Lung CT Research Articles

Deep learning-based computerized diagnosis of lung cancer

The Deep-Learning (DL) technique is capturing increasingly flexible in the sector of processing medical images. Rapid and precise lung cancer detection requirements a standardized computer-aided diagnostic (CAD) architecture. For a quick and reliable detection of lung cancer, a standardized CAD framework is required. High-risk patients are advised by the National Lung Screening Trial to undertake standard screenings with low-dose CT to support the early detection of cancer and decrease the consequence of lung cancer death. In this paper, a lung CT scan and probabilistic bilateral convolutional neural networks (PB-CNN)-based automated diagnosis system for lung cancer are developed. The PB-CNN models were trained using sample cases from the LUNA16 dataset. We used existing techniques, such as Decision Trees (DT), Artificial Neural Networks (ANN) and K-Nearest Neighbors (KNN) to detect lung cancer. We employed accuracy, precision, recall, and f-measure in our experimental investigation. The proposed PB-CNN is automatically detecting lung cancer, yielding an acceptable performance.

AN INTELLIGENT MODEL FOR BENIGN AND MALIGNANT PULMONARY NODULE ANALYSIS USING U-NET NETWORKS AND MULTILEVEL ATTENTION MECHANISMS

A key study area throughout the medical sector for image processing and analysis is medical image segmentation. The diagnosis and treatment strategies of doctors may have a solid foundation owing to accurate and effective medical image segmentation. Conventional approaches in this field rely on manual feature extraction, which makes segmentation complex, costs doctors’ time and energy, and involves a subjective evaluation that is readily susceptible to diagnostic errors. Researchers have applied convolutional neural network-based deep learning techniques to the segmentation of medical images as a result of their impressive advancements and successes in the field of computer vision. The research described here uses the U-Net network’s outstanding feature learning capabilities and end-to-end processing mode for lung CT image segmentation via fully convolutional network (FCN) research. However, focusing on valuable, crucial information aspects in the U-Net network is challenging. This study employed multilevel attention mechanisms on the basis of U-Net networks to enhance the model’s accuracy in lung CT image segmentation. These mechanisms were inspired by attention mechanisms. By improving the segmentation accuracy and optimizing the segmentation effect, the new model embeds a self-attention module in front of each upsampling layer in the U-Net model. This module provides more detailed information by stitching the self-attention module of the original image and then suppresses irrelevant and redundant information by using the effect of feature extraction of the upsampling layer. Several additional comparative experiments were conducted on the 2019nCoVR dataset. The outcomes demonstrate the efficacy of the optimized model described in this paper and its application results in improved segmentation effects in lung CT images. Additionally, the new model has distinct advantages over existing approaches that are typical of medical image segmentation, which represent its higher level of lung CT image segmentation.

Application of improved Unet network in the recognition and segmentation of lung CT images in patients with pneumoconiosis

BackgroundPneumoconiosis has a significant impact on the quality of patient survival. This study aims to evaluate the performance and application value of improved Unet network technology in the recognition and segmentation of lesion areas of lung CT images in patients with pneumoconiosis.MethodsA total of 1212 lung CT images of patients with pneumoconiosis were retrospectively included. The improved Unet network was used to identify and segment the CT image regions of the patients’ lungs, and the image data of the granular regions of the lungs were processed by the watershed and region growing algorithms. After random sorting, 848 data were selected into the training set and 364 data into the validation set. The experimental dataset underwent data augmentation and were used for model training and validation to evaluate segmentation performance. The segmentation results were compared with FCN-8s, Unet network (Base), Unet (Squeeze-and-Excitation, SE + Rectified Linear Unit, ReLU), and Unet + + networks.ResultsIn the segmentation of lung CT granular region with the improved Unet network, the four evaluation indexes of Dice similarity coefficient, positive prediction value (PPV), sensitivity coefficient (SC) and mean intersection over union (MIoU) reached 0.848, 0.884, 0.895 and 0.885, respectively, increasing by 7.6%, 13.3%, 3.9% and 6.4%, respectively, compared with those of Unet network (Base), and increasing by 187.5%, 249.4%, 131.9% and 51.0%, respectively, compared with those of FCN-8s, and increasing by 14.0%, 31.2%, 4.7% and 9.7%, respectively, compared with those of Unet network (SE + ReLU), while the segmentation performance was also not inferior to that of the Unet + + network.ConclusionsThe improved Unet network proposed shows good performance in the recognition and segmentation of abnormal regions in lung CT images in patients with pneumoconiosis, showing potential application value for assisting clinical decision-making.

