Higher Area Under The Curve Values Research Articles

Which hemogram-derived indices might be useful in predicting the clinical outcomes of sepsis patients in the intensive care unit?

Purpose: The aim of our study is to investigate the prognostic value of Neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), and platelet-to-lymphocyte ratio (PLR) in determining mortality in patients hospitalized for intensive care unit (ICU) sepsis.
 Materials and Methods: This study retrospectively evaluates all patients hospitalized with sepsis in our ICU between February 2017 and April 2018. In addition to the demographic and clinical characteristics of the patients, complete blood count parameters were also recorded. Demographic and clinical characteristics, mainly NLR, MLR and PLR results, and other laboratory results of patients with sepsis were compared between the ones with and without mortality. 
 Results: Four hundred and eleven patients were included in the study. 55.7% (229/411) of patients with sepsis died and 44.3% (182/411) were discharged alive. NLR, MLR and PLR were higher in the group with mortality compared to the survivor group. The cut-off value for predicting mortality in patients with sepsis was 9.2 for NLR, ≥0.8 for MLR, and ≥187.3 for PLR. The area under the curve (AUC) value for NLR was 0.825, the AUC value for MLR 0.835 and the AUC value for PLR was 0.720.
 Conclusion: High NLR, MLR and PLR values are associated with mortality in sepsis patients hospitalized in ICU, and the most significant parameter for mortality indicator among the three rates was found to be MLR with the highest AUC value.

The role of pattern electroretinograms and optical coherence tomography angiography in the diagnosis of normal-tension glaucoma

In this study, we investigated the correlation between pattern electroretinogram (PERG) and optical coherence tomography angiography (OCTA) parameters for diagnosis in patients with normal-tension glaucoma (NTG). Forty-nine normal individuals (49 eyes) and 60 patients with NTG (60 eyes) were enrolled. OCTA and PERG parameters, such as macular vessel density (VD) and the amplitude of N35–P50 and P50–N95, were measured. Correlation analyses were performed between the parameters, and the area under the curve (AUC) was used to identify their diagnostic ability for NTG. Macular VD and the amplitude of N35–P50 and P50–N95 showed significant differences between the normal individuals and patients with NTG. Correlation between P50 and N95 amplitude and macular VD was significant in the normal and early glaucoma groups. Macular VD showed a higher AUC value (0.730) than that of P50–N95 amplitude (0.645) in the early glaucoma group. In the moderate to severe glaucoma group, the AUC value of the amplitude of P50–N95 (0.907) was higher than that of macular VD (0.876). The results indicate that PERG and OCTA parameters may identify glaucoma in its early stage, based on the severity of glaucomatous damage in patients with NTG.

Prediction of myopia onset with refractive error measured using non-cycloplegic subjective refraction: the WEPrOM Study

Background/aimsTo evaluate the predictive performance of various predictors, including non-cycloplegic refractive error, for risk of myopia onset under pragmatic settings.MethodsThe Wenzhou Medical University Essilor Progression and Onset of Myopia Study...

Comparing the characteristics and predicting the survival of patients with head and neck melanoma versus body melanoma: a population-based study

