True Class Research Articles

Coal structure identification based on geophysical logging data: Insights from Wavelet Transform (WT) and Particle Swarm Optimization Support Vector Machine (PSO-SVM) algorithms

Coal structure is closely related to microscopic and macroscopic properties of coal, and its accurate identification is of great significance to coalbed methane (CBM) reservoir evaluation, hydraulic fracturing performance and production efficiency prediction. The identification of coal structure based on geophysical logging data has become a popular topic as well as a challenge. The emerging Machine Learning (ML) provides convenience for this issue. Under this background, in the case of the Shanxi Formation No. 3 coal seam and the Taiyuan Formation No. 15 coal seam in Mabidong Block, Qinshui Basin, China, this paper proposed a new coal structure identification method based on geophysical logging data using Wavelet Transform (WT) and Particle Swarm Optimization Support Vector Machine (PSO-SVM) algorithms. Our results showed that the vertical resolution of logging data can be effectively improved when sym5 wavelet basis and third level decomposition were selected in wavelet decomposition, and 4.5 was selected as the weighting coefficient in wavelet reconstruction. The coal structure prediction model based on the logging data processed by WT was established by PSO-SVM algorithm, where PSO was used for parameter optimization (optimal penalty factor C and width parameter σ) and SVM with Radial Basis Function (RBF) kernel was used for model establishment. The Hold-out Cross Validation (HO CV) method was used to test the generalization ability of the prediction model, and the accuracy (ACC) of coal structure prediction in the training set and testing set was 94.26% and 88.46%, respectively. The prediction model was applied to identify the coal structure of two coring wells and the predicted coal structure class was consistent with the true coal structure class, confirming the validity of the model. The coal structure prediction results in the whole study area showed that the tectonic conditions control the coal structure. This work provides new insights for coal structure identification.

Deciphering the Differential Diagnosis of Inherited Versus Acquired Bone Marrow Failure Syndromes: A Large Cohort Study

The wide overlaps in clinical and hematologic features of inherited (iBMF) and acquired (aBMF) bone marrow failures are often the cause of diagnostic difficulties and delay in proper therapeutic approaches. Screening tests for iBMF are complex and expensive and have limited availability. Data-driven machine learning algorithms may reveal useful tools in clinical practice to guide differential diagnosis. We collected data about clinical, laboratory and natural history of 597 patients with BMF (55 iBMF; 542 aBMF). iBMFs were typically younger (30 y, 0.1-71 vs 44 y, 2-85, p <0.001) and presented higher baseline blood counts (reticulocytes 62x10^9/L, 31-205 vs 35x10^9/L, 0-216; neutrophiles 1.4x10^9/L, 0-4.3 vs 0.6x10^9/L, 0-4.6; platelets 93x10^9/L, 7-356 vs 23x10^9/L, 1-355, p <0.001), and more cytogenetic aberrations (32% vs 13%, p 0.002). Conversely, PNH clones were detected and confirmed overtime mainly in aBMF (51% vs 9% p <0.001), with median clone size of 0.6% (range 0.02-3.8) in iBMF. By NGS analysis, no differences in acquired myeloid gene mutations incidence were detected between the groups (25% vs 13%, p 0.1). Evolution to MDS occurred in 20% of iBMF vs 11% of aBMF (p 0.04), and to AML in 3.6% vs 1.5% (p 0.2). Time between diagnosis and evolution was shorter in iBMF (2 y, 0-14 vs 5.4 y, 0.2-32 p 0.01). Survival was similar between the 2 groups with median values of 7 y (95%CI, 6.1-7.6), reaching 8 y (95%CI, 6.9-9.2) in transplanted patients. Factors significantly associated with a better survival were female gender (HR: 0.73, 95%CI: 0.59-0.90, p 0.004), BMT (HR: 0.62, 95%CI: 0.47-0.83, p 0.001), number of therapy lines (HR: 0.67, 95%CI: 0.51-0.87, p 0.003), response to 1 st line (HR: 0.53, 95%CI: 0.41-0.69, p < 0.001), and evolution to PNH (HR: 0.61, 95%CI: 0.40-0.96, p 0.03). Conversely, a lower survival was associated with older age: patients aged 31-60 had 2-fold risk of death compared to those under 30 (HR 1.8, 95%CI 1.15-2.75, p <0.001); patients over 60 had a 3.2 times higher risk compared to under 30 (HR 3.2, 95%CI 2.35-4.45, p <0.001). Presence of trilineage cytopenia also predicted poorer survival (HR: 2.92, 95%CI: 2.25-3.79, p <0.001 versus single- or bi-lineage cytopenias). No clear association with survival was found regarding karyotype and presence of mutations at myeloid NGS panel (Table 1). Finally, via a 10-folds cross-validation, we built a multivariate logistic model to predict the probability of having an iBMF. A dataset composed of 139 patients (120 aBMF and 19 iBMF) was used to train and validate the model, while subjects with missing information were excluded from the analysis. The obtained classifier was tested on an independent test set, resulting in 96% (81-100) accuracy, with 26/27 patients correctly assigned to their true class, 100% sensitivity and 96% specificity and having area under the ROC curve (AUC) equal to 0.97. Factors showing an impact on the predictive model were as follows: higher Hb levels (OR 2.35, range 0.98-5.60, p 0.05), reticulocytes (OR 1.01, range 0.99-1.04, p 0.29), platelets (OR 1.05, range 1.01-1.09, p 0.005), and erythropoietin (OR 1.0, range 1.00-1.01, p 0.009) were associated with IBMF; whilst older age (OR 0.91, range 0.84-0.99, p 0.03), normal NGS myeloid panel (OR 0.09, range 0.01-1.52, p 0.09), and the presence of a PNH clone (OR 0.46, range 0.04-4.83, p 0.52) showed a negative association (Figure 1). In conclusion, iBMF and aBMF differ in clinical presentation and risk of clonal evolution, and deserve specific treatments (IST in aBMF vs BMT and androgens in iBMF) and special considerations (i.e. choice of BMT donor and conditioning in iBMF). The proposed model includes few and routinely available variables and may be a practical tool to be implemented in the clinic, though it will require validation on an independent cohort to confirm its feasibility and prediction accuracy.

