Pre-trained Convolutional Network Research Articles

No-reference image quality assessment based on visual explanation images and deep transfer learning

Quantifying image quality in the absence of a reference image continues to be a challenge despite the introduction of numerous no-reference image quality assessments (NR-IQA) in recent years. Unlike most existing NRIQA methods, this paper proposes an efficient NR-IQA method based on deep visual interpretations. Specifically, the main components of the proposed method are: i) generating a pseudo-reference image (PRI) for the input distorted images, ii) employing a pretrained convolutional network to extract feature maps from the distorted image and the corresponding PRI, iii) producing visual explanation images (VEIs) by using the feature maps of the distorted image and the corresponding PRI, iv) measuring the similarity between the two VEIs using an image similarity metric, and v) employing a non-linear mapping function for quality score alignment. In our experiments, we evaluated the efficacy of the proposed method across various forms of distortion using four benchmark datasets (LIVE, SIQAD, CSIQ, and TID2013). The proposed approach demonstrates parity with the latest methods, as evidenced by comparisons with both hand-crafted NR-IQA and deep learning-based approaches.

Deep transfer learning for detection of breast arterial calcifications on mammograms: a comparative study

IntroductionBreast arterial calcifications (BAC) are common incidental findings on routine mammograms, which have been suggested as a sex-specific biomarker of cardiovascular disease (CVD) risk. Previous work showed the efficacy of a pretrained convolutional network (CNN), VCG16, for automatic BAC detection. In this study, we further tested the method by a comparative analysis with other ten CNNs.Material and methodsFour-view standard mammography exams from 1,493 women were included in this retrospective study and labeled as BAC or non-BAC by experts. The comparative study was conducted using eleven pretrained convolutional networks (CNNs) with varying depths from five architectures including Xception, VGG, ResNetV2, MobileNet, and DenseNet, fine-tuned for the binary BAC classification task. Performance evaluation involved area under the receiver operating characteristics curve (AUC-ROC) analysis, F1-score (harmonic mean of precision and recall), and generalized gradient-weighted class activation mapping (Grad-CAM++) for visual explanations.ResultsThe dataset exhibited a BAC prevalence of 194/1,493 women (13.0%) and 581/5,972 images (9.7%). Among the retrained models, VGG, MobileNet, and DenseNet demonstrated the most promising results, achieving AUC-ROCs > 0.70 in both training and independent testing subsets. In terms of testing F1-score, VGG16 ranked first, higher than MobileNet (0.51) and VGG19 (0.46). Qualitative analysis showed that the Grad-CAM++ heatmaps generated by VGG16 consistently outperformed those produced by others, offering a finer-grained and discriminative localization of calcified regions within images.ConclusionDeep transfer learning showed promise in automated BAC detection on mammograms, where relatively shallow networks demonstrated superior performances requiring shorter training times and reduced resources.Relevance statementDeep transfer learning is a promising approach to enhance reporting BAC on mammograms and facilitate developing efficient tools for cardiovascular risk stratification in women, leveraging large-scale mammographic screening programs.Key points• We tested different pretrained convolutional networks (CNNs) for BAC detection on mammograms.• VGG and MobileNet demonstrated promising performances, outperforming their deeper, more complex counterparts.• Visual explanations using Grad-CAM++ highlighted VGG16’s superior performance in localizing BAC.Graphical

Fused feature extract method for Φ-OTDR event recognition based on VGGish transfer learning