Advanced lung imaging with photon-counting detectors: Insights from thermoluminescence dosimetry

Rationale and ObjectivesThis study investigates the dose burden of photon-counting detector (PCD) lung CT with ultra-high-resolution (UHR) and standard mode using organ-based tube current modulation (OBTCM). Materials and MethodsAn anthropomorphic Alderson-Rando phantom was scanned in UHR and standard mode with and without OBTCM on three dose levels (IQ 5, 20, 50). Effective radiation dose was determined by thermoluminescent dosimetry in 13 measurement sites and compared with the calculated effective dose derived from the dose-length product. Image quality was evaluated subjectively by six radiologists using an equidistant 7-point scale and objectively by means of modulation transfer function analysis. ResultsMeasured effective radiation exposure was lower in UHR and OBTCM studies than in standard mode (IQ 5: 0.34–0.36, IQ 20: 1.57–1.70, IQ 50: 3.76–3.99 mSv). Compared with the “calculated effective dose”, the radiation exposure measured with thermoluminescence dosimetry was 131–170% higher. Noise in UHR mode was rated lower than in standard (all p≤0.042) and OBTCM images (all p≤0.028) for all dose levels, while image sharpness was deemed highest for UHR protocols (all p≤0.042). The use of OBTCM had no significant effect on either dimension of subjective image quality (all p≥0.999). Modulation transfer function analysis confirmed the highest spatial frequency in UHR datasets (all p≤0.016). ConclusionIn PCD-CT of the lung, full field-of-view UHR imaging entails no dose disadvantage over standard mode despite superior image quality. OBTCM possesses moderate dose saving potential. Thermoluminescence dosimetry yielded considerably higher effective doses than those calculated from dose-length products.

Deep Learning Radiomics Features of Mediastinal Fat and Pulmonary Nodules on Lung CT Images Distinguish Benignancy and Malignancy.

This study investigated the relationship between mediastinal fat and pulmonary nodule status, aiming to develop a deep learning-based radiomics model for diagnosing benign and malignant pulmonary nodules. We proposed a combined model using CT images of both pulmonary nodules and the fat around the chest (mediastinal fat). Patients from three centers were divided into training, validation, internal testing, and external testing sets. Quantitative radiomics and deep learning features from CT images served as predictive factors. A logistic regression model was used to combine data from both pulmonary nodules and mediastinal adipose regions, and personalized nomograms were created to evaluate the predictive performance. The model incorporating mediastinal fat outperformed the nodule-only model, with C-indexes of 0.917 (training), 0.903 (internal testing), 0.942 (external testing set 1), and 0.880 (external testing set 2). The inclusion of mediastinal fat significantly improved predictive performance (NRI = 0.243, p < 0.05). A decision curve analysis indicated that incorporating mediastinal fat features provided greater patient benefits. Mediastinal fat offered complementary information for distinguishing benign from malignant nodules, enhancing the diagnostic capability of this deep learning-based radiomics model. This model demonstrated strong diagnostic ability for benign and malignant pulmonary nodules, providing a more accurate and beneficial approach for patient care.