The Performance of Flash Replenishment Contrast-Enhanced Ultrasound for the Qualitative Assessment of Kidney Lesions in Patients with Chronic Kidney Disease.

We investigated the accuracy of CEUS for characterizing cystic and solid kidney lesions in patients with chronic kidney disease (CKD). Cystic lesions are assessed using Bosniak criteria for computed tomography (CT) and magnetic resonance imaging (MRI); however, in patients with moderate to severe kidney disease, CT and MRI contrast agents may be contraindicated. Contrast-enhanced ultrasound (CEUS) is a safe alternative for characterizing these lesions, but data on its performance among CKD patients are limited. We performed flash replenishment CEUS in 60 CKD patients (73 lesions). Final analysis included 53 patients (63 lesions). Four readers, blinded to true diagnosis, interpreted each lesion. Reader evaluations were compared to true lesion classifications. Performance metrics were calculated to assess malignant and benign diagnoses. Reader agreement was evaluated using Bowker's symmetry test. Combined reader sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for diagnosing malignant lesions were 71%, 75%, 45%, and 90%, respectively. Sensitivity (81%) and specificity (83%) were highest in CKD IV/V patients when grouped by CKD stage. Combined reader sensitivity, specificity, PPV, and NPV for diagnosing benign lesions were 70%, 86%, 91%, and 61%, respectively. Again, in CKD IV/V patients, sensitivity (81%), specificity (95%), and PPV (98%) were highest. Inter-reader diagnostic agreement varied from 72% to 90%. In CKD patients, CEUS is a potential low-risk option for screening kidney lesions. CEUS may be particularly beneficial for CKD IV/V patients, where kidney preservation techniques are highly relevant.

A Review of the F-Measure: Its History, Properties, Criticism, and Alternatives

Methods to classify objects into two or more classes are at the core of various disciplines. When a set of objects with their true classes is available, a supervised classifier can be trained and employed to decide if, for example, a new patient has cancer or not. The choice of performance measure is critical in deciding which supervised method to use in any particular classification problem. Different measures can lead to very different choices, so the measure should match the objectives. Many performance measures have been developed, and one of them is the F-measure, the harmonic mean of precision and recall. Originally proposed in information retrieval, the F-measure has gained increasing interest in the context of classification. However, the rationale underlying this measure appears weak, and unlike other measures, it does not have a representational meaning. The use of the harmonic mean also has little theoretical justification. The F-measure also stresses one class, which seems inappropriate for general classification problems. We provide a history of the F-measure and its use in computational disciplines, describe its properties, and discuss criticism about the F-Measure. We conclude with alternatives to the F-measure, and recommendations of how to use it effectively.

