Deep Learning-based Segmentation Research Articles

CellSium: versatile cell simulator for microcolony ground truth generation.

To train deep learning-based segmentation models, large ground truth datasets are needed. To address this need in microfluidic live-cell imaging, we present CellSium, a flexibly configurable cell simulator built to synthesize realistic image sequences of bacterial microcolonies growing in monolayers. We illustrate that the simulated images are suitable for training neural networks. Synthetic time-lapse videos with and without fluorescence, using programmable cell growth models, and simulation-ready 3D colony geometries for computational fluid dynamics are also supported. CellSium is free and open source software under the BSD license, implemented in Python, available at github.com/modsim/cellsium (DOI: 10.5281/zenodo.6193033), along with documentation, usage examples and Docker images. Supplementary data are available at Bioinformatics Advances online.

Comparison of Two-Dimensional- and Three-Dimensional-Based U-Net Architectures for Brain Tissue Classification in One-Dimensional Brain CT.

Brain tissue segmentation plays a crucial role in feature extraction, volumetric quantification, and morphometric analysis of brain scans. For the assessment of brain structure and integrity, CT is a non-invasive, cheaper, faster, and more widely available modality than MRI. However, the clinical application of CT is mostly limited to the visual assessment of brain integrity and exclusion of copathologies. We have previously developed two-dimensional (2D) deep learning-based segmentation networks that successfully classified brain tissue in head CT. Recently, deep learning-based MRI segmentation models successfully use patch-based three-dimensional (3D) segmentation networks. In this study, we aimed to develop patch-based 3D segmentation networks for CT brain tissue classification. Furthermore, we aimed to compare the performance of 2D- and 3D-based segmentation networks to perform brain tissue classification in anisotropic CT scans. For this purpose, we developed 2D and 3D U-Net-based deep learning models that were trained and validated on MR-derived segmentations from scans of 744 participants of the Gothenburg H70 Cohort with both CT and T1-weighted MRI scans acquired timely close to each other. Segmentation performance of both 2D and 3D models was evaluated on 234 unseen datasets using measures of distance, spatial similarity, and tissue volume. Single-task slice-wise processed 2D U-Nets performed better than multitask patch-based 3D U-Nets in CT brain tissue classification. These findings provide support to the use of 2D U-Nets to segment brain tissue in one-dimensional (1D) CT. This could increase the application of CT to detect brain abnormalities in clinical settings.

Automated segmentation of articular disc of the temporomandibular joint on magnetic resonance images using deep learning

Temporomandibular disorders are typically accompanied by a number of clinical manifestations that involve pain and dysfunction of the masticatory muscles and temporomandibular joint. The most important subgroup of articular abnormalities in patients with temporomandibular disorders includes patients with different forms of articular disc displacement and deformation. Here, we propose a fully automated articular disc detection and segmentation system to support the diagnosis of temporomandibular disorder on magnetic resonance imaging. This system uses deep learning-based semantic segmentation approaches. The study included a total of 217 magnetic resonance images from 10 patients with anterior displacement of the articular disc and 10 healthy control subjects with normal articular discs. These images were used to evaluate three deep learning-based semantic segmentation approaches: our proposed convolutional neural network encoder-decoder named 3DiscNet (Detection for Displaced articular DISC using convolutional neural NETwork), U-Net, and SegNet-Basic. Of the three algorithms, 3DiscNet and SegNet-Basic showed comparably good metrics (Dice coefficient, sensitivity, and positive predictive value). This study provides a proof-of-concept for a fully automated deep learning-based segmentation methodology for articular discs on magnetic resonance images, and obtained promising initial results, indicating that the method could potentially be used in clinical practice for the assessment of temporomandibular disorders.

A Hybrid Optimization Approach with Deep Learning Technique for the Classification of Dental Caries

Due to the wealth of data available from different radiographic images, detecting dental caries has traditionally been a difficult undertaking. Numerous techniques have been developed to enhance image quality for quicker caries detection. For the investigation of medical images, deep learning has emerged as the preferred methodology. This study provides a thorough examination of the application of deep learning to object detection, segmentation, and classification. It also examines the literature on deep learning-based segmentation and identification techniques for dental images. To identify dental caries, several techniques have been used to date. However, these techniques are inefficient, inaccurate, and unable to handle a sizable amount of datasets. There is a need for a way that can get around these issues since the prior methods failed to do so. In the domains of medicine and radiology, deep convolutional neural networks (CNN) have produced amazing results in predicting and diagnosing diseases. This new field of healthcare research is developing quickly. The current study's objective was to assess the effectiveness of deep CNN algorithms for dental caries detection and diagnosis on radiographic images. The Convolutional Neural Network (CNN) method, which is based on artificial intelligence, is used in this study to introduce hybrid optimal deep learning, which offers superior performance.

