Lung CT Images Research Articles

Enabling Few-View 3D Tomographic Image Reconstruction by Geometry-Informed Deep Learning

<h3>Purpose/Objective(s)</h3> Deep learning affords enormous opportunities to augment the armamentarium of biomedical imaging, albeit its design and implementation have potential flaws. Fundamentally, most deep learning models are driven entirely by data without consideration of any physical priors, which dramatically increases the complexity of neural networks and limits the application scope and generalizability of the resultant models. Here we establish a geometry-informed deep learning framework for ultra-sparse tomographic image reconstruction. We introduce a novel mechanism for the integration of geometric priors of the imaging system. We demonstrate that the seamless inclusion of known geometric priors is essential to enhance the performance of volumetric computed tomography imaging with ultra-sparse sampling. <h3>Materials/Methods</h3> We propose a geometry-informed deep learning framework for 3D tomographic image reconstruction and illustrate the way that the known imaging geometry is integrated into the dual-domain deep learning. Specifically, the proposed framework consists of three modules: a) 2D projection generation network is developed to learn to generate novel-view projections from the given sparse views; b) geometric back-projection operator transforms the 2D projections to 3D images, referred to as geometry preserving images (GPI), which geometrically relates the pixelated 2D input data to the corresponding ray lines in 3D space; and c) 3D image refinement network learns to refine the GPI to generate the final 3D images. To evaluate feasibility of the proposed approach, we conduct experiments on a dataset containing 1018 lung CT images, where 80% and 20% data are used for training and testing. The projection images are digitally produced from CT images using geometry consistent with a clinical on-board cone-beam CT system for radiation therapy. <h3>Results</h3> We deploy the trained model on the held-out testing set for few-view 3D image reconstruction. The reconstructed results are compared with ground truth qualitatively and quantitatively. For single-/two-/three-view reconstruction, the average NRMSE / SSIM / PSNR values over all testing data are 0.368 / 0.734 / 20.770, 0.300 / 0.807 / 22.687 and 0.274 / 0.838 / 23.669, respectively. By visualizing the reconstructed CT images, we observe the proposed model can generate images closely to the targets although the anatomic structures of different patients have a large variance, indicating the potential of the proposed model for volumetric imaging even with few views. Moreover, experiments also show the proposed model can also be generalized to multi-view 3D image reconstruction and outperform deep models without geometry priors. <h3>Conclusion</h3> We present a novel geometry-integrated deep learning model for volumetric imaging with ultra-sparse sampling. The study opens new avenues for data-driven biomedical imaging and promises to provide substantially improved imaging tools for various clinical imaging and image-guided interventions.

Automatic segmentation of organs-at-risks of nasopharynx cancer and lung cancer by cross-layer attention fusion network with TELD-Loss.

Radiotherapy is one of the main treatments of nasopharyngeal cancer (NPC) and lung cancer. Accurate segmentation of organs at risks (OARs) in CT images is a key step in radiotherapy planning for NPC and lung cancer. However, the segmentation of OARs is influenced by the highly imbalanced size of organs, which often results in very poor segmentation results for small and difficult-to-segment organs. In addition, the complex morphological changes and fuzzy boundaries of OARs also pose great challenges to the segmentation task. In this paper, we propose a cross-layer attention fusion network (CLAF-CNN) to solve the problem of accurately segmentingOARs. In CLAF-CNN, we integrate the spatial attention maps of the adjacent spatial attention modules to make the segmentation targets more accurately focused, so that the network can capture more target-related features. In this way, the spatial attention modules in the network can be learned and optimized together. In addition, we introduce a new Top-K exponential logarithmic Dice loss (TELD-Loss) to solve the imbalance problem in OAR segmentation. The TELD-Loss further introduces a Top-K optimization mechanism based on Dice loss and exponential logarithmic loss, which makes the network pay more attention to small organs and difficult-to-segment organs, so as to enhance the overall performance of the segmentationmodel. We validated our framework on the OAR segmentation datasets of the head and neck and lung CT images in the StructSeg 2019 challenge. Experiments show that the CLAF-CNN outperforms the state-of-the-art attention-based segmentation methods in the OAR segmentation task with average Dice coefficient of 79.65% for head and neck OARs and 88.39% for lungOARs. This work provides a new network named CLAF-CNN which contains cross-layer spatial attention map fusion architecture and TELD-Loss for OAR segmentation. Results demonstrated that the proposed method could obtain accurate segmentation results for OARs, which has a potential of improving the efficiency of radiotherapy planning for nasopharynx cancer and lung cancer.

