Thoracic CT Images Research Articles

Under pressure – Association of the arm position and leading circulatory structure behind the pressure point in cardiopulmonary resuscitation patients

BackgroundThoracic computed tomography scans (CT) are used by several study groups to investigate the circulatory structures (heart and vessels) located behind the pressure point for chest compressions. Yet, it remains unclear how the positioning of these structures is influenced by factors such as intubation, the respiratory cycle and arm positioning. MethodsWe retrospectively analyzed data of adult patients with in- or out-of-hospital cardiac arrest who underwent thoracic CT imaging within one year before or up to six months after arrest. A region of interest (ROI) behind the pressure point was defined. The largest structure within this region was defined as “leading circulatory structure”, which was the primary outcome. Airway status (intubated versus spontaneous breathing), respiratory cycle (inspiration, expiration, resting expiratory position), and arm position (up over the head versus down beside the trunk) served as covariates in an ordinal regression model. ResultsAmong 500 initially screened patients, 411 (82.2 %) were included in the analysis. There was a significant association between the arm position and the leading circulatory structure behind the pressure point. However, no association was found with airway status or respiratory cycle. The most frequently identified leading circulatory structure was the left atrium (arms up: 41.8 %, down: 50.7 %), followed by the ascending aorta (up: 23.8 % vs. down: 16.7 %). The left ventricle was the leading structure in only one case (0.2 %, arms down). ConclusionThis study shows that arm position is significantly associated with the leading circulatory structure behind the pressure point for chest compressions in cardiac arrest.

Quantum-enhanced hybrid feature engineering in thoracic CT image analysis for state-of-the-art nodule classification: an advanced lung cancer assessment

The intricate nature of lung cancer treatment poses considerable challenges upon diagnosis. Early detection plays a pivotal role in mitigating its escalating global mortality rates. Consequently, there are pressing demands for robust and dependable early detection and diagnostic systems. However, the technological limitations and complexity of the disease make it challenging to implement an efficient lung cancer screening system. AI-based CT image analysis techniques are showing significant contributions to the development of computer-assisted detection (CAD) systems for lung cancer screening. Various existing research groups are working on implementing CT image analysis systems for assessing and classifying lung cancer. However, the complexity of different structures inside the CT image is high and comprehension of significant information inherited by them is more complex even after applying advanced feature extraction and feature selection techniques. Traditional and classical feature selection techniques may struggle to capture complex interdependencies between features. They may get stuck in local optima and sometimes require additional exploration strategies. Traditional techniques may also struggle with combinatorial optimization problems when applied to a prominent feature space. This paper proposed a methodology to overcome the existing challenges by applying feature extraction using Vision Transformer (FexViT) and Feature selection using the Quantum Computing based Quadratic unconstrained binary optimization (QC-FSelQUBO) technique. This algorithm shows better performance when compared with other existing techniques. The proposed methodology showed better performance as compared to other existing techniques when evaluated by applying necessary output measures, such as accuracy, Area under roc (receiver operating characteristics) curve, precision, sensitivity, and specificity, obtained as 94.28%, 99.10%, 96.17%, 90.16% and 97.46%. The further advancement of CAD systems is essential to meet the demand for more reliable detection and diagnosis of cancer, which can be addressed by leading the proposed quantum computation and growing AI-based technology ahead.

Uncertain prediction of deformable image registration on lung CT using multi-category features and supervised learning.

The assessment of deformable registration uncertainty is an important task for the safety and reliability of registration methods in clinical applications. However, it is typically done by a manual and time-consuming procedure. We propose a novel automatic method to predict registration uncertainty based on multi-category features and supervised learning. Three types of features, including deformation field statistical features, deformation field physiologically realistic features, and image similarity features, are introduced and calculated to train the random forest regressor for local registration uncertain prediction. Deformation field statistical features represent the numerical stability of registration optimization, which are correlated to the uncertainty of deformation fields; deformation field physiologically realistic features represent the biomechanical properties of organ motions, which mathematically reflect the physiological reality of deformation; image similarity features reflect the similarity between the warped image and fixed image. The multi-category features comprehensively reflect the registration uncertainty. The strategy of spatial adaptive random perturbations is also introduced to accurately simulate spatial distribution of registration uncertainty, which makes deformation field statistical features more discriminative to the uncertainty of deformation fields. Experiments were conducted on three publicly available thoracic CT image datasets. Seventeen randomly selected image pairs are used to train the random forest model, and 9 image pairs are used to evaluate the prediction model. The quantitative experiments on lung CT images show that the proposed method outperforms the baseline method for uncertain prediction of classical iterative optimization-based registration and deep learning-based registration with different registration qualities. The proposed method achieves good performance for registration uncertain prediction, which has great potential in improving the accuracy of registration uncertain prediction.