Boundary-Oriented Binary Building Segmentation Model With Two Scheme Learning for Aerial Images

Various deep learning-based segmentation models have been developed to segment buildings in aerial images. However, the segmentation maps predicted by the conventional convolutional neural network-based methods cannot accurately determine the shapes and boundaries of segmented buildings. In this article, to improve the prediction accuracy for the boundaries and shapes of segmented buildings in aerial images, we propose the boundary-oriented binary building segmentation model (B3SM). To construct the B3SM for boundary-enhanced semantic segmentation, we present two-scheme learning (Schemes I and II), which uses the upsampling interpolation method (USIM) as a new operator and a boundary-oriented loss function (B-Loss). In Scheme I, a raw input image is processed and transformed into a presegmented map. In Scheme II, the presegmented map from Scheme I is transformed into a more fine-grained representation. To connect these two schemes, we use the USIM operator. In addition, the novel B-Loss function is implemented in B3SM to extract the features of the boundaries of buildings effectively. To perform quantitative evaluation of the shapes and boundaries of segmented buildings generated by B3SM, we develop a new metric called the boundary-oriented intersection over union (B-IoU). After evaluating the effectiveness of two-scheme learning, USIM, and B-Loss for building segmentation, we compare the performance of B3SM to those of other state-of-the-art methods using public and custom datasets. The experimental results demonstrate that the B3SM outperforms other state-of-the-art models, resulting in more accurate shapes and boundaries for segmented buildings in aerial images.

Review of Nodule Mineral Image Segmentation Algorithms for Deep-Sea Mineral Resource Assessment

A large number of nodule minerals exist in the deep sea. Based on the factors of difficulty in shooting, high economic cost and high accuracy of resource assessment, large-scale planned commercial mining has not yet been conducted. Only experimental mining has been carried out in areas with high mineral density and obvious benefits after mineral resource assessment. As an efficient method for deep-sea mineral resource assessment, the deep towing system is equipped with a visual system for mineral resource analysis using collected images and videos, which has become a key component of resource assessment. Therefore, high accuracy in deep-sea mineral image segmentation is the primary goal of the segmentation algorithm. In this paper, the existing deep-sea nodule mineral image segmentation algorithms are studied in depth and divided into traditional and deep learning-based segmentation methods, and the advantages and disadvantages of each are compared and summarized. The deep learning methods show great advantages in deep-sea mineral image segmentation, and there is a great improvement in segmentation accuracy and efficiency compared with the traditional methods. Then, the mineral image dataset and segmentation evaluation metrics are listed. Finally, possible future research topics and improvement measures are discussed for the reference of other researchers.

Automated deep learning-based segmentation of COVID-19 lesions from chest computed tomography images.

PurposeThe novel coronavirus COVID-19, which spread globally in late December 2019, is a global health crisis. Chest computed tomography (CT) has played a pivotal role in providing useful information for clinicians to detect COVID-19. However, segmenting COVID-19-infected regions from chest CT results is challenging. Therefore, it is desirable to develop an efficient tool for automated segmentation of COVID-19 lesions using chest CT. Hence, we aimed to propose 2D deep-learning algorithms to automatically segment COVID-19-infected regions from chest CT slices and evaluate their performance.Material and methodsHerein, 3 known deep learning networks: U-Net, U-Net++, and Res-Unet, were trained from scratch for automated segmenting of COVID-19 lesions using chest CT images. The dataset consists of 20 labelled COVID-19 chest CT volumes. A total of 2112 images were used. The dataset was split into 80% for training and validation and 20% for testing the proposed models. Segmentation performance was assessed using Dice similarity coefficient, average symmetric surface distance (ASSD), mean absolute error (MAE), sensitivity, specificity, and precision.ResultsAll proposed models achieved good performance for COVID-19 lesion segmentation. Compared with Res-Unet, the U-Net and U-Net++ models provided better results, with a mean Dice value of 85.0%. Compared with all models, U-Net gained the highest segmentation performance, with 86.0% sensitivity and 2.22 mm ASSD. The U-Net model obtained 1%, 2%, and 0.66 mm improvement over the Res-Unet model in the Dice, sensitivity, and ASSD, respectively. Compared with Res-Unet, U-Net++ achieved 1%, 2%, 0.1 mm, and 0.23 mm improvement in the Dice, sensitivity, ASSD, and MAE, respectively.ConclusionsOur data indicated that the proposed models achieve an average Dice value greater than 84.0%. Two-dimensional deep learning models were able to accurately segment COVID-19 lesions from chest CT images, assisting the radiologists in faster screening and quantification of the lesion regions for further treatment. Nevertheless, further studies will be required to evaluate the clinical performance and robustness of the proposed models for COVID-19 semantic segmentation.