Texture feature-aware consistency for semi-supervised honeycomb lung lesion segmentation

Accurate segmentation of honeycomb lung lesions from lung CT images is crucial for the clinical diagnosis of a wide range of lung diseases. Existing semi-supervised segmentation methods ignore the rich texture feature information in honeycomb lung CT images, resulting in inaccurate predictions in the lesion region. Therefore, this paper proposes a semi-supervised honeycomb lung lesion segmentation framework using texture feature-aware consistency. The framework consists of a shared encoder, a main decoder, and three auxiliary decoders. The different outputs produced by the decoders are used to compute the Grey Level Co-occurrence Matrix and extract texture features, respectively, and further to compute the loss of texture feature-aware consistency. In pseudo-labeling supervision, the uncertainty of each output is computed to bootstrap the pseudo-labeling supervision loss. For the small lesion recognition omission problem, a pseudo-label refinement module is constructed to reuse the multi-scale features of the encoder, focusing on small-scale lesions, and refining the pseudo-labels. Experimental results show that the proposed method outperforms other methods in the honeycomb lung segmentation task, achieving a Dice coefficient of 77.62 using 20% labeled data. Additionally, experiments on publicly available datasets further demonstrate the generalizability of the proposed method. The code is available at https://github.com/Oran9er/TFC-Net-main.

Integrating cardiovascular risk assessment into mobile low-dose CT lung screenings in rural Appalachia: A comprehensive analysis of the relationship between lung cancer risk, coronary artery calcium burden, and cardiovascular risk reduction strategies

ObjectiveMobile low-dose computed tomography (LDCT) lung screenings are part of an outreach program in rural Appalachia to detect early lung cancer. Coronary artery calcium (CAC) scoring on LDCT can identify calcium deposits in coronary arteries and can prompt consideration of risk modification for prevention of cardiovascular disease (CVD) events. It is not known if Lung CT Screening Reporting & Data System (Lung-RADS) scoring correlates with CAC scores. There is no clear guidance for patients undergoing LDCT screenings to receive follow-up regarding CAC or prevention of associated CVD risk. MethodsThis was a retrospective review of mobile LDCT LCS in adults with no known history of CVD. CT images were obtained at 100 kVp with a slice thickness of 3 mm. Agatston CAC scoring was performed retroactively. Lung-RADS scores were categorized as: Negative (1), Benign (2), Probably Benign (3), and Suspicious (4). CAC scoring was grouped as 0, 1–100, 101–399, and ≥400. Descriptive statistics and chi-square analyses were utilized. ResultsA total of 526 LDCT screenings were included. Over 54 % of patients had coronary calcification on LDCT LCS. 161 patients (30.6 %) had a CAC score of ≥100 and 75 patients (14.3 %) had a CAC score ≥400. Of patients with a CAC score ≥100, 7.5 % received referrals for follow-up after the LDCT screen and 9.3 % had additional cardiac testing. Of those with a CAC score ≥100 not already on a statin (45.3 %) and not already on aspirin (63.3 %), few were started within 3 months of LDCT for prevention (8.2 % and 5.9 % respectively). Among patients with a Lung-RADS score of 4, 17 % had a CAC score >400, whereas only 12 % with a Lung-RADS score of 1 fell into the same CAC category. Higher Lung-RADS scores correlated with fewer patients with CAC of 0. A significant correlation was observed between higher Lung-RADS scores and elevated CAC scores (p = 0.02). ConclusionIn patients with no CVD history, coronary artery calcification was frequently identified on mobile LDCT lung screenings in rural communities. Patients with higher probabilities of malignant lung nodules may also be at increased risk for significant coronary artery disease. Calcium scoring from LDCT screenings allowed for simultaneous assessment of lung cancer and CVD risk. Unfortunately, few referrals or CVD prevention medications were initiated. Awareness of CAC score utility, follow-up for identified coronary calcifications, and consideration of primary prevention medications when indicated, would be beneficial in patients undergoing LDCT lung screenings, especially in rural areas with limited healthcare access.