BackgroundPrevious studies reported cutaneous melanoma in head and neck (HNM) differed from those in other regions (body melanoma, BM). Individualized tools to predict the survival of patients with HNM or BM remain insufficient. We aimed at comparing the characteristics of HNM and BM, developing and validating nomograms for predicting the survival of patients with HNM or BM.MethodsThe information of patients with HNM or BM from 2004 to 2015 was obtained from the Surveillance, Epidemiology, and End Results (SEER) database. The HNM group and BM group were randomly divided into training and validation cohorts. We used the Kaplan-Meier method and multivariate Cox models to identify independent prognostic factors. Nomograms were developed via the rms and dynnom packages, and were measured by the concordance index (C-index), the area under the curve (AUC) of the receiver operating characteristic (ROC) curve and calibration plots.ResultsOf 70,605 patients acquired, 21% had HNM and 79% had BM. The HNM group contained more older patients, male sex and lentigo maligna melanoma, and more frequently had thicker tumors and metastases than the BM group. The 5-year cancer-specific survival (CSS) and overall survival (OS) rates were 88.1 ± 0.3% and 74.4 ± 0.4% in the HNM group and 92.5 ± 0.1% and 85.8 ± 0.2% in the BM group, respectively. Eight variables (age, sex, histology, thickness, ulceration, stage, metastases, and surgery) were identified to construct nomograms of CSS and OS for patients with HNM or BM. Additionally, four dynamic nomograms were available on web. The internal and external validation of each nomogram showed high C-index values (0.785–0.896) and AUC values (0.81–0.925), and the calibration plots showed great consistency.ConclusionsThe characteristics of HNM and BM are heterogeneous. We constructed and validated four nomograms for predicting the 3-, 5- and 10-year CSS and OS probabilities of patients with HNM or BM. These nomograms can serve as practical clinical tools for survival prediction and individual health management.

Prediction of Disease Progression of COVID-19 Based upon Machine Learning.

BackgroundSince December 2019, COVID-19 has spread throughout the world. Clinical outcomes of COVID-19 patients vary among infected individuals. Therefore, it is vital to identify patients at high risk of disease progression.MethodsIn this retrospective, multicenter cohort study, COVID-19 patients from Huoshenshan Hospital and Taikang Tongji Hospital (Wuhan, China) were included. Clinical features showing significant differences between the severe and nonsevere groups were screened out by univariate analysis. Then, these features were used to generate classifier models to predict whether a COVID-19 case would be severe or nonsevere based on machine learning. Two test sets of data from the two hospitals were gathered to evaluate the predictive performance of the models.ResultsA total of 455 patients were included, and 21 features showing significant differences between the severe and nonsevere groups were selected for the training and validation set. The optimal subset, with eleven features in the k–nearest neighbor model, obtained the highest area under the curve (AUC) value among the four models in the validation set. D-dimer, CRP, and age were the three most important features in the optimal-feature subsets. The highest AUC value was obtained using a support vector–machine model for a test set from Huoshenshan Hospital. Software for predicting disease progression based on machine learning was developed.ConclusionThe predictive models were successfully established based on machine learning, and achieved satisfactory predictive performance of disease progression with optimal-feature subsets.

Post-operative procalcitonin and C-reactive protein predict pancreatic fistula after laparoscopic pancreatoduodenectomy

BackgroundClinically relevant pancreatic fistula (CRPF) is a serious complication following laparoscopic pancreaticoduodenectomy (LPD). This study aimed to determine if C-reactive protein (CRP) and procalcitonin (PCT) serum levels could be used as early biomarkers to predict CRPF after LPD.MethodsIn this retrospective study, we collected peri-operative data of patients who underwent LPD between January 2019 and November 2019. We compared serum levels of white blood cells (WBC), CRP, and PCT on post-operative days (POD) 1, 2, 3, 5, and 7 between the CRPF and non-CRPF groups and analyzed the predictive risk factors for CRPF.ResultsAmong the 186 patients included in this study, 18 patients (9.7%) developed CRPF, including 15 and 3 patients with grade B and C fistulas, respectively. The mean WBC, CRP, and PCT levels were higher on most PODs in the CRPF group compared to the non-CRPF group. Receiver operating characteristic (ROC) analysis indicated that CRP levels on POD 2, 5, and 7 can predict CRPF development after LPD, with the area under the curve (AUC) value reaching the highest level on POD 2 (AUC 0.794). PCT levels on POD 2, 3, 5, and 7 were highly predictive of CRPF after LPD. The highest AUC value was achieved on POD 3 [PCT > 2.10 ng/ml (AUC 0.951; sensitivity 88.2%, specificity 92.9%, P < 0.001)].ConclusionsBoth CRP and PCT levels can be used to predict CRPF development after LPD, with PCT having a higher predictive value.