Lung cancer detection using an integration of fuzzy K-Means clustering and deep learning techniques for CT lung images

Lung cancer detection using an integration of fuzzy K-Means clustering and deep learning techniques for CT lung images

Deep Learning-Based Approach for Diagnosis of COVID-19 Patients from CT Lung Images

COVID-19 is currently one of the most life-threatening problems around the world. The fast and accurate detection of the COVID-19 infection plays an important role in identifying, making better decisions as well as ensuring treatment for the patients. In this study, we propose an approach to identify COVID-19 patients via deep learning methods. The proposed approach includes two steps: segmentation of lung and classification of patients via analyzing the segmented images. Especially, in the first step, the attention gate is integrated into Unet neural network structure, which can help increase the accuracy of the segmentation tasks. In the second step, the EfficientNet-B4 is applied to classify COVID-19 patients. By using EfficientNet, the performance of the classification process as well as the size of the model are improved. We apply the proposed approach for the database including 130 patients. Experimental results show the desired performances of the proposed approach.

Using Artificial Intelligence for Automatic Segmentation of CT Lung Images in Acute Respiratory Distress Syndrome

Knowledge of gas volume, tissue mass and recruitability measured by the quantitative CT scan analysis (CT-qa) is important when setting the mechanical ventilation in acute respiratory distress syndrome (ARDS). Yet, the manual segmentation of the lung requires a considerable workload. Our goal was to provide an automatic, clinically applicable and reliable lung segmentation procedure. Therefore, a convolutional neural network (CNN) was used to train an artificial intelligence (AI) algorithm on 15 healthy subjects (1,302 slices), 100 ARDS patients (12,279 slices), and 20 COVID-19 (1,817 slices). Eighty percent of this populations was used for training, 20% for testing. The AI and manual segmentation at slice level were compared by intersection over union (IoU). The CT-qa variables were compared by regression and Bland Altman analysis. The AI-segmentation of a single patient required 5–10 s vs. 1–2 h of the manual. At slice level, the algorithm showed on the test set an IOU across all CT slices of 91.3 ± 10.0, 85.2 ± 13.9, and 84.7 ± 14.0%, and across all lung volumes of 96.3 ± 0.6, 88.9 ± 3.1, and 86.3 ± 6.5% for normal lungs, ARDS and COVID-19, respectively, with a U-shape in the performance: better in the lung middle region, worse at the apex and base. At patient level, on the test set, the total lung volume measured by AI and manual segmentation had a R2 of 0.99 and a bias −9.8 ml [CI: +56.0/−75.7 ml]. The recruitability measured with manual and AI-segmentation, as change in non-aerated tissue fraction had a bias of +0.3% [CI: +6.2/−5.5%] and −0.5% [CI: +2.3/−3.3%] expressed as change in well-aerated tissue fraction. The AI-powered lung segmentation provided fast and clinically reliable results. It is able to segment the lungs of seriously ill ARDS patients fully automatically.