A hybrid classification method via keywords screening and attention mechanisms in extreme short text

Short text classification has provoked a vast amount of attention and research in recent decades. However, most existing methods only focus on the short texts that contain dozens of words like Twitter and Microblog, while pay far less attention to the extreme short texts like news headline and search snippets. Meanwhile, contemporary short text classification methods that extend the features via external knowledge sources always introduce lots of useless concepts, which may be detrimental to classification performance. Moreover, unlike traditional short text classification methods, the classification results of extreme short texts are often determined by a few even one or two keywords. To address these problems, we propose a novel hybrid classification method via Keywords Screening and Attention Mechanisms in extreme short text, called KSAM. More specifically, firstly, the attention-based BiLSTM is introduced in our method to enhance the role of keywords. Secondly, we screen the keywords in the extreme short text for obtaining the true class label, and the concepts concerning the keywords are retrieved from external open knowledge sources like DBpedia. Thirdly, the attention mechanisms are introduced to acquire the weight of these retrieved concepts. Finally, conceptual information is utilized to assist the classification of the extreme short text. Extensive experiments have demonstrated the effectiveness of our method compared to other state-of-the-art methods.

TactONet: Tactile Ordinal Network Based on Unimodal Probability for Object Hardness Classification

Hardness is one of the most critical tactile properties for robots to recognize objects. Machine learning methods have shown superior performance in object hardness classification. However, existing machine learning methods for tactile hardness classification cannot use the ordinal information between hardness classes because the one-hot encoding only cares about the correct class and ignores the inter-class relationship. To solve this problem, we propose to generalize the one-hot encoding using unimodal distributions including the Poisson and binomial distributions for tactile ordinal classification problems, resulting in two tactile ordinal networks (TacONet): TacONet-p and TacONet-b. Furthermore, we collect a tactile hardness dataset on the silicone samples with three different shapes (Shapes A, B, C), and each shape samples have thirteen hardness classes ranging from 0A (Shore A scale) to 60A at 5A intervals. We validate the resulting method for tactile hardness classification using a real robot. Experimental results demonstrate that compared with state-of-the-art methods, the proposed method achieves better classification performance in terms of accuracy and quadratic weighted kappa (QWK) on the tactile hardness dataset, reaching a classification accuracy up to 99.5% and a QWK up to 99.9% on Shape C. <i>Note to Practitioners</i>&#x2014;In the field of robotics tactile recognition, hardness classification is one of the most important and common tasks for robots to accurately recognize objects, particularly when the environment is dark or visual sensors are not working. In this paper, we propose a novel tactile ordinal network for tactile hardness classification tasks. The existing machine learning models for tactile hardness classification are trained by minimizing the cross-entropy loss between predicted vectors and one-hot encoding vectors of true classes, which makes the models only care about the correct classes and ignores the inter-class relationship of hardness classes. In other words, these models have the same probability to misclassify a hardness class with any other hardness class. To tackle this problem, we propose to generalize the one-hot encoding method using a unimodal distribution method to encode the true classes. The unimodal distribution encoding vectors can make the model learn the ordinal information between classes. It is proved that the proposed method is able to effectively improve the classification accuracy and QWK in a tactile hardness classification task.

J-score: a robust measure of clustering accuracy.

Clustering analysis discovers hidden structures in a data set by partitioning them into disjoint clusters. Robust accuracy measures that evaluate the goodness of clustering results are critical for algorithm development and model diagnosis. Common problems of clustering accuracy measures include overlooking unmatched clusters, biases towards excessive clusters, unstable baselines, and difficulties of interpretation. In this study, we presented a novel accuracy measure, J-score, to address these issues. Given a data set with known class labels, J-score quantifies how well the hypothetical clusters produced by clustering analysis recover the true classes. It starts with bidirectional set matching to identify the correspondence between true classes and hypothetical clusters based on Jaccard index. It then computes two weighted sums of Jaccard indices measuring the reconciliation from classes to clusters and vice versa. The final J-score is the harmonic mean of the two weighted sums. Through simulation studies and analyses of real data sets, we evaluated the performance of J-score and compared with existing measures. Our results show that J-score is effective in distinguishing partition structures that differ only by unmatched clusters, rewarding correct inference of class numbers, addressing biases towards excessive clusters, and having a relatively stable baseline. The simplicity of its calculation makes the interpretation straightforward. It is a valuable tool complementary to other accuracy measures. We released an R/jScore package implementing the algorithm.