Ecological niche overlap among species of the genus Zaluzania (Asteraceae) from the dry regions of Mexico

Background and aims – The hypothesis of ecological niche conservatism postulates that closely related species share ecologically similar environments; that is, they tend to maintain the characteristics of their fundamental niche over time. The objective of this study is to evaluate the similarity and equivalence of the ecological niches among species of the genus Zaluzania (Asteraceae), characteristic of the Mexican arid and semi-arid regions, to infer their potential niche conservatism. Methods – Based on critically reviewed herbarium occurrence data, potential distribution models for eight species of Zaluzania were generated using the Maxent algorithm. The overlap between potential distribution areas was then evaluated using equivalence and ecological niche parameters implemented in the ENMTools software; for this we quantified the degree of overlap and similarity between the niches using the equivalence (D) and similarity (I) parameters.Key results – The resulting models show that species display areas of high suitability along the Mexican dry regions, as well as overlapping heterogeneous values. All models showed high AUC (Area Under the Curve) values (&gt; 0.8). The D and I values between each pair of species showed low values of overlap.Conclusions – Each species of the genus shows a fundamental niche distinct from their sister species. The genus thus offers an example of niche divergence among species, with each one adapting to different environmental pressures. Our results do not support the hypothesis of niche conservatism in the genus, suggesting that the species evolved in divergent environments.

Ecological niche overlap among species of the genus Zaluzania (Asteraceae) from the dry regions of Mexico

Background and aims – The hypothesis of ecological niche conservatism postulates that closely related species share ecologically similar environments; that is, they tend to maintain the characteristics of their fundamental niche over time. The objective of this study is to evaluate the similarity and equivalence of the ecological niches among species of the genus Zaluzania (Asteraceae), characteristic of the Mexican arid and semi-arid regions, to infer their potential niche conservatism. Methods – Based on critically reviewed herbarium occurrence data, potential distribution models for eight species of Zaluzania were generated using the Maxent algorithm. The overlap between potential distribution areas was then evaluated using equivalence and ecological niche parameters implemented in the ENMTools software; for this we quantified the degree of overlap and similarity between the niches using the equivalence (D) and similarity (I) parameters.Key results – The resulting models show that species display areas of high suitability along the Mexican dry regions, as well as overlapping heterogeneous values. All models showed high AUC (Area Under the Curve) values (&gt; 0.8). The D and I values between each pair of species showed low values of overlap.Conclusions – Each species of the genus shows a fundamental niche distinct from their sister species. The genus thus offers an example of niche divergence among species, with each one adapting to different environmental pressures. Our results do not support the hypothesis of niche conservatism in the genus, suggesting that the species evolved in divergent environments.

Comparison of the different voxel sizes in the estimation of peri-implant fenestration defects using cone beam computed tomography: an ex vivo study

BackgroundTo examine the influence of voxel sizes to detect of peri-implant fenestration defects on cone beam computed tomography (CBCT) images.Materials and methodsThis study performed with three sheep heads both maxilla and mandible and two types of dental implant type 1 zirconium implant (Zr40) (n = 6) and type 2 titanium implant (Ti22) (n = 10). A total of 14 peri-implant fenestrations (8 buccal surfaces, 6 palatal/lingual surface) were created while 18 surfaces (8 buccal, 10 palatal/lingual) were free of fenestrations. Three observers have evaluated the images of fenestration at each site. Images obtained with 0.75 mm3, 0.100 mm3, 0.150 mm3, 0.200 mm3, and 0.400 mm3 voxel sizes. For intra- and inter-observer agreements for each voxel size, Kappa coefficients were calculated.ResultsIntra- and inter-observer kappa values were the highest for 0.150 mm3, and the lowest in 0.75 mm3 and 0.400 mm3 voxel sizes for all types of implants. The highest area under the curve (AUC) values were found higher for the scan mode of 0.150 mm3, whereas lower AUC values were found for the voxel size for 0.400 mm3. Titanium implants had higher AUC values than zirconium with the statistical significance for all voxel sizes (p ≤ 0.05).ConclusionA voxel size of 0.150 mm3 can be used to detect peri-implant fenestration bone defects. CBCT is the most reliable diagnostic tool for peri-implant fenestration bone defects.