Drug-induced liver injury associated with savolitinib: a novel case report and causality assessment

BackgroundSavolitinib, a small molecule inhibitor, has gained approval as the inaugural medication in China that specifically targets MET kinase. Patients with advanced non-small cell lung cancer (NSCLC) who show MET exon 14 skipping now have a new and innovative treatment option available.Case reportIn this case report, we describe a patient who experienced drug-induced liver injury (DILI) due to the administration of savolitinib. After being prescribed with savolitinib (400 mg per day, oral), a 73-year-old male diagnosed with stage IV NSCLC with MET exon 14 skipping mutation experienced an increase in liver enzymes and bilirubin levels according to his laboratory tests conducted one month later. Following a 14-day course of hepatoprotective medication, the liver function reverted back to its normal state. After receiving savolitinib (200 mg per day, oral) for one week, the patient was once again diagnosed with severe liver impairment. Then savolitinib was discontinued and received treatment with hepatoprotective drugs for one week. Following the restoration of normal liver function, another attempt was made to administer a small amount of savolitinib (100 mg per day, oral). Thus far, the patient has been followed up and there has been no recurrence of liver damage. Additionally, the lung CT scan revealed ongoing tumor shrinkage with no apparent indications of spreading or metastasis. The Roussel Uclaf Causality Assessment Method (RUCAM) determined that savolitinib was “highly probable” cause of DILI. Moderate-severe was determined to be the extent of DILI severity.ConclusionTo the best of our understanding, this is the initial instance of DILI resulting from the use of savolitinib as a standalone treatment in a real-world setting. During the administration of savolitinib, healthcare professionals should carefully consider the potential occurrence of DILI. Administering the patient with a small amount of savolitinib resulted in a remarkable response against the tumor, leading us to speculate that the effectiveness of savolitinib might be associated with its plasma concentration. Studying the pharmacokinetics and pharmacodynamics (PK/PD) of savolitinib is beneficial for tailoring and accurately prescribing the medication to each individual.

Compound heterozygous CFTR variants (Q1352H and 5T; TG13) in a Chinese patient with cystic fibrosis

Cystic fibrosis (CF) is an autosomal recessive inherited disease caused by variants of cystic fibrosis transmembrane conductance regulation (CFTR) gene. This report presents a case of a Chinese boy diagnosed with CF, attributed to the presence of two specific CFTR gene variations: 4056G > C (NM_000492.4) (p.Gln1352His, legacy: Q1352H) and c.1210-34TG[13]T[5] (NM_000492.4)(legacy: 5T; TG13). A ten-year-old boy was admitted to the hospital due to recurrent pneumonia, cough, and intermittent fever for seven years. Lung auscultation revealed rales, and a lung CT scan indicated parenchymal transformation with infection in both lungs. Whole Exome Sequencing (WES) identified two CFTR gene variants, Q1352H and 5T; TG13, which were significantly associated with clinical phenotype. Following a two-year course of azithromycin combined with inhalation therapy with budesonide, the patient experienced no further episodes of respiratory infections. Moreover, significant improvements were observed in pulmonary function, pulmonary infection, and bronchiectasis. The occurrence of combined variations, Q1352H and 5T; TG13, in the CFTR gene is rare and specific to Chinese populations. WES proves to be a valuable diagnostic tool for detecting CFTR gene variants.

Novel Deep CNNs Explore Regions, Boundaries, and Residual Learning for COVID-19 Infection Analysis in Lung CT.