Photon-counting CT in Thoracic Imaging: Early Clinical Evidence and Incorporation Into Clinical Practice.

Photon-counting CT (PCCT) is an emerging advanced CT technology that differs from conventional CT in its ability to directly convert incident x-ray photon energies into electrical signals. The detector design also permits substantial improvements in spatial resolution and radiation dose efficiency and allows for concurrent high-pitch and high-temporal-resolution multienergy imaging. This review summarizes (a) key differences in PCCT image acquisition and image reconstruction compared with conventional CT; (b) early evidence for the clinical benefit of PCCT for high-spatial-resolution diagnostic tasks in thoracic imaging, such as assessment of airway and parenchymal diseases, as well as benefits of high-pitch and multienergy scanning; (c) anticipated radiation dose reduction, depending on the diagnostic task, and increased utility for routine low-dose thoracic CT imaging; (d) adaptations for thoracic imaging in children; (e) potential for further quantitation of thoracic diseases; and (f) limitations and trade-offs. Moreover, important points for conducting and interpreting clinical studies examining the benefit of PCCT relative to conventional CT and integration of PCCT systems into multivendor, multispecialty radiology practices are discussed.

Comparison of Initial Thoracic CT Images of COVID-19 Patients with Non-Variant, Alpha, Delta, and Omicron Variants: A Retrospective Study

Background:: CT findings and Ground glass opacity (GGO) volumes may differ between SARS CoV-2 non-variant, alpha, delta, and omicron variants. Objective:: To compare the thoracic CT findings, GGO volumes, and GGOs’ lung uptake rates among patients with COVID-19 variants. Methods:: Thoracic CT images of 83 patients with non-variant, 78 patients with alpha variant, 93 patients with delta variant, and 73 patients with omicron variant having positive Real-Time Polymerase Chain Reaction test results were analyzed retrospectively. GGO volumes and lung volumes were calculated by using the Cavalieri Principle. Differences in CT findings, ground-glass opacity volumes, and lung involvement rates between non-variant and variant groups were evaluated. Results:: There were significant differences found in the incidence of GGOs (p < 0.001), air bronchogram (p = 0.007), reticulation (p = 0.002) and subpleural lines, and linear opacities (p = 0.034) between non-variant and variant groups. GGO uptake rates (ground glass opacity volumes × 100 ÷ lung volume) were 8.88% in the non-variant, 4.83% in the alpha variant, 3.50% in the delta variant, and 2.02% in the omicron variant. In estimating variant groups, it was determined that the increase in the rate of GGOs in the right lung increased the probability of having an omicron variant, whereas the presence of nodules decreased it. The possibility of the delta variant increased with an increase in the rate of ground glass opacities in the left lung. Conclusion:: Thoracic CT findings solely can be helpful in distinguishing COVID-19 variants. Decreased frequency of uptake rates of GGOs suggested that the severity of COVID-19 disease was gradually decreasing.

Pulmonary hemorrhage as a presentation of AL amyloidosis secondary to multiple myeloma: a case report