Multi-Spectral Imaging for Weed Identification in Herbicides Testing

A new methodology to help to improve the efficiency of herbicide assessment is explained. It consists of an automatic tool to quantify the percentage of weeds and plants of interest (sunflowers) that are present in a given area. Images of the crop field taken from Sequoia camera were used. Firstly, the quality of the images of each band is improved. Later, the resulting multi-spectral images are classified into several classes (soil, sunflower and weed) through a novel algorithm implemented in e-Cognition software. Obtained results of the proposed classifications have been compared with two deep learning-based segmentation methods (U-Net and FPN).

A Deep Learning-Based Segmentation System for Rapid Onsite Cytologic Pathology Evaluation of Pancreatic Masses: A Retrospective, Multicenter, Diagnostic Study

A Deep Learning-Based Segmentation System for Rapid Onsite Cytologic Pathology Evaluation of Pancreatic Masses: A Retrospective, Multicenter, Diagnostic Study

Remote-Sensing Image Segmentation Based on Implicit 3-D Scene Representation

Remote sensing image segmentation, as a challenging but fundamental task, has drawn increasing attention in the remote sensing field. Recent advances in deep learning have greatly boosted research on this task. However, the existing deep learning-based segmentation methods heavily rely on a large amount of pixel-wise labeled training data, and the labeling process is time-consuming and labor-intensive. In this paper, we focus on the scenario that leverages the 3D structure of multi-view images and a limited number of annotations to generate accurate novel view segmentation. Under this scenario, we propose a novel method for remote sensing image segmentation based on implicit 3D scene representation, which generates arbitrary-view segmentation output from limited segmentation annotations. The proposed method employs a two-stage training strategy. In the first stage, we optimize the implicit neural representations of a 3D scene and encode their multi-view images into a neural radiance field. In the second stage, we transform the scene color attribute into semantic labels and propose a ray-convolution network to aggregate local 3D consistency cues across different locations. We also design a color-radiance network to help our method generalize to unseen views. Experiments on both synthetic and real-world data suggest that our method significantly outperforms deep convolutional networks (CNN)-based methods and other view synthesis-based methods. We also show that the proposed method can be applied as a novel data augmentation approach that benefits CNN-based segmentation methods.

Automatic upper airway segmentation in static and dynamic MRI via anatomy-guided convolutional neural networks.

Upper airway segmentation on MR images is a prerequisite step for quantitatively studying the anatomical structure and function of the upper airway and surrounding tissues. However, the complex variability of intensity and shape of anatomical structures and different modes of image acquisition commonly used in this application makes automatic upper airway segmentation challenging. In this paper, we develop and test a comprehensive deep learning-based segmentation system for use on MR images to address this problem. In our study, both static and dynamic MRI data sets are utilized, including 58 axial static 3D MRI studies, 22 mid-retropalatal dynamic 2D MRI studies, 21 mid-retroglossal dynamic 2D MRI studies, 36 mid-sagittal dynamic 2D MRI studies, and 23 isotropic dynamic 3D MRI studies, involving a total of 160 subjects and over 20000 MRI slices. Samples of static and 2D dynamic MRI data sets were randomly divided into training, validation, and test sets by an approximate ratio of 5:2:3. Considering that the variability of annotation data among 3D dynamic MRIs was greater than for other MRI data sets, we increased the ratio of training data for these data to improve the robustness of the model. We designed a unified framework consisting of the following procedures. For static MRI, a generalized region-of-interest (GROI) strategy is applied to localize the partitions of nasal cavity and other portions of upper airway in axial data sets as two separate subobjects. Subsequently, the two subobjects are segmented by two separate 2D U-Nets. The two segmentation results are combined as the whole upper airway structure. The GROI strategy is also applied to other MRI modes. To minimize false-positive and false-negative rates in the segmentation results, we employed a novel loss function based explicitly on these rates to train the segmentation networks. An inter-reader study is conducted to test the performance of our system in comparison to human variability in ground truth (GT) segmentation of these challenging structures. The proposed approach yielded mean Dice coefficients of 0.84±0.03, 0.89±0.13, 0.84±0.07, and 0.86±0.05 for static 3D MRI, mid-retropalatal/mid-retroglossal 2D dynamic MRI, mid-sagittal 2D dynamic MRI, and isotropic dynamic 3D MRI, respectively. The quantitative results show excellent agreement with manual delineation results. The inter-reader study results demonstrate that the segmentation performance of our approach is statistically indistinguishable from manual segmentations considering the inter-reader variability in GT. The proposed method can be utilized for routine upper airway segmentation from static and dynamic MR images with high accuracy and efficiency. The proposed approach has the potential to be employed in other dynamic MRI-related applications, such as lung or heart segmentation.