Automatic Identification of Fake Profiles in Online Social Networks

In the modern era, the social life of individuals has shifted towards becoming connected with online informal communities. These communities have revolutionized the way we pursue our social life, making it easier to connect with them and keep up with their updates. However, with the rapid growth of these communities, there has been a rise in issues such as false profiles and online bullying. There is currently no practical solution to address these issues. To address this issue, we have developed a system with which the programmed identification of false profiles is feasible and efficient. This system uses order methods such as support vector machine, nave bayes and decision trees to classify the profiles into false or true classes. As this is a software-based recognition method, it can be easily connected to online informal organizations that have a large number of profiles whose profiles cannot be physically inspected. In today’s world, users are bombarded with unnecessary information during surfing, which are posted by false users. Research has shown that 20% to 40% of profiles in social networks like Facebook are fake profiles. This detection of fake profiles leads to solution using frameworks. Keywords: Online Social Network, Fake Profiles, Classification, Natural Language Processing.

K-Fold Cross-Validation Can Significantly Over-Estimate True Classification Accuracy in Common EEG-Based Passive BCI Experimental Designs: An Empirical Investigation.

In passive BCI studies, a common approach is to collect data from mental states of interest during relatively long trials and divide these trials into shorter "epochs" to serve as individual samples in classification. While it is known that using k-fold cross-validation (CV) in this scenario can result in unreliable estimates of mental state separability (due to autocorrelation in the samples derived from the same trial), k-fold CV is still commonly used and reported in passive BCI studies. What is not known is the extent to which k-fold CV misrepresents true mental state separability. This makes it difficult to interpret the results of studies that use it. Furthermore, if the seriousness of the problem were clearly known, perhaps more researchers would be aware that they should avoid it. In this work, a novel experiment explored how the degree of correlation among samples within a class affects EEG-based mental state classification accuracy estimated by k-fold CV. Results were compared to a ground-truth (GT) accuracy and to "block-wise" CV, an alternative to k-fold which is purported to alleviate the autocorrelation issues. Factors such as the degree of true class separability and the feature set and classifier used were also explored. The results show that, under some conditions, k-fold CV inflated the GT classification accuracy by up to 25%, but block-wise CV underestimated the GT accuracy by as much as 11%. It is our recommendation that the number of samples derived from the same trial should be reduced whenever possible in single-subject analysis, and that both the k-fold and block-wise CV results are reported.

Improving Uncertainty Quantification of Deep Classifiers via Neighborhood Conformal Prediction: Novel Algorithm and Theoretical Analysis

Safe deployment of deep neural networks in high-stake real-world applications require theoretically sound uncertainty quantification. Conformal prediction (CP) is a principled framework for uncertainty quantification of deep models in the form of prediction set for classification tasks with a user-specified coverage (i.e., true class label is contained with high probability). This paper proposes a novel algorithm referred to as Neighborhood Conformal Prediction (NCP) to improve the efficiency of uncertainty quantification from CP for deep classifiers (i.e., reduce prediction set size). The key idea behind NCP is to use the learned representation of the neural network to identify k nearest-neighbor calibration examples for a given testing input and assign them importance weights proportional to their distance to create adaptive prediction sets. We theoretically show that if the learned data representation of the neural network satisfies some mild conditions, NCP will produce smaller prediction sets than traditional CP algorithms. Our comprehensive experiments on CIFAR-10, CIFAR-100, and ImageNet datasets using diverse deep neural networks strongly demonstrate that NCP leads to significant reduction in prediction set size over prior CP methods.