Abstract Background The convergence of pulmonary hemorrhage, pulmonary amyloidosis, and multiple myeloma is uncommon. Amyloidosis can affect the pulmonary parenchyma in a diffuse, tracheobronchial, or parenchymal pattern and may rarely be associated with pulmonary hemorrhage. Additionally, pulmonary amyloidosis is not a frequent manifestation of multiple myeloma. We present a case of a male patient with pulmonary hemorrhage as the initial manifestation of AL pulmonary amyloidosis, and ultimately, confirmation of multiple myeloma through bone marrow biopsy. Case presentation The clinical case involves a 60-year-old male with no significant medical history, who was admitted presenting a clinical picture evolving over 6 months characterized by hemoptoic cough, accompanied by dyspnea, a decrease in functional capacity, and constitutional symptoms. Thoracic CT images revealed multilobar ground-glass opacities with suspected alveolar hemorrhage. In response to this clinical presentation, bronchoalveolar lavage with cytology was performed, revealing the presence of hemosiderin-laden macrophages. Given the complexity of the case, further investigation included a wedge biopsy of the lung. The pathological report indicated an atypical lymphoplasmacytoid proliferation with deposits of eosinophilic amorphous material, suggestive of amyloidosis. Congo red staining confirmed the presence of amyloid material. Elevated Kappa light chains were detected in both serum and urine, with an increased K/L ratio. Immunoglobulins G and M were found to be decreased. As part of the comprehensive assessment, a bone marrow biopsy was conducted, confirming the diagnosis of multiple myeloma with 10% atypical plasma cells. In light of this diagnosis, appropriate treatment has been initiated to address this intricate medical condition effectively. Conclusion The present case report provides an illustrative perspective on an uncommon presentation of pulmonary amyloidosis secondary to multiple myeloma, with the initial manifestation being pulmonary hemorrhage. The findings from both the physical examination and laboratory tests were consistent with pulmonary amyloidosis, and definitive confirmation of the multiple myeloma diagnosis was achieved through bone marrow biopsy. This case highlights the significance of considering pulmonary amyloidosis as a potential cause of hemoptysis, especially in patients with associated risk factors for multiple myeloma. Early recognition of this clinical association is pivotal for precise diagnosis and prompt therapeutic intervention. The complexity of this case underscores the importance of a comprehensive diagnostic approach in unraveling intricate medical conditions.

Quantitative CT Evaluation of Bone Mineral Density in the Thoracic Spine on 18F-Fluorocholine PET/CT Imaging in Patients With Primary Hyperparathyroidism

Introduction: Measurement of bone mineral density (BMD) with quantitative CT (QCT) carries several advantages over other densitometric techniques, including superior assessment of the spine. As most QCT studies evaluated the lumbar spine, measurements of the thoracic spine are limited. We performed QCT analysis of the thoracic spine in a cohort of patients with primary hyperparathyroidism. Materials and methods: This study was a retrospective QCT analysis of the thoracic spine on 18F-fluorocholine PET/CT scans in patients with primary hyperparathyroidism patients between March 2018 and December 2022. Correlations between QCT-derived BMD or Hounsfield units (HU) and demographic data, laboratory parameters, results from histopathological examination after parathyroidectomy and results of DXA imaging were analyzed, when available. Results: In 189 patients, mean QCT-derived BMD at the thoracic spine was 85.6 mg/cm3. Results from recent DXA were available in 122 patients. Mean thoracic QCT-derived BMD and HU were significantly correlated with DXA-derived BMD in lumbar spine, total hip and femoral neck and with the lowest T-score at DXA imaging. Only weak correlations were found with BMI or 18F-fluorocholine uptake, while no significant correlations were found with adenoma weight, PTH or calcium levels. Conclusion: Our study confirms correlation between QCT-derived BMD in the thoracic spine with age and DXA-derived BMD measurements within a population of patients with primary hyperparathyroidism. Establishment of reference BMD values for individual thoracic vertebrae, may allow direct osteoporosis classification on thoracic CT imaging.

Cavitary pulmonary lesion wall thickness, presence of additional nodules, and intralesional contrast enhancement are associated with malignancy in dogs and cats.