COVID-19 poses a global health crisis, necessitating precise diagnostic methods for timely containment. However, accurately delineating COVID-19-affected regions in lung CT scans is challenging due to contrast variations and significant texture diversity. In this regard, this study introduces a novel two-stage classification and segmentation CNN approach for COVID-19 lung radiological pattern analysis. A novel Residual-BRNet is developed to integrate boundary and regional operations with residual learning, capturing key COVID-19 radiological homogeneous regions, texture variations, and structural contrast patterns in the classification stage. Subsequently, infectious CT images undergo lesion segmentation using the newly proposed RESeg segmentation CNN in the second stage. The RESeg leverages both average and max-pooling implementations to simultaneously learn region homogeneity and boundary-related patterns. Furthermore, novel pixel attention (PA) blocks are integrated into RESeg to effectively address mildly COVID-19-infected regions. The evaluation of the proposed Residual-BRNet CNN in the classification stage demonstrates promising performance metrics, achieving an accuracy of 97.97%, F1-score of 98.01%, sensitivity of 98.42%, and MCC of 96.81%. Meanwhile, PA-RESeg in the segmentation phase achieves an optimal segmentation performance with an IoU score of 98.43% and a dice similarity score of 95.96% of the lesion region. The framework's effectiveness in detecting and segmenting COVID-19 lesions highlights its potential for clinical applications.

CBCT-based synthetic CT image generation using a diffusion model for CBCT-guided lung radiotherapy.

Although cone beam computed tomography (CBCT) has lower resolution compared to planning CTs (pCT), its lower dose, higher high-contrast resolution, and shorter scanning time support its widespread use in clinical applications, especially in ensuring accurate patient positioning during the image-guided radiation therapy (IGRT) process. While CBCT is critical to IGRT, CBCT image quality can be compromised by severe stripe and scattering artifacts. Tumor movement secondary to respiratory motion also decreases CBCT resolution. In order to improve the image quality of CBCT, we propose a Lung Diffusion Model (L-DM) framework. Our proposed algorithm is based on a conditional diffusion model trained on pCT and deformed CBCT (dCBCT) image pairs to synthesize lung CT images from dCBCT images and benefit CBCT-based radiotherapy. dCBCT images were used as the constraint for the L-DM. The image quality and Hounsfield unit (HU) values of the synthetic CTs (sCT) images generated by the proposed L-DM were compared to three selected mainstream generation models. We verified our model in both an institutional lung cancer dataset and a selected public dataset. Our L-DM showed significant improvement in the four metrics of mean absolute error (MAE), peak signal-to-noise ratio (PSNR), normalized cross-correlation (NCC), and structural similarity index measure (SSIM). In our institutional dataset, our proposed L-DM decreased the MAE from 101.47 to 37.87HU and increased the PSNR from 24.97 to 29.89 dB, the NCC from 0.81 to 0.97, and the SSIM from 0.80 to 0.93. In the public dataset, our proposed L-DM decreased the MAE from 173.65 to 58.95HU, while increasing the PSNR, NCC, and SSIM from 13.07 to 24.05 dB, 0.68 to 0.94, and 0.41 to 0.88, respectively. The proposed L-DM significantly improved sCT image quality compared to the pre-correction CBCT andthree mainstream generative models. Our model can benefit CBCT-based IGRT and other potential clinical applications as it increases the HU accuracy and decreases the artifacts from input CBCT images.

Using Telehealth to Increase Lung Cancer Screening Referrals for At-Risk Veterans in Rural Communities.

At-risk rural veterans have low rates of lung cancer screening (LCS). This proof-of-principle quality improvement project aimed to determine whether a telehealth intervention would increase referrals for at-risk veterans living in the rural upper Midwest and attending a smoking cessation program to LCS with low-dose computed tomography (LDCT) of the chest. Sixty-eight of 74 LCS-eligible rural veterans who self-enrolled in a smoking cessation program were contacted by telephone. Those who agreed to enroll in LCS were referred to LDCT and followed for 4 months. At the conclusion of the intervention, the number of referrals and screenings performed were tabulated. LDCT reports were reviewed and scored according to Lung CT Screening Reporting and Data System (Lung-RADS) version 1.1. Only 3 of 74 LCS-eligible veterans (4%) underwent LDCT before initiation of this telehealth intervention. By the conclusion of this 4-month project, 19 of 74 veterans (26%) underwent LDCT. Forty-one veterans were successfully contacted and 29 agreed to participate in LCS. Of those who agreed to participate, 19 underwent LDCT within 4 months. Of the veterans who received LDCT, 10 were diagnosed with Lung-RADS 1, 7 with Lung-RADS 2, 1 with Lung-RADS 3, and 1 with Lung-RADS 4B. Annual follow-up LDCT or referral for further evaluation were pursued in each case. Collectively, these data suggest that telehealth intervention could increase referrals of at-risk rural veterans to a centralized LCS program at a regional US Department of Veterans Affairs medical facility.