Thanks to the development of artificial intelligence algorithms, the event recognition of distributed optical fiber sensing systems has achieved high classification accuracy on many deep learning models. However, the large-scale samples required for the deep learning networks are difficult to collect for the optical fiber vibration sensing systems in actual scenarios. An overfitting problem due to insufficient data in the network training process will reduce the classification accuracy. In this paper, we propose a fused feature extract method suitable for the small dataset of Φ-OTDR systems. The high-dimensional features of signals in the frequency domain are extracted by a transfer learning method based on the VGGish framework. Combined with the characteristics of 12 different acquisition points in the space, the spatial distribution characteristics of the signal can be reflected. Fused with the spatial and temporal features, the features undergo a sample feature correction algorithm and are used in a SVM classifier for event recognition. Experimental results show that the VGGish, a pre-trained convolutional network for audio classification, can extract the knowledge features of Φ-OTDR vibration signals more efficiently. The recognition accuracy of six types of intrusion events can reach 95.0% through the corrected multi-domain features when only 960 samples are used as the training set. The accuracy is 17.7% higher than that of the single channel trained on VGGish without fine-tuning. Compared to other CNNs, such as ResNet, the feature extract method proposed can improve the accuracy by at least 4.9% on the same dataset.

Radiomics and Clinical Data for the Diagnosis of Incidental Pulmonary Nodules and Lung Cancer Screening: Radiolung Integrative Predictive Model

IntroductionEarly diagnosis of lung cancer (LC) is crucial to improve survival rates. Radiomics models hold promise for enhancing LC diagnosis. This study assesses the impact of integrating a clinical and a radiomic model based on deep learning to predict the malignancy of pulmonary nodules (PN). MethodologyProspective cross-sectional study of 97 PNs from 93 patients. Clinical data included epidemiological risk factors and pulmonary function tests. The region of interest of each chest CT containing the PN was analysed. The radiomic model employed a pre-trained convolutional network to extract visual features. From these features, 500 with a positive standard deviation were chosen as inputs for an optimised neural network. The clinical model was estimated by a logistic regression model using clinical data. The malignancy probability from the clinical model was used as the best estimate of the pre-test probability of disease to update the malignancy probability of the radiomic model using a nomogram for Bayes’ theorem. ResultsThe radiomic model had a positive predictive value (PPV) of 86%, an accuracy of 79% and an AUC of 0.67. The clinical model identified DLCO, obstruction index and smoking status as the most consistent clinical predictors associated with outcome. Integrating the clinical features into the deep-learning radiomic model achieves a PPV of 94%, an accuracy of 76% and an AUC of 0.80. ConclusionsIncorporating clinical data into a deep-learning radiomic model improved PN malignancy assessment, boosting predictive performance. This study supports the potential of combined image-based and clinical features to improve LC diagnosis.

Detection of activities in bathrooms through deep learning and environmental data graphics images

Automatic detection activities in indoor spaces has been and is a matter of great interest. Thus, in the field of health surveillance, one of the spaces frequently studied is the bathroom of homes and specifically the behaviour of users in the said space, since certain pathologies can sometimes be deduced from it. That is why, the objective of this study is to know if it is possible to automatically classify the main activities that occur within the bathroom, using an innovative methodology with respect to the methods used to date, based on environmental parameters and the application of machine learning algorithms, thus allowing privacy to be preserved, which is a notable improvement in relation to other methods. For this, the methodology followed is based on the novel application of a pre-trained convolutional network for classifying graphs resulting from the monitoring of the environmental parameters of a bathroom. The results obtained allow us to conclude that, in addition to being able to check whether environmental data are adequate for health, it is possible to detect a high rate of true positives (around 80%) in some of the most frequent and important activities, thus facilitating its automation in a very simple and economical way.

Deep Reinforcement Learning for Efficient Digital Pap Smear Analysis

In August 2020, the World Health Assembly launched a global initiative to eliminate cervical cancer by 2030, setting three primary targets. One key goal is to achieve a 70% screening coverage rate for cervical cancer, primarily relying on the precise analysis of Papanicolaou (Pap) or digital Pap smears. However, the responsibility of reviewing Pap smear samples to identify potentially cancerous cells primarily falls on pathologists—a task known to be exceptionally challenging and time-consuming. This paper proposes a solution to address the shortage of pathologists for cervical cancer screening. It leverages the OpenAI-GYM API to create a deep reinforcement learning environment utilizing liquid-based Pap smear images. By employing the Proximal Policy Optimization algorithm, autonomous agents navigate Pap smear images, identifying cells with the aid of rewards, penalties, and accumulated experiences. Furthermore, the use of a pre-trained convolutional neuronal network like Res-Net50 enhances the classification of detected cells based on their potential for malignancy. The ultimate goal of this study is to develop a highly efficient, automated Papanicolaou analysis system, ultimately reducing the need for human intervention in regions with limited pathologists.