COVIDPRO-NET: a prognostic tool to detect COVID 19 patients from lung X-ray and CT images using transfer learning and Q-deformed entropy

ABSTRACT The humankind had faced several pandemic outbreaks, and coronavirus illness (COVID-19) caused by severe, acute respiratory syndrome coronavirus 2, is designated an emergency by the World Health Organization (WHO). Recognition of COVID-19 is a challenging task. The most commonly used methods are X-ray and CT scans images to inspect COVID-19 patients. It requires specialised medical professionals to report each patient’s health manually. It is found that COVID-19 shows considerable similarity to pneumonia lung disease. Thus, knowledge learned from a model to diagnose pneumonia can be translated to identify COVID-19. Transfer learning method offers a drastic performance when compared with results from conventional classification. In this study, Image pre-processing is done to alleviate intensity variations between medical images. These processed images undergo a feature extraction which is accomplished using Q-deformed entropy and deep learning extraction. The feature extraction techniques are employed to remove abnormal markers from images, noise impedance from tissues and lesions. The traits acquired are integrated to differentiate between COVID-19, pneumonia and healthy cases. The primary aim of this model is to produce an image processing tool for medical professionals. The model results to inspect how a healthy or COVID-19 individual outperforms conventional models. The maximum accuracy of the collected data set is 99.68%.

Deep Learning-Based Approaches to Improve Classification Parameters for Diagnosing COVID-19 from CT Images.

Patients infected with the COVID-19 virus develop severe pneumonia, which generally leads to death. Radiological evidence has demonstrated that the disease causes interstitial involvement in the lungs and lung opacities, as well as bilateral ground-glass opacities and patchy opacities. In this study, new pipeline suggestions are presented, and their performance is tested to decrease the number of false-negative (FN), false-positive (FP), and total misclassified images (FN + FP) in the diagnosis of COVID-19 (COVID-19/non-COVID-19 and COVID-19 pneumonia/other pneumonia) from CT lung images. A total of 4320 CT lung images, of which 2554 were related to COVID-19 and 1766 to non-COVID-19, were used for the test procedures in COVID-19 and non-COVID-19 classifications. Similarly, a total of 3801 CT lung images, of which 2554 were related to COVID-19 pneumonia and 1247 to other pneumonia, were used for the test procedures in COVID-19 pneumonia and other pneumonia classifications. A 24-layer convolutional neural network (CNN) architecture was used for the classification processes. Within the scope of this study, the results of two experiments were obtained by using CT lung images with and without local binary pattern (LBP) application, and sub-band images were obtained by applying dual-tree complex wavelet transform (DT-CWT) to these images. Next, new classification results were calculated from these two results by using the five pipeline approaches presented in this study. For COVID-19 and non-COVID-19 classification, the highest sensitivity, specificity, accuracy, F-1, and AUC values obtained without using pipeline approaches were 0.9676, 0.9181, 0.9456, 0.9545, and 0.9890, respectively; using pipeline approaches, the values were 0.9832, 0.9622, 0.9577, 0.9642, and 0.9923, respectively. For COVID-19 pneumonia/other pneumonia classification, the highest sensitivity, specificity, accuracy, F-1, and AUC values obtained without using pipeline approaches were 0.9615, 0.7270, 0.8846, 0.9180, and 0.9370, respectively; using pipeline approaches, the values were 0.9915, 0.8140, 0.9071, 0.9327, and 0.9615, respectively. The results of this study show that classification success can be increased by reducing the time to obtain per-image results through using the proposed pipeline approaches.

Analysis of 239 ordinary and severe cases of COVID-19: Clinical features and treatment.

This study retrospectively analyzed the clinical characteristics of patients with new coronavirus infection (COVID-19) and summarizes the treatment experience. A total of 239 COVID-19 patients admitted to the Hajar Hospital, Aja University of Medical Science, Tehran, Iran from March 22, 2020 to May 18, 2020 were selected, including 176 cases in the ordinary group and 63 cases in the severe/critical group. We collected and compared the clinical data of the two groups of patients, including general conditions, clinical symptoms, signs, laboratory tests, lung CT imaging and prognosis, and analyzed the treatment plans of the two groups. The mean age of 239 COVID-19 patients was 48.1±17.6 years, including 132 males. Patients in the severe and critically ill groups were older than the normal group, with more males and more underlying diseases. The difference was statistically significant (p < 0.05). The lymphocyte (LYM) counts and albumin (ALB) counts of the severe and critically ill groups were more significantly lower than those of the normal group; while the percentage of neutrophils (NEU), C-reactive protein (CRP), D-dimer, and lactate The increase of lactate dehydrogenase (LDH) and urea nitrogen (BUN) was more significant, and the difference was statistically significant (p < 0.05). Patients in the severe and critically ill groups received more antiviral drugs, glucocorticoids, and nasal catheters than those in the normal group, and the difference was statistically significant (p < 0.05). Also, we observed that the most radiological finding was bilateral ground-glass opacity in both groups, however, the rate of typical abnormalities in both chest CT scan and chest x-ray was significantly higher in sever/critical group except air-bronchogram. Taken together, we showed that combination of oseltamivir and glucocorticosteroid such as dexamethasone was very effective in severe patients.