To investigate the CT features of cavitary pulmonary lesions and determine their utility to differentiate malignant from benign lesions. This retrospective study included cases from 5 veterinary medical centers between January 1 2010, and December 31, 2020. Inclusion criteria included having a gas-filled cavitary pulmonary lesion on thoracic CT and definitive diagnosis by either cytology or histopathology. Forty-two animals (27 dogs and 15 cats) were included in this study. Medical records systems/imaging databases were searched, and cases meeting inclusion criteria were selected. The CT studies were interpreted by a third-year radiology resident, and findings were reviewed by a board-certified veterinary radiologist. 7 of the 13 lesion characteristics investigated were not statistically associated with the final diagnosis of the lesion, whereas 6 were statistically associated. Those that were associated included the presence of intralesional contrast enhancement, type of intralesional contrast enhancement (heterogenous and homogenous analyzed separately), presence of additional nodules, wall thickness of the lesion at its thickest point, and wall thickness at the thinnest point. Results from the present study showed that thoracic CT imaging of cavitary pulmonary lesions can be used to further refine the list of differential diagnoses. Based on this data set, in lesions that have heterogenous contrast enhancement, additional pulmonary nodules, and wall thickness > 40 mm at their thickest point, it would be reasonable to consider malignant neoplastic disease higher on the list of differentials than other causes.

Clinical Outcomes of Synchronous Lung Stereotactic Body Radiation Therapy Lesions via Single-Isocenter/Multi-Lesion Treatments

Lung cancer patients with synchronous primary or oligometastatic (< 5 lesions) with associated co-morbidities may not retain their treatment position for the extended stereotactic body radiotherapy (SBRT) treatment times with individual isocenter plans for each lesion due to discomfort or shortness of breath. SBRT using a single-isocenter/multi-lesion (SIML) volumetric arc therapy (VMAT) plan with flattening filter free (FFF) beam could significantly shorten overall treatment time, improve patient comfort and compliance and clinic efficiency. We report clinical results of treating multiple lung metastases or synchronous primary lung cancers with SIML lung SBRT. Eighty patients with synchronous primary lung cancers or oligometastatic lung tumors (two, n = 61; three, n = 10; four, n = 4; five n = 5; total treated lesions, n = 193) were simulated with abdominal compression and/or 4D-CT MIP images and treated with a highly-conformal SIML VMAT lung SBRT plans via 3-4 non-coplanar arcs. Common prescriptions were 50-55 Gy/5 fractions and 54 Gy/3 fractions prescribed 70-80% isodose line to the each PTV. Acuros dose calculation for 6MV-FFF beam was used for tissue heterogeneity corrections. RTOG-0618/0813 criteria were used to dose constraints to organs at risk (OAR) and target conformality. Treatment was delivered every other day or twice weekly with CBCT-guidance, adjustments made with 6DOF couch corrections on a medical linear accelerator, and treatment delivery time within 15 minutes. Local control rates and toxicity profile was evaluated using CTCAE v. 5.0 grading for pneumonitis, rib fracture and chest wall pain. All plans met RTOG-0618/0813 requirements for each tumor coverage, dose to OAR including normal lung and ribs. Mean follow up after last fraction of treatment was 16.9 months (range, 1.0-54.2 months). PTV volume ranged from 2.17 to 167.8 cc with mean volume of 16.1 cc. Of the 80 patients treated, 71 had adequate post-treatment thoracic CT imaging to assess local control. Local control was achieved in 167/175 (95.4%) of treated and followed lesions. CTCAE grade 1 asymptomatic pneumonitis was noted on thoracic CT scans in 42/71 (59.2%) of patients and occurred, on average 4.8 months after SBRT. Symptomatic pneumonitis and rib fracture did not occur in any patient. CTCAE grade 2 chest wall pain occurred in 4/80 (5.0%) treated patients and was managed conservatively with over-the-counter NSAIDS or acetaminophen. SIML lung SBRT for synchronous primary lung cancers or multiple lung metastases can be used as a variant to traditional multiple isocenter SBRT or chronologically separate treatment courses, and has excellent local control rates and low toxicity profile in our patient population. It can help improve comfort and compliance of the patients who have difficulty lying still for an extended treatment course, and significantly reduces treatment time via isocenter shifts/repeated CBCTs for image guidance, thus improving clinic efficiency.