Abstract 14467: Machine Learning for Identifying Salient Regions on the Coronary Artery on CT for Radiomic Feature Extraction and Characterization of CAD

Introduction: Radiomics is a quantitative approach that allows for high throughput extraction of image features such as texture. It has been used for characterization of coronary artery disease (CAD). However, the reliability of radiomic features (RFs) varies with the image segmentation quality. We present a new strategy to establish a segmentation confidence map spatially across the coronary arteries in the dataset. We then evaluate the quality of RFs in salient (high confidence) and non-salient (low confidence) regions in the coronary artery. Purpose: To identify salient regions in the coronary artery, providing high segmentation certainty as well as a site for informative radiomic feature extraction, useful in CAD characterization. Methods: In our approach, we first applied a deep learning-based segmentation model (U-Net with ResNet backbone) to segment the coronary arteries (Fig. 1a). Next, we constructed a mean probability atlas from the segmentation probability maps, using fiducial registration (Fig. 1b). Finally, we identified salient and non-salient regions within the segmentation (Fig. 1c) by masking the mean probability atlas with the predicted segmentation and then thresholding them. Human expert-generated manual segmentation was used as ground truth. RFs—shape (26), first order (19) and second order (75)—were extracted from salient and non-salient regions for comparative correlation analysis with features from ground truth segmentation. A higher correlation implies that these features are equally informative as those from the manual segmentation. Results: The model achieved a dice score of 79.8% on the hold-out set. The RFs extracted only from the salient regions (Proximal portion of RCA and LAD) of the coronary artery display strong correlations with RFs extracted from manual segmentations (Fig 1d). Conclusion: We present a framework to identify regions for optimal extraction of RFs for improved CAD characterization.

Urban infrastructure audit: an effective protocol to digitize signalized intersections by mining street view images

ABSTRACT Auditing and mapping traffic infrastructure is a crucial task in urban management. For example, signalized intersections play an essential role in transportation management; however, effectively identifying these intersections remains unsolved. Traditionally, signalized intersection data are manually collected through field audits or checking street view images (SVIs), which is time-consuming and labor-intensive. This study proposes an effective protocol to identify signalized intersections using road networks and SVIs. First, we propose a six-step geoprocessing model to generate an intersection feature layer from road networks. Second, we utilize up to three nearest SVIs to capture streetscapes at each intersection. Then, a deep learning-based image segmentation model is adopted to recognize traffic light-related pixels from each SVI. Last, we design a post-processing step to generate new features characterizing SVIs’ segmentation results at each intersection and build a decision tree model to determine the traffic control type. Results demonstrate that the proposed protocol can effectively identify signalized intersections with an overall accuracy of 97.05%. It also proves the effectiveness of SVIs for auditing urban infrastructures. This study can directly benefit transportation agencies by providing a ready-to-use smart audit and mapping solution for large-scale identification and mapping of signalized intersections.

Deep Learning-Based Segmentation and Uncertainty Assessment for Automated Analysis of Myocardial Perfusion MRI Datasets Using Patch-Level Training and Advanced Data Augmentation.