Classification and feature selection methods based on fitting logistic regression to PU data

In our work, we examine the classification methods where the positive and unlabeled data are considered and where the conditional distribution of the true class label given the feature vector is governed by the model of logistic regression. Our first objective is to compute and compare the selected metrics allowing for the quality assessment of these methods. In this context, we investigate four methods of the posterior probability estimation, where the risk of logistic loss function is optimized: the naive approach, the weighted likelihood approach, as well as the quite recently proposed methods – the joint approach, and the LassoJoint method. The corresponding evaluations are basically performed for 13 machine learning models on some chosen – both low- and high-dimensional – datasets. Some of the mentioned machine learning model schemes have been directly borrowed from literature and some have been obtained through some modifications in the existing procedures. Our second goal is to establish the most stable and efficient approach for the posterior probability estimation. Moreover, we use the AdaSampling scheme for comparison of the considered classification methods. We also conducted comparisons of feature selection procedures – the Mutual Information-Based feature selection method and the LassoJoint approach. The current article is an enhancement of the conference paper Furmańczyk et al. (2022).

Relaxed least square regression with ℓ2,1-norm for pattern classification

This work aims to address two issues that often exist in least square regression (LSR) models for classification tasks, which are (1) learning a compact projection matrix for feature selection and (2) adopting relaxed regression targets. To this end, we first propose a sparse regularized LSR framework for feature selection by introducing the [Formula: see text] regularizer. Second, we utilize two different strategies to relax the strict regression targets based on the sparse framework. One way is to exploit the [Formula: see text]-dragging technique. Another strategy is to directly learn the labels from the inputs and constrain the distance between true and false classes simultaneously. Hence, more feasible regression schemes are constructed, and the models will be more flexible. Further, efficient iterative methods are derived to optimize the proposed models. Various experiments on image databases intend to manifest our proposed models have outstanding recognition capability compared with many state-of-the-art classifiers.

Radiogenomics of muscle invasive bladder cancer.

e16587 Background: Accurate staging of Muscle Invasive Bladder Cancer (MIBC) is essential to ensure optimal treatment (e.g., radical cystectomy vs. bladder preservation). However, currently about 1/3 of patients are overstaged and 1/3 are understaged. There is a pressing need of accurate noninvasive staging for MIBC to assist clinical decision making. Methods: In this preliminary study, an extensive radiogenomic analysis of MIBC cases was performed to identify potential indicators of MIBC stages, and to automate the staging process. A total of 28 MRI scans and their matched transcriptomic profiles were obtained from subjects diagnosed with MIBC at different stages. In the genomics, a bulk RNA-seq from FFPE tissues was performed that identified 3 major clusters, high, intermediate, and low infiltrating immune cells and fibroblasts. This was followed by a bioinformatic analyses, such as pathway enrichment and gene set enrichment analyses, 15 most significant signatures were identified for stage categorization of MIBC (Fig. 1). In the radiomics, several hundred image-based features from bladder tumor in MRI were extracted and analyzed. The tumors were manually outlined and grouped into lower-stage (stage I and II) and higher-stage (stage III and IV) by the trained radiologists. Using various statistical tests, several radiomic features (e.g., heterogeneity, inverse contrast, entropy) were inferred as significant at discriminating the two groups (Fig. 2). A naïve Bayes machine learning model was then used to integrate such features and perform automated binary classification of cases into their true class. A 4-fold cross validation was performed in which a unique 75% and 25% of 28 cases were used for training and testing of the model respectively. Results: The model performance was assessed in terms of prediction (binary classification) of stages as mean sensitivity, specificity, and accuracy, reaching 84%, 86%, and 86% respectively. Conclusions: The radiogenomics of MIBC provides insight into features associated with cancer stages. Assisted with AI, the discriminative features can accurately stage MIBC. Source of Funding: Cedars-Sinai Cancer Internal Funds: Developmental Funds for Investigator Initiated Trials Using Advanced Imaging Methods.

A Regression Model and a Combination of Academic and Non-Academic Features to Predict Student Academic Performance

Predicting academic performance provides an effective way for students and faculties to monitor their academic progress. The identification of the most significant features was a key outcome of this research, and the college/university databases from online learning platforms are the main academic data sets used to ascertain performance. However, previous research emphasized the addition of other significant features in the prediction of academic performance. Universities’ organizational features include non-academic essential data used in determining student success, but no research has utilized this data to predict student academic performance. Generally, to evaluate binary classification, the relationship between the predicted classifications and the true classifications is analyzed, this approach can lead to the loss of important information from the data. Furthermore, to avoid such loss, this research proposes a regression model by comparing six regression algorithms, and combining academic and non-academic features for prediction student academic performance. Among the various models examined, the gradient-boosted trees regression model demonstrated the lowest error rate. The proposed features have been observed to have a significant impact on academic performance, making them suitable for use in predictions.