Assessing Thoracic Vertebral Bone Mineral Density (T8-T10) for Osteoporosis Diagnosis During CT Lung Cancer Screening in Older Adults.

Osteoporosis diagnosis often utilizes quantitative computed tomography (QCT). This study explored the validity of applying lumbar bone mineral density (LBMD) standards to thoracic vertebrae (T8-T10) for osteoporosis detection during CT lung cancer screenings. This study investigated the utility of thoracic BMD (BMD-T8-T10) for detecting osteoporosis in older persons during CT lung cancer screening. We studied 701 participants who underwent QCT scans for both LBMD and BMD-T8-T10. Osteoporosis was diagnosed using ACR criteria based on LBMD. We determined BMD-T8-T10 thresholds via a receiver operating characteristic (ROC) curve and translated BMD-T8+T9+T10 to LBMD (TTBMD) using linear regression. Kappa test was used to evaluate the accuracy of BMD-T8-T10 thresholds and TTBMD in diagnosing osteoporosis. Raw BMD-T8-T10 poorly identified osteoporosis (kappa = 0.51). ROC curve analysis identified BMD-T8-T10 thresholds for osteopenia (138 mg/cm3) and osteoporosis (97 mg/cm3) with areas under the curve of 0.97 and 0.99, respectively. We normalized BMD-T8-T10 to TTBMD based on the formula: TTBMD = 0.9 × BMD-T8-T10 - 2.56. These thresholds (kappa = 0.74) and TTBMD performed well in detecting osteoporosis/osteopenia (kappa = 0.74). Both calculating BMD-T8-T10 threshold (138.0 mg/cm3 for osteopenia and 97 mg/cm3 for osteoporosis) and normalizing BMD-T8-T10 to LBMD demonstrated good performance in identifying osteoporosis in older adults during CT lung cancer screening.

Study on lung CT image segmentation algorithm based on threshold-gradient combination and improved convex hull method

Lung images often have the characteristics of strong noise, uneven grayscale distribution, and complex pathological structures, which makes lung image segmentation a challenging task. To solve this problems, this paper proposes an initial lung mask extraction algorithm that combines threshold and gradient. The gradient used in the algorithm is obtained by the time series feature extraction method based on differential memory (TFDM), which is obtained by the grayscale threshold and image grayscale features. At the same time, we also proposed a lung contour repair algorithm based on the improved convex hull method to solve the contour loss caused by solid nodules and other lesions. Experimental results show that on the COVID-19 CT segmentation dataset, the advanced lung segmentation algorithm proposed in this article achieves better segmentation results and greatly improves the consistency and accuracy of lung segmentation. Our method can obtain more lung information, resulting in ideal segmentation effects with improved accuracy and robustness.

Shape prior-constrained deep learning network for medical image segmentation

We propose a shape prior representation-constrained multi-scale features fusion segmentation network for medical image segmentation, including training and testing stages. The novelty of our training framework lies in two modules comprised of the shape prior constraint and the multi-scale features fusion. The shape prior learning model is embedded into a segmentation neural network to solve the problems of low contrast and neighboring organs with intensities similar to the target organ. The latter can provide both local and global contexts to address the issues of large variations in patient postures as well as organ’s shape. In the testing stage, we propose a circular collaboration framework strategy which combines a shape generator auto-encoder network model with a segmentation network model, allowing the two models to collaborate with each other, resulting in a cooperative effect that leads to accurate segmentations. Our proposed method is evaluated and demonstrated on the ACDC MICCAI’17 Challenge Dataset, CT scans datasets, namely, in COVID-19 CT lung, and LiTS2017 liver from three different datasets, and its results are compared with the recent state of the art in these areas. Our method ranked 1st on the ACDC Dataset in terms of Dice score and achieved very competitive performance on COVID-19 CT lung and LiTS2017 liver segmentation.