A discovery of screening markers for rheumatoid arthritis by liquid chromatography mass spectrometry: A metabolomic approach

This study aimed to discover serum metabolite biomarkers for potential use in screening for rheumatoid arthritis (RA). The sera from 43 healthy controls (HCs) and 49 RA patients were globally analyzed using high-performance liquid chromatography- tandem mass spectrometry. Molecular features (MFs) from samples were analyzed using volcano plots, partial least squares discriminant analysis, and variable importance in projection scores to select candidates. The spectra of candidate MFs were matched with the METLIN database. We confirmed the association between candidates and RA and analyzed the receiver-operating characteristic (ROC) curves. We selected a total of 57 candidate MFs that had a fold change ≥1.5, P value ≤.05, and over 80% of frequency. Among them, 18 MFs were identified as metabolites with the METLIN database. Six metabolites (dehydroepiandrosterone sulfate, androsterone sulfate, γ-linolenic acid, 9[E],11[E]-conjugated linoleic acid, docosahexaenoic acid, and docosapentaenoic acid [22n-3]) out of the 18 were associated with mechanisms of RA and were selected as final candidates. ROC curve analysis revealed their area under the curve (AUC) values were all above 0.75 and the combined AUC of the six candidates was 0.89. Using six candidates as a marker set showed potential in distinguishing RA patients from HCs, based on high AUC values. Therefore, we propose that a marker set of these six candidates has potential clinical application in RA screening.

Walking Characteristics In Individuals With Stroke Differ Based On Walking Speed, Endurance And Daily Steps

PURPOSE: Walking is an ideal means of obtaining physical activity, yet people with stroke take few daily steps. The purpose of this study was to examine how the walking characteristics of bouts per day, maximum steps per bout and time spent walking differ between individuals with various walking speeds, walking endurance and daily steps. Additionally, we aimed to identify cutoff values for differentiating active and inactive ambulators (i.e. those who do and do not achieve physical activity guidelines through walking). METHODS: Stepping data from 252 individuals with chronic stroke (>6 months) with mean age of 63 (13) years and step count of 4,277 (3,064) steps per day were analyzed. Individuals were placed into previously established levels of ambulation (i.e. household ambulators, limited community and unlimited community ambulators), based on walking speed, walking endurance and daily steps (via two days of StepWatch activity monitoring). Differences in walking characteristics were assessed between ambulation levels (e.g. household vs. community ambulators). Linear regression determined which characteristics best predicted daily step counts. Receiver Operating Characteristic (ROC) curves and area under the curve (AUC) determined which variable was most accurate in classifying active (>5,500 steps) and inactive (<5,500 steps) individuals. RESULTS: Regardless of categorization by walking speed, walking endurance or daily steps, household ambulators had significantly fewer bouts per day, maximum steps per bout and time spent walking compared to both limited and unlimited community ambulators (p = <0.001). Only 81 (32%) participants obtained >5,500 steps per day. The two highest AUC values were 0.91 (95% CI 0.88, 0.95) for maximum steps per bout and 0.83 (95% CI 0.78, 0.88) for bouts per day. Cutoff values of 648 maximum steps per bout or 53 bouts were used to differentiate active and inactive ambulators. CONCLUSIONS: Walking characteristics differed based on an individual’s walking speed, walking endurance and daily steps. Differences in daily steps between household and community ambulators are largely due to shorter and fewer walking bouts. Interventions aimed at improving walking after stroke should promote increased walking bouts of any length to increase physical activity after stroke.