Automated Prediction of Osteoarthritis Level in Human Osteochondral Tissue Using Histopathological Images.

Osteoarthritis (OA) is the most common arthritis and the leading cause of lower extremity disability in older adults. Understanding OA progression is important in the development of patient-specific therapeutic techniques at the early stage of OA rather than at the end stage. Histopathology scoring systems are usually used to evaluate OA progress and the mechanisms involved in the development of OA. This study aims to classify the histopathological images of cartilage specimens automatically, using artificial intelligence algorithms. Hematoxylin and eosin (HE)- and safranin O and fast green (SafO)-stained images of human cartilage specimens were divided into early, mild, moderate, and severe OA. Five pre-trained convolutional networks (DarkNet-19, MobileNet, ResNet-101, NasNet) were utilized to extract the twenty features from the last fully connected layers for both scenarios of SafO and HE. Principal component analysis (PCA) and ant lion optimization (ALO) were utilized to obtain the best-weighted features. The support vector machine classifier was trained and tested based on the selected descriptors to achieve the highest accuracies of 98.04% and 97.03% in HE and SafO, respectively. Using the ALO algorithm, the F1 scores were 0.97, 0.991, 1, and 1 for the HE images and 1, 0.991, 0.97, and 1 for the SafO images for the early, mild, moderate, and severe classes, respectively. This algorithm may be a useful tool for researchers to evaluate the histopathological images of OA without the need for experts in histopathology scoring systems or the need to train new experts. Incorporating automated deep features could help to improve the characterization and understanding of OA progression and development.

PD01-01 ENHANCED KINEMATICS AND MULTI-MODALITY LEARNING TO IMPROVE AUTOMATION OF SUTURING SKILLS ASSESSMENT

You have accessJournal of UrologyCME1 Apr 2023PD01-01 ENHANCED KINEMATICS AND MULTI-MODALITY LEARNING TO IMPROVE AUTOMATION OF SUTURING SKILLS ASSESSMENT Loc Trinh, Timothy N. Chu, Ahmed Ghazi, John W. Davis, Brian J. Miles, Clayton Lau, Anand Malpani, Yan Liu, and Andrew J. Hung Loc TrinhLoc Trinh More articles by this author , Timothy N. ChuTimothy N. Chu More articles by this author , Ahmed GhaziAhmed Ghazi More articles by this author , John W. DavisJohn W. Davis More articles by this author , Brian J. MilesBrian J. Miles More articles by this author , Clayton LauClayton Lau More articles by this author , Anand MalpaniAnand Malpani More articles by this author , Yan LiuYan Liu More articles by this author , and Andrew J. HungAndrew J. Hung More articles by this author View All Author Informationhttps://doi.org/10.1097/JU.0000000000003218.01AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract INTRODUCTION AND OBJECTIVE: Previous work have shown that surgeons’ technical skill scores are strong predictors of urinary continence recovery following robot-assisted radical prostatectomy (RARP). Computer vision techniques can help automate the evaluation of suturing skills. In this study, we capture enhanced kinematics (granular motion tracking) of surgical instruments and leverage complementary information between videos and instrument kinematics to improve the performance of automated skill assessments. METHODS: 43 surgeons across 5 institutions completed 5 VR exercises on the Surgical Science™ Flex VR simulator. Instrument kinematic data were captured at 30 Hz, including up to 70 features (XYZ + rotations of each instrument, needle, target, and camera). Videos were segmented into three sub-stitch phases (Needle Handling, Driving, and Withdrawal) and labeled by 3 human raters after standardized training with a validated suturing assessment method. The learning framework extracted features from RGB color and optical flow frames (video) using a pre-trained convolutional network (Alexnet). Visual features are concatenated to kinematic features and encoded using a Transformer encoder. The self-attention mechanism within the Transformer correlated information across both modality (video + kinematic) for all timesteps (Figure 1a). RESULTS: This study included 35 expert surgeons (median caseload 362 [IQR 250-1250]) and 8 training surgeon 64 [ 25-75]) with a total 3448 virtual sutures captured and labelled. Enhanced kinematics models outperformed basic kinematics from previous studies across all domains (+1-12% improvement in AUC; Figure 1b). Video models outperformed enhanced kinematics model on two sub-skills (Needle Hold Angle and Driving Smoothness), demonstrating the usefulness of videos. Multi-modal learning with video and kinematic data in a Transformer encoder yields the best performance for the key sub-skills of Needle driving and Needle withdrawal domains (+1-5.5% improvement in AUC; Figure 1b). CONCLUSIONS: This study demonstrated the best performance of automated assessment of suturing technical skills to date. Enhanced kinematics information and advanced multi-modal learning techniques can improve automated assessment and provide accurate feedback to training surgeons. Source of Funding: Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under Award Number R01CA251579. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health © 2023 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 209Issue Supplement 4April 2023Page: e62 Advertisement Copyright & Permissions© 2023 by American Urological Association Education and Research, Inc.MetricsAuthor Information Loc Trinh More articles by this author Timothy N. Chu More articles by this author Ahmed Ghazi More articles by this author John W. Davis More articles by this author Brian J. Miles More articles by this author Clayton Lau More articles by this author Anand Malpani More articles by this author Yan Liu More articles by this author Andrew J. Hung More articles by this author Expand All Advertisement PDF downloadLoading ...

