision tree model and logistic regression.MethodsWe implemented an observational study and recruited 512 psoriasis patients in Shanghai Skin Diseases Hospital from 2021 to 2022. We used face-to-face questionnaire interview and physical examination to collect data. Influencing factors of treatment efficacy were analyzed by using logistic regression, and decision tree model based on the CART algorithm. The receiver operator curve (ROC) was plotted for model evaluation and the statistical significance was set at P < 0.05.ResultsThe 512 patients were predominately males (72.1%), with a median age of 47.5 years. In this study, 245 patients achieved ≥ 75% improvement in psoriasis area and severity index (PASI) score in week 8 and was identified as treatment success (47.9%). Logistic regression analysis showed that patients with senior high school and above, without psoriasis family history, without tobacco smoking and alcohol drinking had higher percentage of treatment success in patients with psoriasis. The final decision tree model contained four layers with a total of seventeen nodes. Nine classification rules were extracted and five factors associated with treatment efficacy were screened, which indicated tobacco smoking was the most critical variable for treatment efficacy prediction. Model evaluation by ROC showed that the area under curve (AUC) was 0.79 (95%CI: 0.75 ~ 0.83) both for logistic regression model (0.80 sensitivity and 0.69 specificity) and decision tree model (0.77 sensitivity and 0.73 specificity).ConclusionPsoriasis patients with higher education, without tobacco smoking, alcohol drinking and psoriasis family history had better treatment efficacy. Decision tree model had similar predicting effect with the logistic regression model, but with higher feasibility due to the nature of simple, intuitive, and easy to understand.

The Utility of the Surgical Apgar Score in Assessing the Risk of Perioperative Complications Following Spinal Fusion Surgery for Pediatric Patients with Scoliosis and Cerebral Palsy.

Cohort study. Validate the Surgical Apgar Score (SAS) as a means of predicting perioperative major complications occurring within 30 days after scoliosis surgery in pediatric patients with cerebral palsy (CP). A patient's SAS, which is composed of three commonly recorded intraoperative variables, predicts postoperative complications after various types of spine surgery. This has not; however, been studied in pediatric patients with scoliosis and CP, a population that experiences a high incidence of complications after corrective spinal surgery. Pediatric CP patients who underwent spinal correction surgery were included in this study. Patient background, surgical variables, and perioperative complications occurring within 30 days after surgery were collected. Patients were divided into 4 groups based on their SAS: SAS 0-4, SAS 5-6, SAS 7-8, SAS 9-10. The incidences of perioperative complications for each group were compared using a receiver operating characteristic analysis. Area under curve (AUC) is reported. A total of 111 patients met the inclusion criteria. There were no death cases. There were 44 (39.6%) perioperative major complications in 37 (33.3%) patients that occurred within 30 days after spine surgery. The most frequent perioperative complications were pulmonary issues (13.5%). The incidence of perioperative major complication in each SAS group was as follows: SAS 0-4; 51.6%, SAS 5-6; 30.2%, SAS 7-8; 18.5%, SAS 9-10; 0/0. When the SAS 7-8 group was set as the reference, there was no significant difference compared to SAS 5-6 (P=0.34), while the incidence rate was significantly increased in SAS 0-4 (P=0.02). The AUC was 0.65 (95% Confidence Interval: 0.54-0.75). Overall, there were 37 (33.3%) patients with CP who had a major complication within 30 days after spinal surgery. Lower SAS, with the 0-4 group being the cutoff, were associated with significantly higher complication rates than higher SAS groups.

HLFSRNN-MIL: A Hybrid Multi-Instance Learning Model for 3D CT Image Classification

At present, many diseases are diagnosed by computer tomography (CT) image technology, which affects the health of the lives of millions of people. In the process of disease confrontation, it is very important for patients to detect diseases in the early stage by deep learning of 3D CT images. The paper offers a hybrid multi-instance learning model (HLFSRNN-MIL), which hybridizes high-low frequency feature fusion (HLFFF) with sequential recurrent neural network (SRNN) for CT image classification tasks. Firstly, the hybrid model uses Resnet-50 as the deep feature. The main feature of the HLFSRNN-MIL lies in its ability to make full use of the advantages of the HLFFF and SRNN methods to make up for their own weakness; i.e., the HLFFF can extract more targeted feature information to avoid the problem of excessive gradient fluctuation during training, and the SRNN is used to process the time-related sequences before classification. The experimental study of the HLFSRNN-MIL model is on two public CT datasets, namely, the Cancer Imaging Archive (TCIA) dataset on lung cancer and the China Consortium of Chest CT Image Investigation (CC-CCII) dataset on pneumonia. The experimental results show that the model exhibits better performance and accuracy. On the TCIA dataset, HLFSRNN-MIL with Residual Network (ResNet) as the feature extractor achieves an accuracy (ACC) of 0.992 and an area under curve (AUC) of 0.997. On the CC-CCII dataset, HLFSRNN-MIL achieves an ACC of 0.994 and an AUC of 0.997. Finally, compared with the existing methods, HLFSRNN-MIL has obvious advantages in all aspects. These experimental results demonstrate that HLFSRNN-MIL can effectively solve the disease problem in the field of 3D CT images.