In this work, we develop a patch-level training approach and a task-driven intensity-based augmentation method for deep-learning-based segmentation of motion-corrected perfusion cardiac magnetic resonance imaging (MRI) datasets. Further, the proposed method generates an image-based uncertainty map thanks to a novel spatial sliding-window approach used during patch-level training, hence allowing for uncertainty quantification. Using the quantified uncertainty, we detect the out-of-distribution test data instances so that the end-user can be alerted that the test data is not suitable for the trained network. This feature has the potential to enable a more reliable integration of the proposed deep learning-based framework into clinical practice. We test our approach on external MRI data acquired using a different acquisition protocol to demonstrate the robustness of our performance to variations in pulse-sequence parameters. The presented results further demonstrate that our deep-learning image segmentation approach trained with the proposed data-augmentation technique incorporating spatiotemporal (2D+time) patches is superior to the state-of-the-art 2D approach in terms of generalization performance.

3D shape sensing and deep learning-based segmentation of strawberries

Automation and robotisation of the agricultural sector are seen as a viable solution to socio-economic challenges faced by this industry. This technology often relies on intelligent perception systems providing information about crops, plants and the entire environment. The challenges faced by traditional 2D vision systems can be addressed by modern 3D vision systems which enable straightforward localisation of objects, size and shape estimation, or handling of occlusions. So far, the use of 3D sensing was mainly limited to indoor or structured environments. In this paper, we evaluate modern sensing technologies including stereo and time-of-flight cameras for 3D perception of shape in agriculture and study their usability for segmenting out soft fruit from background based on their shape. To that end, we propose a novel 3D deep neural network which exploits the organised nature of information originating from the camera-based 3D sensors. We demonstrate the superior performance and efficiency of the proposed architecture compared to the state-of-the-art 3D networks. Through a simulated study, we also show the potential of the 3D sensing paradigm for object segmentation in agriculture and provide insights and analysis of what shape quality is needed and expected for further analysis of crops. The results of this work should encourage researchers and companies to develop more accurate and robust 3D sensing technologies to assure their wider adoption in practical agricultural applications.

A deep learning-based segmentation pipeline for profiling cellular morphodynamics using multiple types of live cell microscopy

SummaryTo accurately segment cell edges and quantify cellular morphodynamics from live-cell imaging data, we developed a deep learning-based pipeline termed MARS-Net (multiple-microscopy-type-based accurate and robust segmentation network). MARS-Net utilizes transfer learning and data from multiple types of microscopy to localize cell edges with high accuracy. For effective training on distinct types of live-cell microscopy, MARS-Net comprises a pretrained VGG19 encoder with U-Net decoder and dropout layers. We trained MARS-Net on movies from phase-contrast, spinning-disk confocal, and total internal reflection fluorescence microscopes. MARS-Net produced more accurate edge localization than the neural network models trained with single-microscopy-type datasets. We expect that MARS-Net can accelerate the studies of cellular morphodynamics by providing accurate pixel-level segmentation of complex live-cell datasets.

Deep learning-based segmentation of lithium-ion battery microstructures enhanced by artificially generated electrodes

Accurate 3D representations of lithium-ion battery electrodes, in which the active particles, binder and pore phases are distinguished and labeled, can assist in understanding and ultimately improving battery performance. Here, we demonstrate a methodology for using deep-learning tools to achieve reliable segmentations of volumetric images of electrodes on which standard segmentation approaches fail due to insufficient contrast. We implement the 3D U-Net architecture for segmentation, and, to overcome the limitations of training data obtained experimentally through imaging, we show how synthetic learning data, consisting of realistic artificial electrode structures and their tomographic reconstructions, can be generated and used to enhance network performance. We apply our method to segment x-ray tomographic microscopy images of graphite-silicon composite electrodes and show it is accurate across standard metrics. We then apply it to obtain a statistically meaningful analysis of the microstructural evolution of the carbon-black and binder domain during battery operation.

A Transfer Learning Approach for Improving OAR Segmentation in the Adaptive Therapy or Retreatment of Head and Neck Cancer