Improving lung nodule segmentation in thoracic CT scans through the ensemble of 3D U-Net models.

The current study explores the application of 3D U-Net architectures combined with Inception and ResNet modules for precise lung nodule detection through deep learning-based segmentation technique. This investigation is motivated by the objective of developing a Computer-Aided Diagnosis (CAD) system for effective diagnosis and prognostication of lung nodules in clinical settings. The proposed method trained four different 3D U-Net models on the retrospective dataset obtained from AIIMS Delhi. To augment the training dataset, affine transformations and intensity transforms were utilized. Preprocessing steps included CT scan voxel resampling, intensity normalization, and lung parenchyma segmentation. Model optimization utilized a hybrid loss function that combined Dice Loss and Focal Loss. The model performance of all four 3D U-Nets was evaluated patient-wise using dice coefficient and Jaccard coefficient, then averaged to obtain the average volumetric dice coefficient (DSCavg) and average Jaccard coefficient (IoUavg) on a test dataset comprising 53 CT scans. Additionally, an ensemble approach (Model-V) was utilized featuring 3D U-Net (Model-I), ResNet (Model-II), and Inception (Model-III) 3D U-Net architectures, combined with two distinct patch sizes for further investigation. The ensemble of models obtained the highest DSCavg of 0.84 ± 0.05 and IoUavg of 0.74 ± 0.06 on the test dataset, compared against individual models. It mitigated false positives, overestimations, and underestimations observed in individual U-Net models. Moreover, the ensemble of models reduced average false positives per scan in the test dataset (1.57 nodules/scan) compared to individual models (2.69-3.39 nodules/scan). The suggested ensemble approach presents a strong and effective strategy for automatically detecting and delineating lung nodules, potentially aiding CAD systems in clinical settings. This approach could assist radiologists in laborious and meticulous lung nodule detection tasks in CT scans, improving lung cancer diagnosis and treatment planning.

A position-enhanced sequential feature encoding model for lung infections and lymphoma classification on CT images.

Differentiating pulmonary lymphoma from lung infections using CT images is challenging. Existing deep neural network-based lung CT classification models rely on 2D slices, lacking comprehensive information and requiring manual selection. 3D models that involve chunking compromise image information and struggle with parameter reduction, limiting performance. These limitations must be addressed to improve accuracy and practicality. We propose a transformer sequential feature encoding structure to integrate multi-level information from complete CT images, inspired by the clinical practice of using a sequence of cross-sectional slices for diagnosis. We incorporate position encoding and cross-level long-range information fusion modules into the feature extraction CNN network for cross-sectional slices, ensuring high-precision feature extraction. We conducted comprehensive experiments on a dataset of 124 patients, with respective sizes of 64, 20 and 40 for training, validation and testing. The results of ablation experiments and comparative experiments demonstrated the effectiveness of our approach. Our method outperforms existing state-of-the-art methods in the 3D CT image classification problem of distinguishing between lung infections and pulmonary lymphoma, achieving an accuracy of 0.875, AUC of 0.953 and F1 score of 0.889. The experiments verified that our proposed position-enhanced transformer-based sequential feature encoding model is capable of effectively performing high-precision feature extraction and contextual feature fusion in the lungs. It enhances the ability of a standalone CNN network or transformer to extract features, thereby improving the classification performance. The source code is accessible at https://github.com/imchuyu/PTSFE .

Prediction of early-phase cytomegalovirus pneumonia in post-stem cell transplantation using a deep learning model.