Multi-station automatic classification of seismic signatures from the Lascar volcano database

Abstract. This study was aimed to build a multi-station automatic classification system for volcanic seismic signatures such as hybrid, long period, tremor, tectonic, and volcano–tectonic events. This system was based on a probabilistic model made using transfer learning, which has, as the main tool, a pre-trained convolutional network named AlexNet. We designed five experiments using different datasets with data that were real, synthetic, two different combinations of these (combined 1 and combined 2), and a balanced subset without synthetic data. The experiment presented the highest scores when a process of data augmentation was introduced into processing sequence. Thus, the lack of real data in some classes (imbalance) dramatically affected the quality of the results, because the learning step (training) was overfitted to the more numerous classes. To test the model stability with variable inputs, we implemented a k-fold cross-validation procedure. Under this approach, the results reached high predictive performance, considering that only the percentage of recognition of the tectonic events (TC) class was partially affected. The results obtained showed the performance of the probabilistic model, reaching high scores over different test datasets. The most valuable benefit of using this technique was that the use of volcano seismic signals from multiple stations provided a more generalizable model which, in the near future, can be extended to multi-volcano database systems. The impact of this work is significant in the evaluation of hazard and risk by monitoring the dynamic evolution of volcanic centers, which is crucial for understanding the stages in a volcano’s eruptive cycle.

Towards Task-Generic Image Compression: A Study of Semantics-Oriented Metrics

Instead of being observed by human, multimedia data are now more and more fed into machines to perform different kinds of semantic analysis. One image may be analyzed multiple times by different machine vision algorithms for different purposes. While machine vision-oriented image compression has been studied, the existing methods are usually driven by a specific machine vision task, and may not be applicable for other tasks. We address the task-generic image compression, in the hope that an image is compressed once but used multiple times for different tasks, all with satisfactory performance. Our study is based on the end-to-end learned image compression. We focus ourselves on the distortion metric, i.e., finding out a task-agnostic metric to estimate the quality of reconstructed images. On the one hand, we study deep feature distance as the metric, which transforms images into a latent space by a pretrained convolutional network -- the latent space is believed to be more aligned to semantics -- and calculates distance in the latent space. On the other hand, inspired by the saliency mechanism, we study an importance-weighted pixel distance as the metric, where the weights are generated to reflect the importance of the pixels to semantics. Moreover, we combine the two distances into one metric to investigate their complementary nature. An extensive set of experiments are performed to evaluate these metrics. Experimental results show that the combined metric performs the best, and leads to 20.79%~42.69% bits saving under the same semantic analysis performance, compared to the same network but optimized for signal fidelity. Interestingly, we observe that using the combined metric also improves the visual quality of the reconstructed images.