Dense GAN and multi-layer attention based lesion segmentation method for COVID-19 CT images

As the COVID-19 virus spreads around the world, testing and screening of patients have become a headache for governments. With the accumulation of clinical diagnostic data, the imaging big data features of COVID-19 are gradually clear, and CT imaging diagnosis results become more important. To obtain clear lesion information from the CT images of patients' lungs is helpful for doctors to adopt effective medical methods, and at the same time, is helpful to screen the patients with real infection. Deep learning image segmentation is widely used in the field of medical image segmentation. However, there are some challenges in using deep learning to segment the lung lesions of COVID-19 patients. Since image segmentation requires the labeling of lesion information on a pixel by pixel basis, most professional radiologists need to screen and diagnose patients on the front line, and they do not have enough energy to label a large amount of image data. In this paper, an improved Dense GAN to expand data set is developed, and a multi-layer attention mechanism method, combined with U-Net's COVID-19 pulmonary CT image segmentation, is proposed. The experimental results showed that the segmentation method proposed in this paper improved the segmentation accuracy of COVID-19 pulmonary medical CT image by comparing with other image segmentation methods.

Deep learning for semi-automated unidirectional measurement of lung tumor size in CT

BackgroundPerforming Response Evaluation Criteria in Solid Tumor (RECISTS) measurement is a non-trivial task requiring much expertise and time. A deep learning-based algorithm has the potential to assist with rapid and consistent lesion measurement.PurposeThe aim of this study is to develop and evaluate deep learning (DL) algorithm for semi-automated unidirectional CT measurement of lung lesions.MethodsThis retrospective study included 1617 lung CT images from 8 publicly open datasets. A convolutional neural network was trained using 1373 training and validation images annotated by two radiologists. Performance of the DL algorithm was evaluated 244 test images annotated by one radiologist. DL algorithm’s measurement consistency with human radiologist was evaluated using Intraclass Correlation Coefficient (ICC) and Bland-Altman plotting. Bonferroni’s method was used to analyze difference in their diagnostic behavior, attributed by tumor characteristics. Statistical significance was set at p < 0.05.ResultsThe DL algorithm yielded ICC score of 0.959 with human radiologist. Bland-Altman plotting suggested 240 (98.4 %) measurements realized within the upper and lower limits of agreement (LOA). Some measurements outside the LOA revealed difference in clinical reasoning between DL algorithm and human radiologist. Overall, the algorithm marginally overestimated the size of lesion by 2.97 % compared to human radiologists. Further investigation indicated tumor characteristics may be associated with the DL algorithm’s diagnostic behavior of over or underestimating the lesion size compared to human radiologist.ConclusionsThe DL algorithm for unidirectional measurement of lung tumor size demonstrated excellent agreement with human radiologist.

Elaboration of a New Radiomics Strategy Based on Peritumoral and Intratumoral Features from Lung CT Images for the Pretherapy Prediction of the Histological Subtype in Patients with Nonsmall-Cell Lung Cancer