Diagnostic challenges exist for CMV pneumonia in post-hematopoietic stem cell transplantation (post-HSCT) patients, despite early-phase radiographic changes. The study aims to employ a deep learning model distinguishing CMV pneumonia from COVID-19 pneumonia, community-acquired pneumonia, and normal lungs post-HSCT. Initially, 6 neural network models were pre-trained with COVID-19 pneumonia, community-acquired pneumonia, and normal lung CT images from Kaggle's COVID multiclass dataset (Dataset A), then Dataset A was combined with the CMV pneumonia images from our center, forming Dataset B. We use a few-shot transfer learning strategy to fine-tune the pre-trained models and evaluate model performance in Dataset B. 34 cases of CMV pneumonia were found between January 2018 and December 2022 post-HSCT. Dataset A contained 1681 images of each subgroup from Kaggle. Combined with Dataset A, Dataset B was initially formed by 98 images of CMV pneumonia and normal lung. The optimal model (Xception) achieved an accuracy of 0.9034. Precision, recall, and F1-score all reached 0.9091, with an AUC of 0.9668 in the test set of Dataset B. This framework demonstrates the deep learning model's ability to distinguish rare pneumonia types utilizing a small volume of CT images, facilitating early detection of CMV pneumonia post-HSCT.

Correlation between Severity of Lung Disease in COVID-19 Infection and Acute CNS Abnormalities

Abstract Background Sever acute respiratory syndrome corona virus 2 (SARS-COV 2) began in Wuhan,China,in December 2019 and has rapidely spread around the world to become pandemic,as COVID-19 can produce pathologic changes in chest imaging,COVID-19 related brain imaging findings are also well known. Aim of the Work The aim of this current study is to correlate retrospectively between the severity of lung affection in patients with COVID-19 infections and development of acute CNS abnormalities. Patients and Methods This study is Retrospective study, the study was conducted at Ain Shams University Hospitals over a course of six months from march 2022 to September 2022 and the main source of data for this study was the archived CT and MRI imaging of the patients referred to the department of Radiology. Our inclusion criteria were patients that diagnosed with sever acute respiratory syndrome corona virus 2 infection with acute neurologic manifestation and available chest CT and brain imaging. The 5 lobes of the lungs were individually scored on a scale 0-5 (0 correspond to no involvement and 5 &amp;gt; 75% involvement),A CT lung severity score was determined as the sum of lung involvement,ranging from 0 (no involvement) to 25 (maximum involvement). Results A total 55 patients met the inclusion criteria,In attempt to describe and evaluate the brain and lung imaging correlation in patients with COVID-19, Compared to 20 patients without acute abnormal findings on neuroimaging,there were 35 patients with acute abnormal findings on neuroimaging and those patients were found to have significantly higher mean CT lung severity score (p &amp;lt; 0.0001) and they are more likely to present with ischemic stroke (14, 40%), A threshold of CT lung severity score &amp;gt; 18 was found to be 97% sensitive and 90% specific for acute abnormal findings on neuroimaging, the neuroimaging hallmark of these patients were acute ischemic infarcts (40%),invasive fungal sinusitis (22.9.%), Isolated cavernous sinus thrombosis,PRES,and vasculitis (each one 8.6%),cerebral venous sinus thrombosis other than cavernous sinus (5.7%),cerebritis and abnormal white matter signal (each one 2.8%).The predominant CT chest findings were peripheral ground -glass opacities with or with out consolidation. Conclusion The CT lung severity score may be predictive of acute abnormalities on neuroimaging in patients with COVID-19 with neurologic manifestation. This can be used as predictive tool in patients management to improve clinical outcome.

Yorkshire Lung Screening Trial (YLST) pathway navigation study: a protocol for a nested randomised controlled trial to evaluate the effect of a pathway navigation intervention on lung cancer screening uptake

IntroductionLung cancer is the most common cause of cancer death globally. In 2022 the UK National Screening Committee recommended the implementation of a national targeted lung cancer screening programme, aiming...