Hybrid convolution and transformer network for coupler fracture failure pattern segmentation recognition in heavy-haul trains

Couplers are an important component of heavy-haul trains. Effective failure analysis and fracture surface pattern recognition of the fractured coupler are of great significance for improving the safety of railway transportation. An end-to-end hybrid convolution and transformer (HCT) segmentation network for automatic semantic segmentation recognition of fracture surface failure patterns in metal couplers is proposed in this study. To achieve local modeling, the proposed method replaces handcrafted features with a pre-trained convolutional network for automatic feature extraction from the coupler's input image. Further, an HCT module is proposed for feature fusion and global modeling to improve the accuracy of different failure pattern semantic segmentation. A multi-scale loss function (MS-Loss) is proposed to improve the network performance while accelerating the convergence of the network. The experimental results show that the proposed HCT network achieves the highest mean intersection over union (mIoU) on the DJTU-Seg (91.28 %) and DWTT-Seg (86.37 %) datasets and achieves a detection speed of 51 fps on a single GPU. The proposed method can detect in real-time while avoiding contamination from human subjective factors. It provides a new pattern recognition method for the brittle fracture inspection of heavy-haul train couplers and even the whole metal material fracture measurement solution, offering promising applications.

ConViT: improving vision transformers with soft convolutional inductive biases* *This article is an updated version of: D’Ascoli S, Touvron H, Leavitt M L, Morcos A S, Biroli G and Sagun L 2021 ConViT: improving vision transformers with soft convolutional inductive biases Proc. 38th Int. Conf. Machine Learning vol 139 ed M Meila and T Zhang pp 2286–96.

Convolutional architectures have proven to be extremely successful for vision tasks. Their hard inductive biases enable sample-efficient learning, but come at the cost of a potentially lower performance ceiling. Vision transformers rely on more flexible self-attention layers, and have recently outperformed CNNs for image classification. However, they require costly pre-training on large external datasets or distillation from pre-trained convolutional networks. In this paper, we ask the following question: is it possible to combine the strengths of these two architectures while avoiding their respective limitations? To this end, we introduce gated positional self-attention (GPSA), a form of positional self-attention which can be equipped with a ‘soft’ convolutional inductive bias. We initialize the GPSA layers to mimic the locality of convolutional layers, then give each attention head the freedom to escape locality by adjusting a gating parameter regulating the attention paid to position versus content information. The resulting convolutional-like ViT architecture, ConViT, outperforms the DeiT (Touvron et al 2020 arXiv:2012.12877) on ImageNet, while offering a much improved sample efficiency. We further investigate the role of locality in learning by first quantifying how it is encouraged in vanilla self-attention layers, then analyzing how it has escaped in GPSA layers. We conclude by presenting various ablations to better understand the success of the ConViT. Our code and models are released publicly at https://github.com/facebookresearch/convit.

Deep Learning Cascaded Feature Selection Framework for Breast Cancer Classification: Hybrid CNN with Univariate-Based Approach