Abstract 5766: Prospective evaluation of cell-free DNA fragmentomes for lung cancer detection

Abstract Background: Less than 10% of eligible persons undergo annual lung cancer screening by low-dose computed tomography (LDCT). Greater uptake of LDCT is hampered in part by its cost, inaccessibility, and balance of benefit to risk. A blood-based, low-cost, widely available initial blood test could boost screening participation and improve the net benefit of screening, if it were sensitive for cancer detection and affordable. The DELFI (DNA evaluation of fragments for early interception) technology uses low-coverage, whole-genome sequencing and machine learning to identify patterns of circulating cell-free DNA (cfDNA) fragmentation indicative of cancer. We report initial results of the cfDNA analysis from DELFI-L101 (NCT04825834), a prospective, observational, national case-control study to train and test DELFI classifiers for lung cancer detection. Methods: Eligible participants were adults ≥50 years old with current or previous smoking histories of ≥20 pack-years and recent or planned thoracic CT imaging. At enrollment, medical history was recorded and blood samples were collected for DELFI analysis. A classifier for lung cancer detection was developed using repeated 10-fold cross-validation. A split study approach for the purposes of independent validation of the classifier is forthcoming. Results: The study cohort included 242 patients with lung cancer and 652 individuals without cancer. Study participants largely represented those of a lung cancer screening population, with 45% stage I/IA. Most participants were ≥65 years old with roughly equal proportions of men and women. There was broad representation across lung cancer risk factors among both cases and controls. The cross-validated area under the receiver operator characteristic curve (AUC) was 0.81 for lung cancer detection. AUCs for adenocarcinoma and squamous cell carcinoma were not significantly different, but the AUC for small cell lung cancer was significantly higher than that for adenocarcinoma (p&amp;lt;.001) and squamous cell carcinoma (p=.02). Clinically meaningful sensitivity to detect all stages of disease was achieved. Conclusions: A classifier developed using samples collected prospectively distinguished between lung cancer cases and controls with robust cross-validated performance across all stages and lung cancer subtypes. A cfDNA DELFI fragmentome test could represent an affordable, high-performing blood test that may improve lung cancer screening. Citation Format: Peter J. Mazzone, Kwok-Kin Wong, Jun-Chieh J. Tsay, Harvey I. Pass, Anil Vachani, Allison Ryan, Jacob Carey, Debbie Jakubowski, Tony Wu, Yuhua Zong, Carter Portwood, Keith Lumbard, Joseph Catallini, Nicholas C. Dracopoli, Tara Maddala, Peter B. Bach, Robert B. Scharpf, Victor E. Velculescu. Prospective evaluation of cell-free DNA fragmentomes for lung cancer detection. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5766.

Clinical observation of neutralizing antibody DXP-604 for treatment critical COVID-19 patients infected with SARS-CoV-2 Delta variant.

From January to March 2022, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Delta (B.1.617.2) infection was prevalent in Yuzhou and Zhengzhou. DXP-604 is a broad-spectrum antiviral monoclonal antibody, which has excellent viral neutralization ability in vitro and a long half-life in vivo, with good biosafety and tolerability. Preliminary results showed that DXP-604 can accelerate recovery from Coronavirus disease 2019 (COVID-19) caused by SARS-CoV-2 Delta variant in hospitalized patients with mild to moderate clinical symptoms. However, the efficacy of DXP-604 has not been fully studied in high-risk severe patients. Here, we prospectively enrolled 27 high-risk patients, two groups were divided, in addition to receiving standard of care (SOC), 14 of them additionally received the neutralizing antibody DXP-604 therapy, and another 13 intensive care unit (ICU) patients simultaneously underwent SOC as a control group matched for age, gender, and clinical type. The results revealed lower C-reactive protein, interleukin-6, lactic dehydrogenase and neutrophil counts, and higher lymphocyte and monocyte counts from Day 3 post-DXP-604 treatment compared with SOC treatment. Besides, thoracic CT images showed improvements in lesion areas and degrees, along with changes in blood inflammatory factors. Moreover, DXP-604 reduced the invasive mechanical ventilation and mortality of high-risk SARS-CoV-2 infected patients. The ongoing clinical trials of DXP-604 neutralizing antibody will clarify its utility as a new attractive countermeasure for high-risk COVID-19.

Baseline serum vitamin A and vitamin C levels and their association with disease severity in COVID-19 patients.