Background: Pretherapy prediction of the histological subtype in patients with nonsmall-cell lung cancer (NSCLC) is of critical importance for determining optimal therapeutic strategy. This study aimed to evaluate a new radiomics strategy that incorporates peritumoral and intratumoral features extracted from lung CT images with ensemble learning for pretreatment prediction of two major subtypes of NSCLC, lung squamous cell carcinoma (LUSC) and lung adenocarcinoma (LUAD).\ Methods: A total of 105 patients (47 LUSC and 58 LUAD) were involved in this study and were divided into training (n=73) and testing (n=32) cohorts. All patients underwent a preoperative CT scan. Seven categories of radiomics features were calculated from the intra- and peritumoral regions, constituting 3078 features from each patient’s dataset. Student’s t -tests in combination with three widely used feature selection methods, respectively, were adopted for dimension reduction and optimal feature subset determination. An ensemble classifier that was generated with five commonly used machine learning classifiers, was employed with the optimal features for classification model development, and the performance was quantitatively evaluated using the classification model with both training and testing cohorts for the prediction of LUSC and LUAD. Findings: The classification models developed by using optimal feature subsets determined from intratumoral region and peritumoral region with the ensemble classifier achieved mean area under the curve (AUC) of 0.87, 0.83 in the training cohort and 0.66, 0.60 in the testing cohort, respectively. The model developed by using the optimal feature subset selected from both intra- and peritumoral regions with the ensemble classifier achieved great performance improvement, with AUC of 0.87 and 0.78 in both cohorts, respectively. Interpretations: The proposed new radiomics strategy that extracts image features from the intra- and peritumoral regions with ensemble learning, could greatly improve the diagnostic performance for the histological subtype stratification in patients with NSCLC. Funding: This work was supported by the National Natural Science Foundation of China (No. 81901698), and Young Eagle plane of High Ambition Project (No. 2020CYJHXXP). Declaration of Interest: None to declare. Ethical Approval: This retrospective study was approved by the institutional ethics review board of Xijing Hospital, and informed content was waived.

RETRACTED ARTICLE: Advanced lung cancer prediction based on blockchain material using extended CNN

Lung cancer progresses rapidly, so early detection is crucial the initially use the Internet of Things for early-stage lung cancer diagnosis (IoT). Training for the IoT devices is in real time is crucial they have IoT connectivity all over the world, and great trust in the model's accuracy, both of data. Early detection of disease will save hundreds of thousands of lives every year it completely seals off the airways, which could lead to infection. Malignant growth was first observed with picture-processing and learning systems. The signs of lung cancer usually don't appear until the disease has spread. This is where medical attention becomes most challenging Initially you might find your fingers curving, or ends rising, or pain when trying to swallow followed by a whistling sound, a hoarseness, a weight gain in your face, and/upper chest swelling Some cancer causes key signs include regular or worsening chest pain, bloody or rust-colored sputum, worsened hunger, and shortness of breath. A lung CT image can be used to classify and organize the lung knobs, and also to assess their threat level ECNN features parameters like precision, time intricacy, but is not as sensitive in the preparation as previous frameworks They have irregular cells that form a tumour. Uncontrolled lung growth and destruction Types of lung cancer grow as this happens; they begin to develop until they form a tumour. Lung cells are damaged when they come into contact with airborne toxins. The novel proposed + approach is ECNN.

DUDA-Net: a double U-shaped dilated attention network for automatic infection area segmentation in COVID-19 lung CT images.

PurposeThe global health crisis caused by coronavirus disease 2019 (COVID-19) is a common threat facing all humankind. In the process of diagnosing COVID-19 and treating patients, automatic COVID-19 lesion segmentation from computed tomography images helps doctors and patients intuitively understand lung infection. To effectively quantify lung infections, a convolutional neural network for automatic lung infection segmentation based on deep learning is proposed.MethodThis new type of COVID-19 lesion segmentation network is based on a U-Net backbone. First, a coarse segmentation network is constructed to extract the lung areas. Second, in the encoding and decoding process of the fine segmentation network, a new soft attention mechanism, namely the dilated convolutional attention (DCA) mechanism, is introduced to enable the network to focus on better quantitative information to strengthen the network's segmentation ability in the subtle areas of the lesions.ResultsThe experimental results show that the average Dice similarity coefficient (DSC), sensitivity (SEN), specificity (SPE) and area under the curve of DUDA-Net are 87.06%, 90.85%, 99.59% and 0.965, respectively. In addition, the introduction of a cascade U-shaped network scheme and DCA mechanism can improve the DSC by 24.46% and 14.33%, respectively.ConclusionThe proposed DUDA-Net approach can automatically segment COVID-19 lesions with excellent performance, which indicates that the proposed method is of great clinical significance. In addition, the introduction of a coarse segmentation network and DCA mechanism can improve the COVID-19 segmentation performance.