Can neutrophil-to-lymphocyte and monocyte-to-lymphocyte ratios be useful markers for predicting missed abortion in the first trimester of pregnancy?

To explore whether neutrophil-to-lymphocyte ratio (NLR) and monocyte-to-lymphocyte ratio (MLR) could predict missed abortion (MA) at 7-13 week's gestation. A total of 363 women with a diagnosis of MA and 232 women with normal pregnancy at 7-13 week's gestation, who visited our hospital from June 2012 to May 2018 were retrospectively analyzed. At 7 week's gestation, total and differential leukocyte counts, NLR and MLR were compared between women with MA (n = 69) and normal controls (n = 53). The receiver operating characteristic curve was used to select the optimal indicator and its cut-off point. The role of the optimal indicator was further assessed at 8-13 week's gestation. The mean white blood cell counts, the mean neutrophil counts, the median monocyte counts, the mean NLR and the median MLR in women with MA were significantly lower than those in normal controls (P < 0.05, respectively).The neutrophil counts had the highest area under the curve (AUC) value of 0.772 (95% confidence interval 0.675-0.869) with a cut-off value of 4.870 × 109 /L, and the sensitivity was 72.46%, the specificity was 69.81%, positive predictive value was 75.76%, and negative predictive value was 66.07%. In addition, the neutrophil counts were also significantly lower in MA groups than those in normal controls at 8-13 week's gestation, and all had the highest AUC values. Neutrophil counts may predict MA in the first trimester of pregnancy, which may provide a promising marker to diagnose missed abortion as early as 7 week's gestation.

Complement and other immune-related factors in cervicovaginal fluid associated with intra-amniotic infection/inflammation and spontaneous preterm delivery in women with preterm labor.

To investigate whether complement and other immune-related proteins in cervicovaginal fluid (CVF) can predict intra-amniotic infection and/or inflammation (IAI) and spontaneous preterm delivery (SPTD, < 34.0weeks) in women with preterm labor (PTL) and to compare the predictive abilities of these biomarkers with that of amniotic fluid (AF) white blood cells (WBCs). We designed a retrospective cohort study of 145 women with PTL at 23.0-33.6weeks who underwent amniocentesis. AF was cultured and assayed for WBC count and interleukin-6 (IL-6). CVF samples were obtained at the time of amniocentesis. CVF was assayed for complement C3a and C5a, IGFBP-1, and MMP-9 by ELISA. In the multivariate analysis, elevated CVF levels of C5a and IGFBP-1 were significantly associated with IAI and SPTD at < 34weeks, while those of C3a were associated with IAI, but not SPTD, even after adjusting for other baseline confounders. For C3a, C5a, and IGFBP-1 in the CVF, area under the curve (AUC) values were statistically similar to that of AF WBCs for detecting IAI, whereas these CVF biomarkers had similar or higher AUC values than AF WBCs for predicting SPTD at < 34weeks. However, univariate analysis showed no significant correlation between high CVF MMP-9 and IAI or SPTD at < 34weeks. In women with PTL, the CVF levels of C3a, C5a, and IGFBP-1 may be useful as novel non-invasive predictors of IAI and SPTD at < 34weeks. These biomarkers (especially IGFBP-1) have similar or better diagnostic performance compared to AF WBCs.

Evaluation of multiple screening methods for cervical cancers in rural areas of Xinjiang, China.