<h3>Purpose/Objective(s)</h3> Deep learning-based segmentation has achieved great progress compared with traditional method. However, the generalizability is still an unsolved problem, leading to suboptimal contours for some cases that requires manual editing. A big source of error lies in the abnormal patient anatomy that was not well covered by the training data. In adaptive or retreatment settings, a well-contoured CT is available. However, the current DLSEG algorithm does not allow such information to be utilized. Our proposed solution is to use transfer learning method to tune a trained DLSEG model toward individual patient using the prior segmented scans. <h3>Materials/Methods</h3> A HN DLSEG model was trained on 2015 MICCAI Head and Neck dataset. The model was applied on HNSCC-3DCT-RT dataset from TCIA which contains 3 CT scans at pre-, mid-, and end- treatment course. The DLSEG result was compared with the ground truth contour in the study. Dice similarity coefficient (DSC), 95% Hausdorff distance (HD95), and mean surface distance (MSD) were computed for BrainStem, Mandible, Parotid, SpinalCord and OpticNerve. The standard deviations of these metrics were computed, both within the same patient (IntraPatient variation) and across patients (InterPatient variation). Then, the DLSEG was re-tuned using transfer learning on the pre- treatment CT scan of each individual patient. The re-tuned model was applied to the mid- and end- treatment CT. The segmentation accuracy was evaluated. <h3>Results</h3> All structures showed a much larger InterPatient variation than IntraPatient variation, indicating that patient specific anatomical feature is one of the main contributors to segmentation accuracy. When transfer learning is applied. Nearly all patients saw improvement in contouring accuracy. Statistically significant improvement was observed for BrainStem, Mandible, Parotid and Parotid_L. <h3>Conclusion</h3> Transfer learning is a promising approach to incorporate the existing segmentation on prior CT to improve the segmentation accuracy in the adaptive or re-treatment settings.

Development of an Automatic and Physician-Specific OAR Segmentation Framework for Radiotherapy Treatment Planning

Contouring organs at risk (OARs) is a key step of radiotherapy treatment planning. In this study, a deep learning-based framework was developed to automatize the whole OAR contouring process contributing to the clinical practice.This auto-segmentation framework consisted of a pre-processing module, a deep learning-based segmentation module, and a post-processing module. We focused on the auto-segmentation in head and neck, and pelvis sites, and CT images with manual OAR contours from 67 and 54 patient cases were collected from our clinical database for the deep learning, respectively. 16 head and neck OARs and 11 pelvis OARs which were contoured in most patient cases were selected for auto-segmentation. In the pre-processing module, the original image data was extracted from the clinical DICOM Image file, and then processed for the segmentation network learning. A U-Net like network was designed in the auto-segmentation module, which was designed to consider the contour consistency among adjacent image slices and speed up the network convergency. Manual OAR contours were used as the ground truth for the network training, and a weighted Dice loss function was utilized to handle the missing organ contours in some patient cases. The segmentation results would be converted to a DCIOM Structure file in the post-processing module for treatment planning.The accuracy of the current auto-segmentation framework in terms of the Dice similarity coefficient (DSC) is shown in the table. Once a clinical DICOM Image data is sent to the framework, a corresponding DICOM Structure file would be automatically generated containing the OAR segmentation results. The overall operation takes about 1-2 min per case without the need for additional manual interference.A deep learning-based framework was developed for the auto-segmentation of head and neck, and pelvis OARs with promising contouring accuracy and operation time, which is now ready to be applied in our clinical practice. Moreover, this framework is being developed toward a physician-specific framework to comply with the contour style of each radiation oncologist separately.

Lung Segmentation and Characterization in COVID-19 Patients for Assessing Pulmonary Thromboembolism: An Approach Based on Deep Learning and Radiomics

The COVID-19 pandemic is inevitably changing the world in a dramatic way, and the role of computed tomography (CT) scans can be pivotal for the prognosis of COVID-19 patients. Since the start of the pandemic, great care has been given to the relationship between interstitial pneumonia caused by the infection and the onset of thromboembolic phenomena. In this preliminary study, we collected n = 20 CT scans from the Polyclinic of Bari, all from patients positive with COVID-19, nine of which developed pulmonary thromboembolism (PTE). For eight CT scans, we obtained masks of the lesions caused by the infection, annotated by expert radiologists; whereas for the other four CT scans, we obtained masks of the lungs (including both healthy parenchyma and lesions). We developed a deep learning-based segmentation model that utilizes convolutional neural networks (CNNs) in order to accurately segment the lung and lesions. By considering the images from publicly available datasets, we also realized a training set composed of 32 CT scans and a validation set of 10 CT scans. The results obtained from the segmentation task are promising, allowing to reach a Dice coefficient higher than 97%, posing the basis for analysis concerning the assessment of PTE onset. We characterized the segmented region in order to individuate radiomic features that can be useful for the prognosis of PTE. Out of 919 extracted radiomic features, we found that 109 present different distributions according to the Mann–Whitney U test with corrected p-values less than 0.01. Lastly, nine uncorrelated features were retained that can be exploited to realize a prognostic signature.