With the help of machine learning, many of the problems that have plagued mammography in the past have been solved. Effective prediction models need many normal and tumor samples. For medical applications such as breast cancer diagnosis framework, it is difficult to gather labeled training data and construct effective learning frameworks. Transfer learning is an emerging strategy that has recently been used to tackle the scarcity of medical data by transferring pre-trained convolutional network knowledge into the medical domain. Despite the well reputation of the transfer learning based on the pre-trained Convolutional Neural Networks (CNN) for medical imaging, several hurdles still exist to achieve a prominent breast cancer classification performance. In this paper, we attempt to solve the Feature Dimensionality Curse (FDC) problem of the deep features that are derived from the transfer learning pre-trained CNNs. Such a problem is raised due to the high space dimensionality of the extracted deep features with respect to the small size of the available medical data samples. Therefore, a novel deep learning cascaded feature selection framework is proposed based on the pre-trained deep convolutional networks as well as the univariate-based paradigm. Deep learning models of AlexNet, VGG, and GoogleNet are randomly selected and used to extract the shallow and deep features from the INbreast mammograms, whereas the univariate strategy helps to overcome the dimensionality curse and multicollinearity issues for the extracted features. The optimized key features via the univariate approach are statistically significant (p-value ≤ 0.05) and have good capability to efficiently train the classification models. Using such optimal features, the proposed framework could achieve a promising evaluation performance in terms of 98.50% accuracy, 98.06% sensitivity, 98.99% specificity, and 98.98% precision. Such performance seems to be beneficial to develop a practical and reliable computer-aided diagnosis (CAD) framework for breast cancer classification.

Physiological-physical feature fusion for automatic voice spoofing detection

Biometric speech recognition systems are often subject to various spoofing attacks, the most common of which are speech synthesis and speech conversion attacks. These spoofing attacks can cause the biometric speech recognition system to incorrectly accept these spoofing attacks, which can compromise the security of this system. Researchers have made many efforts to address this problem, and the existing studies have used the physical features of speech to identify spoofing attacks. However, recent studies have shown that speech contains a large number of physiological features related to the human face. For example, we can determine the speaker’s gender, age, mouth shape, and other information by voice. Inspired by the above researches, we propose a spoofing attack recognition method based on physiological-physical features fusion. This method involves feature extraction, a densely connected convolutional neural network with squeeze and excitation block (SE-DenseNet), and feature fusion strategies. We first extract physiological features in audio from a pre-trained convolutional network. Then we use SE-DenseNet to extract physical features. Such a dense connection pattern has high parameter efficiency, and squeeze and excitation blocks can enhance the transmission of the feature. Finally, we integrate the two features into the classification network to identify the spoofing attacks. Experimental results on the ASVspoof 2019 data set show that our model is effective for voice spoofing detection. In the logical access scenario, our model improves the tandem decision cost function and equal error rate scores by 5% and 7%, respectively, compared to existing methods.

A Deep Learning Approach to Clustering Visual Arts

Clustering artworks is difficult for several reasons. On the one hand, recognizing meaningful patterns based on domain knowledge and visual perception is extremely hard. On the other hand, applying traditional clustering and feature reduction techniques to the highly dimensional pixel space can be ineffective. To address these issues, in this paper we propose DELIUS: a DEep learning approach to cLustering vIsUal artS. The method uses a pre-trained convolutional network to extract features and then feeds these features into a deep embedded clustering model, where the task of mapping the input data to a latent space is jointly optimized with the task of finding a set of cluster centroids in this latent space. Quantitative and qualitative experimental results show the effectiveness of the proposed method. DELIUS can be useful for several tasks related to art analysis, in particular visual link retrieval and historical knowledge discovery in painting datasets.

Pre-trained network-based transfer learning: A small-sample machine learning approach to nuclear power plant classification problem

Some research topics belonging to classification problems in the nuclear industry, such as fault diagnosis and accident identification, can be solved by feature extraction and subsequent application of statistical machine learning classifiers. Recently, deep neural network-based methods with automatic feature extraction and high accuracy have gained wide attention. They usually require large-scale training data, however, plant fault or accident data are scarce or difficult to obtain. This paper proposes a convolutional network (CNN)-based transfer learning method to solve this problem. The network's shallow layer is derived from a pre-trained CNN based on the ImageNet database to automatically extract features, and the deep layer is customized to match the classification problem. Data in non-image formats are converted to image formats and subsequently used to train the network. Case studies of rotating machines fault diagnosis show that the proposed method requires only limited training data to achieve high accuracy.