COVSeg-NET: A deep convolution neural network for COVID-19 lung CT image segmentation.

COVID‐19 is a new type of respiratory infectious disease that poses a serious threat to the survival of human beings all over the world. Using artificial intelligence technology to analyze lung images of COVID‐19 patients can achieve rapid and effective detection. This study proposes a COVSeg‐NET model that can accurately segment ground glass opaque lesions in COVID‐19 lung CT images. The COVSeg‐NET model is based on the fully convolutional neural network model structure, which mainly includes convolutional layer, nonlinear unit activation function, maximum pooling layer, batch normalization layer, merge layer, flattening layer, sigmoid layer, and so forth. Through experiments and evaluation results, it can be seen that the dice coefficient, sensitivity, and specificity of the COVSeg‐NET model are 0.561, 0.447, and 0.996 respectively, which are more advanced than other deep learning methods. The COVSeg‐NET model can use a smaller training set and shorter test time to obtain better segmentation results.

COVID-19 Detection based on Lung CT Images using Convolutional Neural Network Architecture

Abstract: A CNN (Convolutional Neural Network) is a class of deep neural networks that are most used for analyzing visual imagery. This is the most widely used deep learning algorithm for image and video recognition, image classification, image segmentation, medical image analysis and natural language processing, and many more. But, training the Neural Network from scratch requires more training time and requires a lot of data, and the performance of the model is affected. ‘Transfer learning’ is the best approach where existing architectures that have been trained on the large datasets are then fine-tuned for our customized dataset. This reduces the time to train and often results in better performance. The main objective of this paper is to detect COVID-19 from lung CT images using the VGG16 CNN (Convolutional Neural Network) model from the ImageNet. Index Terms: CNN, Transfer learning, VGG16, and ImageNet.

Reducing False-Positives in Lung Nodules Detection Using Balanced Datasets.

Malignant pulmonary nodules are one of the main manifestations of lung cancer in early CT image screening. Since lung cancer may have no early obvious symptoms, it is important to develop a computer-aided detection (CAD) system to assist doctors to detect the malignant pulmonary nodules in the early stage of lung cancer CT diagnosis. Due to the recent successful applications of deep learning in image processing, more and more researchers have been trying to apply it to the diagnosis of pulmonary nodules. However, due to the ratio of nodules and non-nodules samples used in the training and testing datasets usually being different from the practical ratio of lung cancer, the CAD classification systems may easily produce higher false-positives while using this imbalanced dataset. This work introduces a filtering step to remove the irrelevant images from the dataset, and the results show that the false-positives can be reduced and the accuracy can be above 98%. There are two steps in nodule detection. Firstly, the images with pulmonary nodules are screened from the whole lung CT images of the patients. Secondly, the exact locations of pulmonary nodules will be detected using Faster R-CNN. Final results show that this method can effectively detect the pulmonary nodules in the CT images and hence potentially assist doctors in the early diagnosis of lung cancer.

An Early Prediction of Lung Cancer using CT Scan Images

Lung cancer is a common occurrence type in a population and one amonglethal cancers. Recently, out of several research presented by diverse health agencies; it is obvious that the fatality ratio is rising due todelayeddiagnosis of lung cancer. Hence, an artificial intelligence-based diagnosis is required to find out the onset of lung nodule micro-calcification, which may support the doctors and radiologists to accurately predict it through image processing methods. In this paper, a novel technique is proposed to identify the nodule micro-calcification pattern by using its physical features. The physical features that considered are the reflection coefficients and mass densities of the binned CT image of lung. The physical features measurements reiteratesonce again the existence of malignant nodule. Then, by applying the methods of thresholding and in interpolation of physical features, a three-dimensional (3D) projected image of the region of interest (ROI) is achieved in respect of physical dimensions. Thus, the nodule size is calculated from 3D projection. This concept is used to verify how best in classification with 100 malignant images (the nodule presence) and 10 normal images (the nodule absence). Apart size measurement, the proposed method supports SVM classifier to act for excellentclassification from normal and malignantinput imagesby just using two physical features. The classifier exhibited an accuracy of 98%.