An automated method to discover true events and classification of intracellular Ca2+ profiles for endothelium in situ injury assay.

Introduction: Endothelial cells (ECs), being located at the interface between flowing blood and vessel wall, maintain cardiovascular homeostasis by virtue of their ability to integrate chemical and physical cues through a spatio-temporally coordinated increase in their intracellular Ca2+ concentration ([Ca2+]i). Endothelial heterogeneity suggests the existence of spatially distributed functional clusters of ECs that display different patterns of intracellular Ca2+ response to extracellular inputs. Characterizing the overall Ca2+ activity of the endothelial monolayer in situ requires the meticulous analysis of hundreds of ECs. This complex analysis consists in detecting and quantifying the true Ca2+ events associated to extracellular stimulation and classifying their intracellular Ca2+ profiles (ICPs). The injury assay technique allows exploring the Ca2+-dependent molecular mechanisms involved in angiogenesis and endothelial regeneration. However, there are true Ca2+ events of nearly undetectable magnitude that are almost comparable with inherent instrumental noise. Moreover, undesirable artifacts added to the signal by mechanical injury stimulation complicate the analysis of intracellular Ca2+ activity. In general, the study of ICPs lacks uniform criteria and reliable approaches for assessing these highly heterogeneous spatial and temporal events. Methods: Herein, we present an approach to classify ICPs that consists in three stages: 1) identification of Ca2+ candidate events through thresholding of a feature termed left-prominence; 2) identification of non-true events, known as artifacts; and 3) ICP classification based upon event temporal location. Results: The performance assessment of true-events identification showed competitive sensitivity = [0.9995, 0.9831], specificity = [0.9946, 0.7818] and accuracy = [0.9978, 0.9579] improvements of 2x and 14x, respectively, compared with other methods. The ICP classifier enhanced by artifact detection showed 0.9252 average accuracy with the ground-truth sets provided for validation. Discussion: Results indicate that our approach ensures sturdiness to experimental protocol maneuvers, besides it is effective, simple, and configurable for different studies that use unidimensional time dependent signals as data. Furthermore, our approach would also be effective to analyze the ICPs generated by other cell types, other dyes, chemical stimulation or even signals recorded at higher frequency.

Feature selection for high dimensional microarray gene expression data via weighted signal to noise ratio.

Feature selection in high dimensional gene expression datasets not only reduces the dimension of the data, but also the execution time and computational cost of the underlying classifier. The current study introduces a novel feature selection method called weighted signal to noise ratio (WSNR) by exploiting the weights of features based on support vectors and signal to noise ratio, with an objective to identify the most informative genes in high dimensional classification problems. The combination of two state-of-the-art procedures enables the extration of the most informative genes. The corresponding weights of these procedures are then multiplied and arranged in decreasing order. Larger weight of a feature indicates its discriminatory power in classifying the tissue samples to their true classes. The current method is validated on eight gene expression datasets. Moreover, results of the proposed method (WSNR) are also compared with four well known feature selection methods. We found that the (WSNR) outperform the other competing methods on 6 out of 8 datasets. Box-plots and Bar-plots of the results of the proposed method and all the other methods are also constructed. The proposed method is further assessed on simulated data. Simulation analysis reveal that (WSNR) outperforms all the other methods included in the study.

Controlling False-Positives in Automatic Lung Nodule Detection by Adding 3D Cuboid Attention to a Convolutional Neural Network

Background and ObjectiveFalse-positive diagnoses often occur during pulmonary nodule detection, which directly affects the prognosis of patients. The development of intelligent computing and the wide application of deep learning technology have brought about the intelligent detection of lung nodules, which has greatly improved the detection efficiency and detection accuracy. MethodsThis paper explores a convolutional neural network filled with a 3D cuboid attention module for lung nodule detection comprehensively encompassing nodules in the cross, coronal, and sagittal planes of CT sections. First, 3D cuboid attention module extracts image features in order from the plane, space, and volume directions of CT slices (PSV-AM). Second, the model adopts a structure named UPSV-Net of 9 modules to expose the lesion area as much as possible. The PSV-AM is embedded into the network to guide it in extracting the structural features of nodules. At the same time, the features extracted from the shallow and deep layers are associatively concatenated. Thirdly, on the basis of the convolution module shared from the detection branch, deepening the convolution path to form the true and false nodule classification network, which is committed to reducing the false positive rate. ResultsThe model study was validated on the 2016 Lung Nodule Analysis challenge dataset. On the test bench, the detection accuracy is 96.15%, and the sensitivity is 97.6%. Subsequently, the sensitivity to the true and false nodules reached 97.61% and the average false-positive detection rate (FPDR) was 0.893, which was comparable to the diagnosis results of clinicians.