We aimed to investigate the association between the serum concentrations of Vitamin A and Vitamin C and the severity of the COVID-19. Methods:Fifty-three consecutive PCR (+) COVID-19 patients admitted to a dedicated ward were enrolled in this study. Blood samples for serum Vitamin A and C measurements were drawn from all participants upon admission. All subjects underwent thoracic CT imaging prior to hospitalization. CT severity score (CT-SS) was then calculated for determining the extent of pulmonary involvement. A group of healthy volunteers, in whom COVID-19 was ruled out, were assigned to the control group (n=26). These groups were compared by demographic features and serum vitamin A and C levels. The relationship between serum concentrations of these vitamins and pre-defined outcome measures, CT-SS and length of hospitalization (LOH), was also assessed. Results:In COVID-19 patients, serum Vitamin A (ng/ml, 494±96 vs. 698±93; p<0.001) and Vitamin C (ng/ml, 2961 [1991-31718] vs. 3953 [1385-8779]; p=0.007) levels were significantly lower with respect to healthy controls. According to the results of correlation analyses, there was a significant negative association between Vitamin A level and outcome measures (LOH, r=-0.293; p=0.009 and CT-SS, r=-0.289; p=0.010). The negative correlations between Vitamin C level and those measures were even more prominent (LOH, r=-0.478; p<0.001 and CT-SS, r=-0.734: p<0.001). COVID-19 patients had lower baseline serum Vitamin A and Vitamin C levels as compared to healthy controls. In subjects with COVID-19, Vitamin A and Vitamin C levels were negatively correlated with CT-SS and LOH.

Segmenting lung parenchyma from CT images with gray correlation‐based clustering

AbstractLung segmentation, a prerequisite step of lung disease detection in computer‐aided diagnosis system, is a challenging task because of noises, complex structures, as well as large individual differences of lung CT scans. Here, an automatic algorithm for segmenting lungs from thoracic CT images accurately is presented. This scheme consists of three principal steps: image preprocessing, lung extracting and contour correcting. To cope with inhomogeneous intensities of CT images, a novel preprocessing approach based on empirical mode decomposition and bilateral filter is proposed, which has abilities of denoising, smoothing and edge keeping. Lung region is then extracted with a novel gray correlation‐based clustering approach. A new lung contour correction technology is finally employed to repair the concave regions caused by pulmonary nodules, vessels and so on. Experimental results show that the preprocessing approach outperforms other methods on image denoising and smoothing. Meanwhile, the lung segmentation algorithm is tested on a group of lung CT images affected with interstitial lung diseases and achieves a high segmentation accuracy. Compared with several existing lung segmentation methods, this algorithm exhibits a better performance on lung segmentation.

Postkovid stanje - plućno ili sistemsko, reverzibilno ili ireverzibilno?

During the COVID-19 pandemic, many patients were completely recovered, and about 30% of patients had prolonged symptoms, independent from the severity of their clinical picture of the COVID-19. There is no correlation between severe acute COVID and severe post-COVID syndrome. According to the NICE 2020 recommendations, the term long COVID includes symptomatic COVID-19 (period 4-12 weeks after acute disease) and post-COVID-19 syndrome. According to the WHO post-COVID-19 syndrome is a condition in persons with a history of probable or confirmed COVID infection, usually 3 months from the onset of COVID-19, with symptoms lasting at least 2 months, that cannot be explained by an alternative diagnosis. Intolerance to physical exertion, experiencing breathing difficulty and neuropsychiatric disorders are the most common and the longest lasting difficulties. Most common respiratory post-COVID manifestation is persistent dyspnea, while persistent pulmonary fibrosis was extremely rare. In almost half of patients who have had severe COVID pneumonias, there was no absolute correlation between pulmonary function disorders and thoracic CT images, and disorders kept persisting after complete radiographic regression. Dominant manifestation of these disorders is reduced lung diffusion capacity for CO (TLCOc, KCOc), while spirometry values are mostly normal. Pulmonary thromboembolism, myocarditis, anxiety-depressive disorder, forgetfulness and hair loss are some of the serious post-COVID complications. A significant number of patients need physical rehabilitation.