This study is to evaluate the screening methods of cervical cancers for rural females in Kash bachu, Xinjiang, China.A total number of 3000 married females were surveyed, and subjected to the gynecological examination. In these subjects, 1993 females received the careHPV (human papillomavirus) test, while 1007 females underwent the visual inspection with acetic acid (VIA) and visual inspection with Lugol's iodine (VILI). The subjects positive for careHPV detection were subjected to Cervista, Cobas 4800, and Aptima HPV detection, and Thinprep Cytologic Test (TCT). The subjects positive for 1 detection only received colposcopy cervical biopsy.A total of 569 subjects received the cervical biopsy, and the positive rate was 2.3% (69/3000), while the detection rate for CIN (cervical intraepithelial neoplasia) II and above levels was 1.13% (34/3000). Receiver operator characteristic (ROC) curve analysis showed that, the area under the curve (AUC) value for the careHPV test was 0.671, which was higher than the VIA/VILI (0.619), suggesting higher diagnostic value for the careHPV test. For the Cervista, Cobas 4800, Aptima HPV detection, and TCT methods, the highest AUC value was observed for the TCT method, indicating that the TCT method is the most valuable for the cervical cancer screening.The diagnostic value of careHPV test is superior to the VIA/VILA detection method. The TCT method has the greatest value for the cervical cancer screening. The Cervista HPV detection method should be considered where the conditions are limited.

Clinical validation of three cardiovascular magnetic resonance techniques to measure strain and torsion in patients with suspected coronary artery disease

BackgroundSeveral cardiovascular magnetic resonance (CMR) techniques can measure myocardial strain and torsion with high accuracy. The purpose of this study was to compare displacement encoding with stimulated echoes (DENSE), tagging and feature tracking (FT) for measuring circumferential and radial myocardial strain and myocardial torsion in order to assess myocardial function and infarct scar burden both at a global and at a segmental level.Method116 patients with a high likelihood of coronary artery disease (European SCORE > 15%) underwent CMR examination including cine images, tagging, DENSE and late gadolinium enhancement (LGE) in the short axis direction. In total, 97 patients had signs of myocardial disease and 19 had no abnormalities in terms of left ventricular (LV) wall mass index, LV ejection fraction, wall motion, LGE or a history of myocardial infarction. Thirty-four patients had myocardial infarct scar with a transmural LGE extent (transmurality) that exceeded 50% of the wall thickness in at least one segment. Global circumferential strain (GCS) and global radial strain (GRS) was analyzed using FT of cine loops, deformation of tag lines or DENSE displacement.ResultsDENSE and tagging both showed high sensitivity (82% and 71%) at a specificity of 80% for the detection of segments with > 50% LGE transmurality, and receiver operating characteristics (ROC) analysis showed significantly higher area under the curve-values (AUC) for DENSE (0.87) than for tagging (0.83, p < 0.001) and FT (0.66, p = 0.003). GCS correlated with global LGE when determined with DENSE (r = 0.41), tagging (r = 0.37) and FT (r = 0.15). GRS had a low but significant negative correlation with LGE; DENSE r = − 0.10, FT r = − 0.07 and tagging r = − 0.16. Torsion from DENSE and tagging had a weak correlation (− 0.20 and − 0.22 respectively) with global LGE.ConclusionCircumferential strain from DENSE detected segments with > 50% scar with a higher AUC than strain determined from tagging and FT at a segmental level. GCS and torsion computed from DENSE and tagging showed similar correlation with global scar size, while when computed from FT, the correlation was lower.

Modeling flood susceptibility using data-driven approaches of naïve Bayes tree, alternating decision tree, and random forest methods

Floods are one of the most devastating types of disasters that cause loss of lives and property worldwide each year. This study aimed to evaluate and compare the prediction capability of the naïve Bayes tree (NBTree), alternating decision tree (ADTree), and random forest (RF) methods for the spatial prediction of flood occurrence in the Quannan area, China. A flood inventory map with 363 flood locations was produced and partitioned into training and validation datasets through random selection with a ratio of 70/30. The spatial flood database was constructed using thirteen flood explanatory factors. The probability certainty factor (PCF) method was used to analyze the correlation between the factors and flood occurrences. Consequently, three flood susceptibility maps were produced using the NBTree, ADTree, and RF methods. Finally, the area under the curve (AUC) and statistical measures were used to validate the flood susceptibility models. The results indicated that the RF method is an efficient and reliable model in flood susceptibility assessment, with the highest AUC values, positive predictive rate, negative predictive rate, sensitivity, specificity, and accuracy for the training (0.951, 0.892, 0.941, 0.945, 0.886, and 0.915, respectively) and validation (0.925, 0.851, 0.938, 0.945, 0.835, and 0.890, respectively) datasets.