DeepWILD: Wildlife Identification, Localisation and estimation on camera trap videos using Deep learning

Videos and images from camera traps are more and more used by ecologists to estimate the population of species on a territory. It is a laborious work since experts have to analyse massive data sets manually. This takes also a lot of time to filter these videos when many of them do not contain animals or are with human presence. Fortunately, deep learning algorithms for object detection can help ecologists to identify multiple relevant species on their data and to estimate their population. In this study, we propose to go even further by using object detection model to detect, classify and count species on camera traps videos. To this end, we developed a 3-step process: (i) At the first stage, after splitting videos into images, we annotate images by associating bounding boxes to each label thanks to MegaDetector algorithm; (ii) then, we extend MegaDetector based on Faster R-CNN architecture with backbone Inception-ResNet-v2 in order to not only detect the 13 relevant classes but also to classify them; (iii) finally, we design a method to count individuals based on the maximum number of bounding boxes detected. This final stage of counting is evaluated in two different contexts: first including only detection results (i.e. comparing our predictions against the right number of individuals, no matter their true class), then an evolved version including both detection and classification results (i.e. comparing our predictions against the right number in the right class). The results obtained during the evaluation of our model on the test data set are: (i) 73,92% mAP for classification, (ii) 96,88% mAP for detection with a ratio Intersection-Over-Union (IoU) of 0.5 (overlapping ratio between groundtruth bounding box and the detected one), and (iii) 89,24% mAP for detection at IoU = 0.75. Highly represented classes, like humans, have highest values of mAP around 81% whereas less represented classes in the train data set, such as dogs, have lowest values of mAP around 66%. Regarding the proposed counting method, we predicted a count either exact or ± 1 unit for 87% with detection results and for 48% with detection and classification results of our test data set. Our model is also able to detect empty videos. To the best of our knowledge, this is the first study in France about the use of object detection model on a French national park to locate, identify and estimate the population of species from camera trap videos.

Clinical Application of Proximal Arch Cannulation in the Surgical Treatment of Acute Type I Aortic Dissection.

Objective The goal is to determine the best location for inserting a catheter into the aortic arch of patients with a certain type of aortic dissection (DeBakey type I) by analyzing images of the patient's aortic arch before surgery. This analysis will take into account the shape and structure of the patient's aortic arch to find the most optimal location for cannulation. Methods A retrospective analysis was conducted on 100 patients with acute DeBakey type I aortic dissection diagnosed between January 2021 and February 2023, utilizing the Carestream medical imaging software Image Suite V4 (New York, USA). The study included 67 cases that underwent surgery and 33 cases that did not. The study aimed to evaluate the optimal intubation position on the patient's aortic arch by analyzing the true and false lumen classification, true and false lumen area, and hematoma thickness on the patient's aortic arch, as observed in the aortic computed tomography angiography (CTA) conducted upon admission. Results The vascular axis analysis showed a significant difference in the true lumen area among the three regions that were examined (P < 0.001). Zone 1 had a larger true lumen area of 6.40 ± 2.71 cm2 compared to zone 2 with 5.75 ± 2.13 cm2 and zone 3 with 4.85 ± 1.70 cm2, as determined by statistical analysis. In addition, the statistical analysis of hematoma thickness in the three regions where cannulation can be performed revealed a significant difference among the three groups (P = 0.027). Further analysis showed that there was no significant difference between zone 1 and zone 2 (P = 1.000), a significant difference between zone 1 and zone 3 (P < 0.046), and no significant difference between zone 2 and zone 3 (P = 0.080). The difference between zone 1 false lumen thickness of 1.55 ± 0.51 cm and zone 3 false lumen thickness of 1.33 ± 0.55 cm was found to be small. Conclusion Cannulation of the aortic arch is a common strategy used in cardiac surgery. Accurate cannulation is critical to the success of the procedure. The use of CTAprovides valuable guidance for the cannulation procedure. A thorough examination of CTA and precise measurement of relevant parameters can help guide the surgeon to determine the optimal cannulation site. The study found that zone 1 of the aortic arch has the largest area and is the most suitable for cannulation, in accordance with the physiological characteristics and surgical practices of a surgeon. Furthermore, cannulation of the aortic arch has been found to be a safe and effective strategy for cannulation. Overall, careful examination of CTA and accurate measurement of relevant parameters can have a significant guiding effect on the cannulation of the aortic arch, which can lead to improved outcomes in cardiac surgery.