Classification of pneumonia using pre-trained convolutional networks on chest X-Ray images

Pneumonia is an infection that is caused to the people of all ages with mild to severe inflammation of the lung disease. The most common and best method for the diagnosis of pneumonia is chest radiography. But diagnosing pneumonia from chest radiographs is a difficult task, even for radiologists. To overcome, Pre-Trained Convolutional Neural Networks namely Inceptionv3 and Resnet50 are used as a feature extractor. The exacted features are fed into 1D CNN which is classifies into Normal, Bacterial Pneumonia and Viral Pneumonia. When comparing Inceptionv3 with 1D CNN and resnet50 with 1D CNN, it is analyzed that Inceptionv3 with 1D CNN gives the satisfactory results of 96.04%.

Automatic Image Caption Generation Based on Some Machine Learning Algorithms

This paper is dedicated to machine learning, the branches of machine learning, which include the methods for solving this issue, and the practical implementation of the solution to the automatic image description generation. Automatic image caption generation is one of the frequent goals of computer vision. Image description generation models must solve a larger number of complex problems to have this task successfully solved. The objects in the image must be detected and recognized, after which a logical and syntactically correct textual description is generated. For that reason, description generation is a complex problem. It is an extremely important challenge for machine learning algorithms because it represents an impersonation of a complicated human ability to encapsulate huge amounts of highlighted visual pieces of information in descriptive language. The results of the generated descriptions are compared depending on the used pretrained convolutional networks. The BLEU metrics are used to calculate the quality of the image description. Although the solution to the problem of image description automatic generation does provide us with good results, there is yet room for improvement since there are images that are not adequately described.

Dual-Energy CT, Virtual Non-Calcium Bone Marrow Imaging of the Spine: An AI-Assisted, Volumetric Evaluation of a Reference Cohort with 500 CT Scans.

Virtual non-calcium (VNCa) images from dual-energy computed tomography (DECT) have shown high potential to diagnose bone marrow disease of the spine, which is frequently disguised by dense trabecular bone on conventional CT. In this study, we aimed to define reference values for VNCa bone marrow images of the spine in a large-scale cohort of healthy individuals. DECT was performed after resection of a malignant skin tumor without evidence of metastatic disease. Image analysis was fully automated and did not require specific user interaction. The thoracolumbar spine was segmented by a pretrained convolutional neuronal network. Volumetric VNCa data of the spine’s bone marrow space were processed using the maximum, medium, and low calcium suppression indices. Histograms of VNCa attenuation were created for each exam and suppression setting. We included 500 exams of 168 individuals (88 female, patient age 61.0 ± 15.9). A total of 8298 vertebrae were segmented. The attenuation histograms’ overlap of two consecutive exams, as a measure for intraindividual consistency, yielded a median of 0.93 (IQR: 0.88–0.96). As our main result, we provide the age- and sex-specific bone marrow attenuation profiles of a large-scale cohort of individuals with healthy trabecular bone structure as a reference for future studies. We conclude that artificial-intelligence-supported, fully automated volumetric assessment is an intraindividually robust method to image the spine’s bone marrow using VNCa data from DECT.

Early Recognition of Skin Malignancy in Images Based on Convolutional Networks by Using Dynamic System Model

Because of the high mortality rate, increased medical costs, and ongoing global growth in the incidence of this malignancy, early detection has become a top priority. Early detection and treatment of melanoma are critically important; the likelihood of a positive outcome rises dramatically. To address this issue, academic researchers plan to develop a prototype image analysis system based on deep learning to determine whether a lesion is malignant or benign based on dermatoscopy image databases. Pretrained convolutional networks with simple architectures were employed in this study to grasp their design better and to train the given dataset more quickly. Using convolutional neural networks as the basis, this research seeks to develop a deep learning system capable of classifying images. To train our model with the pretrained AlexNet, VGG, and ResNet networks, we will use the learning transfer methodology (or transfer learning), whose architecture we will outline so that it may subsequently be adjusted to our data. In this research work, fairly basic pretrained convolutional networks have been used to understand their architecture and efficiently train the given dataset. However, other networks have much more complex structures or even the same networks used, but with many more layers. For possible future work, it is proposed to use, for example, ResNet‐152, Vgg‐19, or other different networks such as DenseNet or Inception.