Building Risk Prediction Models for Type 2 Diabetes Using Machine Learning Techniques.

IntroductionAs one of the most prevalent chronic diseases in the United States, diabetes, especially type 2 diabetes, affects the health of millions of people and puts an enormous financial burden on the US economy. We aimed to develop predictive models to identify risk factors for type 2 diabetes, which could help facilitate early diagnosis and intervention and also reduce medical costs.MethodsWe analyzed cross-sectional data on 138,146 participants, including 20,467 with type 2 diabetes, from the 2014 Behavioral Risk Factor Surveillance System. We built several machine learning models for predicting type 2 diabetes, including support vector machine, decision tree, logistic regression, random forest, neural network, and Gaussian Naive Bayes classifiers. We used univariable and multivariable weighted logistic regression models to investigate the associations of potential risk factors with type 2 diabetes.ResultsAll predictive models for type 2 diabetes achieved a high area under the curve (AUC), ranging from 0.7182 to 0.7949. Although the neural network model had the highest accuracy (82.4%), specificity (90.2%), and AUC (0.7949), the decision tree model had the highest sensitivity (51.6%) for type 2 diabetes. We found that people who slept 9 or more hours per day (adjusted odds ratio [aOR] = 1.13, 95% confidence interval [CI], 1.03–1.25) or had checkup frequency of less than 1 year (aOR = 2.31, 95% CI, 1.86–2.85) had higher risk for type 2 diabetes.ConclusionOf the 8 predictive models, the neural network model gave the best model performance with the highest AUC value; however, the decision tree model is preferred for initial screening for type 2 diabetes because it had the highest sensitivity and, therefore, detection rate. We confirmed previously reported risk factors and also identified sleeping time and frequency of checkup as 2 new potential risk factors related to type 2 diabetes.

CT-Based Radiomics Can Improve the Prediction of Recurrent Prostate Cancer-Positive Lymph Nodes

In patients with prostate cancer (PC), PSMA-PET/CT is currently the most preferred method for lymph node (LN) staging in certain regions. Globally, CT-based LN evaluation using the short diameter axis is still often used to determine the node status. Optimization of treatment planning according to the affected LNs is crucial for a successful radiation therapy (RT). The purpose of this retrospective study is to develop CT-based radiomic models for an optimized CT-based LN classification using LNs from recurrent PC patients who underwent PSMA-radio-guided surgery with confirmed histology. A total of 160 LNs with confirmed histology (114 positive and 46 negative) from a retrospective cohort of 87 recurrent PC patients were randomized to a training cohort (105 LNs) and a testing cohort (55 LNs). Suspicious LNs on (68Ga-PSMA-11)-PET/CT and adjacent non-diseased templates, were selectively surgically resected using a gamma probe following intravenous application of radioactively labeled PSMA with 99mTechnetium (73 patients) and 111Indium (14 patients). Histologically confirmed LNs were manually segmented using contrast-enhanced diagnostic CT datasets. 86 radiomic parameters including shape, intensity, and local binary pattern (LBP) features were extracted using a software package in python. On the training cohort, the least absolute shrinkage and selection operator (LASSO) algorithm was used to develop three radiomic models: "shape/intensity,” "LBP,” and a combined model. Logistic regression models were developed for clinical parameters including LN short diameter, LN volume, and a CT-based expert evaluation (CT-expert rating). Prediction performance stability was evaluated by bootstrapping (1000 fold). The resulting models were validated using the testing cohort. The LBP radiomic model showed the best prediction performance (area under the curve (AUC)-Training: 0.870, AUC-Testing: 0.836) in comparison to the shape/intensity radiomic model (AUC-Training: 0.838, AUC-Testing: 0.794). The combined radiomic model did not increase the discriminative performance further. Clinical parameters showed overall a lower predictive performance. Clinically, LN short diameter had the highest AUC values (AUC-Training: 0.750, AUC-Testing: 0.732) followed by LN volume (AUC-Training: 0.720, AUC-Testing: 0.571) and CT-expert rating (AUC-Training: 0.650, AUC-Testing: 0.644). The LBP radiomics model performed significantly better than LN short diameter (p=0.0029), LN volume (p=0.0285) and CT-expert rating (p<0.00001). We first developed a CT-based radiomic model to improve the detection of metastatic LNs in recurrent PC patients. The model was successfully internally validated. Such a model could be used to improve radiation therapy planning in case of unavailability of PSMA-PET imaging techniques.