LBA03-05 RADIOGENOMICS OF MUSCLE INVASIVE BLADDER CANCER

You have accessJournal of UrologyCME1 Apr 2023LBA03-05 RADIOGENOMICS OF MUSCLE INVASIVE BLADDER CANCER Touseef Ahmad Qureshi, Xingyu Chen, Yibin Xie, Kaoru Murakami, Toru Sakatani, Makito Miyake, Takashi Kobayashi, Simon Knott, Debiao Li, Charles Rosser, and Hideki Furuya Touseef Ahmad QureshiTouseef Ahmad Qureshi More articles by this author , Xingyu ChenXingyu Chen More articles by this author , Yibin XieYibin Xie More articles by this author , Kaoru MurakamiKaoru Murakami More articles by this author , Toru SakataniToru Sakatani More articles by this author , Makito MiyakeMakito Miyake More articles by this author , Takashi KobayashiTakashi Kobayashi More articles by this author , Simon KnottSimon Knott More articles by this author , Debiao LiDebiao Li More articles by this author , Charles RosserCharles Rosser More articles by this author , and Hideki FuruyaHideki Furuya More articles by this author View All Author Informationhttps://doi.org/10.1097/JU.0000000000003426.05AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract INTRODUCTION AND OBJECTIVE: Accurate staging of Muscle Invasive Bladder Cancer (MIBC) is essential to ensure optimal treatment (e.g., radical cystectomy vs. bladder preservation). However, currently about 1/3 of patients are overstaged and 1/3 are understaged. There is a pressing need of accurate noninvasive staging for MIBC to assist clinical decision making. METHODS: In this preliminary study, an extensive radiogenomic analysis of MIBC cases was performed to identify potential indicators of MIBC stages, and to automate the staging process. A total of 28 MRI scans and their matched transcriptomic profiles were obtained from subjects diagnosed with MIBC at different stages. In the genomics, a bulk RNA-seq from FFPE tissues was performed that identified 3 major clusters, high, intermediate, and low infiltrating immune cells and fibroblasts. This was followed by a bioinformatic analyses, such as pathway enrichment and gene set enrichment analyses, 15 most significant signatures were identified for stage categorization of MIBC (Fig. 1). In the radiomics, several hundred image-based features from bladder tumor in MRI were extracted and analyzed. The tumors were manually outlined and grouped into lower-stage (stage I and II) and higher-stage (stage III and IV) by the trained radiologists. Using various statistical tests, several radiomic features (e.g., heterogeneity, inverse contrast, entropy) were inferred as significant at discriminating the two groups (Fig. 2). A naïve Bayes machine learning model was then used to integrate such features and perform automated binary classification of cases into their true class. A 4-fold cross validation was performed in which a unique 75% and 25% of 28 cases were used for training and testing of the model respectively. RESULTS: The model performance was assessed in terms of prediction (binary classification) of stages as mean sensitivity, specificity, and accuracy, reaching 84%, 86%, and 86% respectively. CONCLUSIONS: The radiogenomics of MIBC provides insight into features associated with cancer stages. Assisted with AI, the discriminative features can accurately stage MIBC. Source of Funding: Cedars-Sinai Cancer Internal Funds: Developmental Funds for Investigator Initiated Trials Using Advanced Imaging Methods © 2023 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 209Issue Supplement 4April 2023Page: e1193 Advertisement Copyright & Permissions© 2023 by American Urological Association Education and Research, Inc.MetricsAuthor Information Touseef Ahmad Qureshi More articles by this author Xingyu Chen More articles by this author Yibin Xie More articles by this author Kaoru Murakami More articles by this author Toru Sakatani More articles by this author Makito Miyake More articles by this author Takashi Kobayashi More articles by this author Simon Knott More articles by this author Debiao Li More articles by this author Charles Rosser More articles by this author Hideki Furuya More articles by this author Expand All Advertisement PDF downloadLoading ...