Automatic detection of solitary pulmonary nodules using superpixels segmentation based iterative clustering approach

Lung disease is the premier cause of cancer deaths all over the world. The lung cancer diagnosis is mediocre at an early stage since it is impotent by the radiologist. Various investigations conducted so far manifest clearly that the nodule segmentation algorithms are inefficient. Thus, this investigation has centralized superpixels segmentation based iterative clustering (CSSBIC) and a sophisticated optimization approach for detailed segmentation of pulmonary nodules. The supreme intent of this research paper is the enhancement of lung CT images to identify the tumor effectively and small-scale anomalous nodules segmentation in the lung area. The initial phase an-isotropic diffusion with masking (AIDME) enhancement techniques which can eliminate the noise discern in the images. The next step is to apply the centralized superpixels segmentation based iterative clustering (CSSBIC) algorithm by using an improved picture sequence nodule for abnormal lung tissue prediction. The lung nodule is essentially retrieved using a GWO, based on deep learning techniques, with advanced ONN (ONN) and advanced CNN (CNN). The average time for segmenting the Nodule Slice order is 1.06 s. The best classification accuracy is 97% via GWO based advanced one nearest neighbour (AONN) and 97.6% by GWO based advanced convolutional neural network (ACNN) classifier.

LNCDS: A 2D-3D cascaded CNN approach for lung nodule classification, detection and segmentation

The early detection of lung cancer is attained with the detection of initial stage nodules (3−30mm) which can exorbitantly increase the 5-year survival rate of lung cancer patients. Nodules are very small size circumscribed structures in the lungs and are difficult to detect due to their size. The identification of nodule is also more challenging due to similarly resembling structures like non-nodules that contains features which could make it identifiable as a nodule. Therefore, to deal with these challenging issues, we proposed a novel approach for segmentation, classification and detection of lung nodules from CT scan images. In our proposed method we involve a maximum intensity projection technique as a part of image preprocessing method. We demonstrated our experimentation and proposed SquExUNet segmentation model and 3D-NodNet classification model on publicly available Lung Image Database Consortium – Image Database Resource Initiative (LIDC), LNDb Challenge Dataset and purely independent Indian Lung CT Image Database (ILCID) clinical dataset. We proposed a 2D-3D cascaded CNN strategy for detection of nodule that yields the accurately segmented and classified nodule. Results obtained with proposed method indicates that we have successfully detected and segmented the lung nodules effectively, compared to existing lung nodule detection and segmentation algorithms. We achieved a Dice-Coefficient metrics of 0.80 for segmentation of nodule and 90.01% Sensitivity for nodule detection.

Content-Based Medical Image Retrieval Using Delaunay Triangulation Segmentation Technique

This article presents a novel technique for retrieval of lung images from the collection of medical CT images. The proposed content-based medical image retrieval (CBMIR) technique uses an automated image segmentation technique called Delaunay triangulation (DT) in order to segment lung organ (region of interest) from the original medical image. The proposed method extracts novel and discriminant features from the segmented lung region instead of extracting novel features from the whole original image. For the extraction of shape features, the authors employ edge histogram descriptor (EHD) and geometric moments (GM), and for the extraction of texture features, the authors use gray-level co-occurrence matrix (GLCM) technique. The shape and texture features are combined to form the hybrid feature which is used for retrieval of similar lung images. The proposed method is evaluated using two benchmark datasets of lung CT images. The simulation results prove that the proposed CBMIR framework shows improved performance in terms of retrieval accuracy and retrieval time.