Easy-to-treat and difficult-to-treat radiological phenotypes in coronavirus disease 2019 pneumonia

INTRODUCTION: Radiological phenotypes are observable radiological patterns or characteristics. Robust data are available regarding the role of high-resolution computed tomography (HRCT) in coronavirus disease 2019 (COVID-19) pneumonia. We evaluated the role of radiological phenotyping in assessing severity and predicting the response to therapy, as well as its association with outcomes in COVID-19 pneumonia. METHODS: This prospective observational study included 3000 COVID-19 reverse transcription polymerase chain reaction-confirmed cases with lung involvement who underwent thoracic HRCT on hospital admission and were categorized as mild, moderate, or severe according to lung segment bilateral involvement (mild 1–7, moderate 8–15, and severe 16–25). Follow-up thoracic CT imaging was also conducted 6 months after hospital discharge. Response to treatment phenotypes was categorized as “easy to treat” or “difficult to treat” based on the response and interventions required in indoor settings, including ventilatory support. Age, gender, comorbidities, laboratory parameters, the use of bilevel-positive airway pressure/noninvasive ventilation, and outcomes (with or without lung fibrosis) were key observations. The Chi-square test was used for statistical analysis. RESULTS: Easy-to-treat and difficult-to-treat radiological response phenotypes were observed in 20% and 80% of the cases, respectively. There were significant associations between the radiological phenotypes and the duration of illness at hospital admission. The duration of illness (&lt;7 days, 7–14 days, and &gt;14 days) could predict the radiological phenotype (P &lt; 0.00001). Laboratory parameters at hospital admission (C-reactive protein, interleukin-6, ferritin, lactate dehydrogenase, and D-dimer) were significantly associated with the radiological phenotypes (P &lt; 0.00001), as were interventions required in indoor units (P &lt; 0.00001). The HRCT severity score at admission was significantly correlated with the radiological phenotype (P &lt; 0.00001). Post-COVID lung fibrosis or sequelae were also significantly associated with the radiological phenotype (P &lt; 0.00001). CONCLUSION: Easy-to-treat and difficult-to-treat phenotypic differentiation had a crucial role during the initial assessment of COVID-19 cases on hospitalization and was used for planning targeted intervention treatments in intensive care units. In addition, phenotypic differentiation had an important role in analyzing the radiological sequelae and predicting final treatment outcomes.

Quantitative analysis of muscle volumes in COVID-19 pneumonia with an automated segmentation system

Aim: The aim of this study was to quantitatively analyze volume of the erector spinae muscle in COVID-19 pneumonia using an artificial intelligence-based automated segmentation program, and to investigate the relationship between pulmonary infiltration ratio and volume of the erector spinae muscle.Methods: In this retrospective study, thoracic CT images of patients who tested positive for SARS-CoV-2 on RT-PCR and had COVID-19 pneumonia were analyzed. Based on the percentage of pulmonary involvement, the study cohort was divided into two groups (Group I: less than 25% involvement and Group II: more than %25 involvement). Volume of the erector spinae muscle and severity of lung involvement were quantitatively analyzed using an artificial intelligence-based automated segmentation program. The data of group I and group II were compared.Results: The study population consisted of 74 subjects; 35 in Group I and 39 in Group II. Significant negative correlations were observed between the total pulmonary infiltration ratio and the volume of the erector spinae muscle. Furthermore, the analysis demonstrated that lung density, total lung infiltration volume, serum C-reactive protein (CRP) level, serum ESR level, and total erector spinae muscle volume can serve as valuable indicators for assessing the severity of lung involvement in patients with COVID-19 pneumonia.Conclusion: Measurement of erector spinae muscle volume may be useful for assessment of pulmonary infiltration in patients with COVID-19 pneumonia

Artificial intelligence-based diagnosis of asbestosis: analysis of a database with applicants for asbestosis state aid