Plasma Exosomal miR-450b-5p as a Possible Biomarker and Therapeutic Target for Transient Ischaemic Attacks in Rats.

Transient ischaemic attack (TIA) and cerebral infarction are difficult to identify within the thrombolytic time window. Blood markers are efficient, economical and noninvasive and can be beneficial in the diagnosis of many diseases. Plasma exosomal biomarkers are rarely reported in TIA. Exosomal microRNAs (miRNAs) were extracted from plasma and cerebrospinal fluid after middle cerebral artery occlusion (MCAo) in rats (0min, 5min, 10min, 2h). Deep sequencing was used to detect exosomal miRNAs in rat plasma and confirm significant differentially expressed miRNAs. Polymerase chain reaction (PCR) was used to detect the differentially expressed miRNAs in plasma and cerebrospinal fluid. Exosomal miRNAs with the same expression trends in plasma and cerebrospinal fluid were selected, and bioinformatics analysis was then carried out. Finally, the area under the curve (AUC) of the receiver operating characteristic (ROC) curve was determined to assess the diagnostic accuracy of miRNAs for TIA in rats. First, high-throughput sequencing was used to detect the expression level of plasma exosome miRNA, and rno-miR-450b-5p with a decreasing expression level was screened. Second, the expression levels of exosomal miRNAs were verified in cerebrospinal fluid and plasma samples by PCR, and the results indicated that exosomal rno-miR-450b-5p was similarly expressed in cerebrospinal fluid and plasma. ROC analysis showed high AUC values for rno-miR-450b-5p (0.880) in the 10min ischaemia rats compared with the control rats. Finally, bioinformatic analysis indicated that exosomal rno-miR-450b-5p may be involved in cerebral ischaemia. Plasma exosomal rno-miR-450b-5p has a high diagnostic value and may become a therapeutic target for rat TIA.

Is it possible to detect cerebral dominance via EEG signals by using deep learning?

Each brain hemisphere is dominant for certain functions such as speech. The determination of speech laterality prior to surgery is of paramount importance for accurate risk prediction. In this study, we aimed to determine speech laterality via EEG signals by using noninvasive machine learning techniques. The retrospective study included 67 subjects aged 18–65 years who had no chronic diseases and were diagnosed as healthy based on EEG examination. The subjects comprised 35 right-hand dominant (speech center located in the left hemisphere) and 32 left-hand dominant individuals (speech center located in the right hemisphere). A spectrogram was created for each of the 18 EEG channels by using various Convolutional Neural Networks (CNN) architectures including VGG16, VGG19, ResNet, MobileNet, NasNet, and DenseNet. These architectures were used to extract features from the spectrograms. The extracted features were classified using Support Vector Machines (SVM) and the classification performances of the CNN models were evaluated using Area Under the Curve (AUC). Of all the CNN models used in the study, VGG16 had a higher AUC value (0.83 ± 0.05) in the determination of speech laterality compared to all other models. The present study is a pioneer investigation into the determination of speech laterality via EEG signals with machine learning techniques, which, to our knowledge, has never been reported in the literature. Moreover, the classification results obtained in the study are promising and lead the way for subsequent studies though not practically feasible.