Quantitative estimation of fenticonazole nitrate by zero-order derivative area under curve spectrophotometric methods in bulk and in-capsule dosage form

The purpose of this study is to establish Zero-order UV-Spectrophotometric - absorbance and Zero Order- Area under curve (AUC) methods for estimation of Fenticonazole Nitrate in bulk and vaginal capsules. Fenticonazole Nitrate is an antifungal drug and it is completely insoluble in water. Methanol was used as solvent for solubilization of Fenticonazole Nitrate. Maximum absorption for Fenticonazole Nitrate was found to be at wavelength 253 nm when dissolved in Methanol. The methods are based upon measurement of absorbance at 253nm and integration of area under curve for analysis of Fenticonazole Nitrate in the wavelength range of 242-262 nm. The drug followed linearity in the concentration range of 5 - 30µg/mL with correlation coefficient value r&amp;#62; 0.99 for both methods. The proposed methods were validated for accuracy (% recovery), precision, repeatability and ruggedness, as per ICH guidelines. The proposed methods were applied for qualitative and quantitative estimation of Fenticonazole Nitrate in vaginal capsules and results were found in good agreement with the label claimed. Developed methods can be used for routine analysis of Fenticonazole Nitrate in bulk and Vaginal Capsules.

Diagnostic utility of transfer learning by using convolutional neural network for cytological diagnosis of malignant effusions.

Cytological analysis of effusion specimens provides critical information regarding the diagnosis and staging of malignancies, thus guiding their treatment and subsequent monitoring. Keeping in view the challenges encountered in the morphological interpretation, we explored convolutional neural networks (CNNs) as an important tool for the cytological diagnosis of malignant effusions. A retrospective review of patients at our institute, over 3.5 years yielded a dataset of 342 effusion samples and 518 images with known diagnoses. Cytological examination and cell block preparation were performed to establish correlation with the gold standard, histopathology. We developed a deep learning model using PyTorch, fine-tuned it on a labelled dataset, and evaluated its diagnostic performance using test samples. The model exhibited encouraging results in the distinction of benign and malignant effusions with area under curve (AUC) of 0.8674, F-measure or F1 score which denotes the harmonic mean of precision and recall, to be 0.8678 thus, demonstrating optimal accuracy of our CNN model. The study highlights the promising potential of transfer learning in enhancing the clinical pathology laboratory efficiency when dealing with malignant effusions.

Combined MR quantitative susceptibility mapping and multi-shell diffusion in Parkinson's disease.

Quantitative susceptibility mapping (QSM), neurite orientation dispersion and density imaging (NODDI), and the g-ratio have separately shown differences between Parkinson's disease (PD) and healthy controls. The g-ratio has, however, not been studied in PD in the substantia nigra (SN) and the putamen. A combination of these methods could also potentially be a complementary imaging biomarker for PD. This study aimed to assess the diagnostic performance of QSM, NODDI, the g-ratio, and a combined QSM-NODDI imaging marker in the SN and putamen of PD patients. In this prospective study, the diagnostic performance of median region of interest values was compared in a cohort of 15 participants with PD and 14 healthy controls after manual segmentation. The diagnostic performance was assessed using the area under curve (AUC) for the receiving operator characteristic. Median QSM in the contralateral SN identified PD with AUC 0.77, and median isotropic volume fraction identified PD in the ipsilateral SN with AUC 0.68. A combined NODDI-QSM marker improved diagnostic performance (AUC 0.80). No significant differences were found in the g-ratio. A combination of median QSM and median isotropic volume fraction improves the differentiation of PD from healthy controls and is a potential biomarker in the diagnostics of PD. This confirms previously reported results indicating that combining QSM and NODDI modestly improves differentiation of PD.