ObjectivesIn many countries, workers who developed asbestosis due to their occupation are eligible for government support. Based on the results of clinical examination, a team of pulmonologists determine the eligibility of patients to these programs. In this Dutch cohort study, we aim to demonstrate the potential role of an artificial intelligence (AI)-based system for automated, standardized, and cost-effective evaluation of applications for asbestosis patients.MethodsA dataset of n = 523 suspected asbestosis cases/applications from across the Netherlands was retrospectively collected. Each case/application was reviewed, and based on the criteria, a panel of three pulmonologists would determine eligibility for government support. An AI system is proposed, which uses thoracic CT images as input, and predicts the assessment of the clinical panel. Alongside imaging, we evaluated the added value of lung function parameters.ResultsThe proposed AI algorithm reached an AUC of 0.87 (p < 0.001) in the prediction of accepted versus rejected applications. Diffusion capacity (DLCO) also showed comparable predictive value (AUC = 0.85, p < 0.001), with little correlation between the two parameters (r-squared = 0.22, p < 0.001). The combination of the imaging AI score and DLCO achieved superior performance (AUC = 0.95, p < 0.001). Interobserver variability between pulmonologists on the panel was estimated at alpha = 0.65 (Krippendorff’s alpha).ConclusionWe developed an AI system to support the clinical decision-making process for the application to the government support for asbestosis. A multicenter prospective validation study is currently ongoing to examine the added value and reliability of this system alongside the clinic panel.Key Points• Artificial intelligence can detect imaging patterns of asbestosis in CT scans in a cohort of patients applying for state aid.• Combining the AI prediction with the diffusing lung function parameter reaches the highest diagnostic performance.• Specific cases with fibrosis but no asbestosis were correctly classified, suggesting robustness of the AI system, which is currently under prospective validation.

Differentiation of malignant from benign pleural effusions based on artificial intelligence

IntroductionThis study aimed to construct artificial intelligence models based on thoracic CT images to perform segmentation and classification of benign pleural effusion (BPE) and malignant pleural effusion (MPE).MethodsA total of...

Hierarchical anatomical structure-aware based thoracic CT images registration

Accurate thoracic CT image registration remains challenging due to complex joint deformations and different motion patterns in multiple organs/tissues during breathing. To combat this, we devise a hierarchical anatomical structure-aware based registration framework. It affords a coordination scheme necessary for constraining a general free-form deformation (FFD) during thoracic CT registration. The key is to integrate the deformations of different anatomical structures in a divide-and-conquer way. Specifically, a deformation ability-aware dissimilarity metric is proposed for complex joint deformations containing large-scale flexible deformation of the lung region, rigid displacement of the bone region, and small-scale flexible deformation of the rest region. Furthermore, a motion pattern-aware regularization is devised to handle different motion patterns, which contain sliding motion along the lung surface, almost no displacement of the spine and smooth deformation of other regions. Moreover, to accommodate large-scale deformation, a novel hierarchical strategy, wherein different anatomical structures are fused on the same control lattice, registers images from coarse to fine via elaborate Gaussian pyramids. Extensive experiments and comprehensive evaluations have been executed on the 4D-CT DIR and 3D DIR COPD datasets. It confirms that this newly proposed method is locally comparable to state-of-the-art registration methods specializing in local deformations, while guaranteeing overall accuracy. Additionally, in contrast to the current popular learning-based methods that typically require dozens of hours or more pre-training with powerful graphics cards, our method only takes an average of 63 s to register a case with an ordinary graphics card of RTX2080 SUPER, making our method still worth promoting. Our code is available at https://github.com/heluxixue/Structure_Aware_Registration/tree/master.

Automatic segmentation of thoracic CT images using three deep learning models

PurposeDeep learning (DL) techniques are widely used in medical imaging and in particular for segmentation. Indeed, manual segmentation of organs at risk (OARs) is time-consuming and suffers from inter- and intra-observer segmentation variability. Image segmentation using DL has given very promising results. In this work, we present and compare the results of segmentation of OARs and a clinical target volume (CTV) in thoracic CT images using three DL models. Materials and methodsWe used CT images of 52 patients with breast cancer from a public dataset. Automatic segmentation of the lungs, the heart and a CTV was performed using three models based on the U-Net architecture. Three metrics were used to quantify and compare the segmentation results obtained with these models: the Dice similarity coefficient (DSC), the Jaccard coefficient (J) and the Hausdorff distance (HD). ResultsThe obtained values of DSC, J and HD were presented for each segmented organ and for the three models. Examples of automatic segmentation were presented and compared to the corresponding ground truth delineations. Our values were also compared to recent results obtained by other authors. ConclusionThe performance of three DL models was evaluated for the delineation of the lungs, the heart and a CTV. This study showed clearly that these 2D models based on the U-Net architecture can be used to delineate organs in CT images with a good performance compared to other models. Generally, the three models present similar performances. Using a dataset with more CT images, the three models should give better results.