Clinical assessment of computed tomography for detecting ingested blister packs: A single-center retrospective study.

Objectives: Blister pack (BP) ingestion poses serious risks, such as gastrointestinal perforation, and accurate localization by computed tomography (CT) is a common practice. However, while it has been reported in vitro that CT visibility varies with the material type of BPs, there have been no reports on this variability in clinical settings. In this study, we investigated the CT detection rates of different BPs in clinical settings. Methods: This single-center retrospective study from 2010 to 2022 included patients who underwent endoscopic foreign body removal for BP ingestion. The patients were categorized into two groups for BP components, the polypropylene (PP) and the polyvinyl chloride (PVC)/polyvinylidene chloride (PVDC) groups. The primary outcome was the comparison of CT detection rates between the groups. We also evaluated whether the BPs contained tablets and analyzed their locations. Results: This study included 61 patients (15 in the PP group and 46 in the PVC/PVDC group). Detection rates were 97.8% for the PVC/PVDC group compared to 53.3% for the PP group, a significant difference (p < 0.01). No cases of BPs composed solely of PP were detected by CT. Blister packs were most commonly found in the upper thoracic esophagus. Conclusions: Even in a clinical setting, the detection rates of PVC and PVDC were higher than that of PP alone. Identifying PP without tablets has proven challenging in clinical. Considering the risk of perforation, these findings suggest that esophagogastroduodenoscopy may be necessary, even if CT detection is negative.

COVID-19 Computed tomography patterns in renal replacement therapy patients / Padrões de tomografia computadorizada da COVID-19 em pacientes em terapia renal substitutiva

ABSTRACT Introduction: Lung diseases are common in patients with end stage kidney disease (ESKD), making differential diagnosis with COVID-19 a challenge. This study describes pulmonary chest tomography (CT) findings in hospitalized ESKD patients on renal replacement therapy (RRT) with clinical suspicion of COVID-19. Methods: ESKD individuals referred to emergency department older than 18 years with clinical suspicion of COVID-19 were recruited. Epidemiological baseline clinical information was extracted from electronic health records. Pulmonary CT was classified as typical, indeterminate, atypical or negative. We then compared the CT findings of positive and negative COVID-19 patients. Results: We recruited 109 patients (62.3% COVID-19-positive) between March and December 2020, mean age 60 ± 12.5 years, 43% female. The most common etiology of ESKD was diabetes. Median time on dialysis was 36 months, interquartile range = 12-84. The most common pulmonary lesion on CT was ground glass opacities. Typical CT pattern was more common in COVID-19 patients (40 (61%) vs 0 (0%) in non-COVID-19 patients, p < 0.001). Sensitivity was 60.61% (40/66) and specificity was 100% (40/40). Positive predictive value and negative predictive value were 100% and 62.3%, respectively. Atypical CT pattern was more frequent in COVID-19-negative patients (9 (14%) vs 24 (56%) in COVID-19-positive, p < 0.001), while the indeterminate pattern was similar in both groups (13 (20%) vs 6 (14%), p = 0.606), and negative pattern was more common in COVID-19-negative patients (4 (6%) vs 12 (28%), p = 0.002). Conclusions: In hospitalized ESKD patients on RRT, atypical chest CT pattern cannot adequately rule out the diagnosis of COVID-19.

RESUMO Introdução: Doenças pulmonares são comuns em pacientes com doença renal em estágio terminal (DRET), dificultando o diagnóstico diferencial com COVID-19. Este estudo descreve achados de tomografia computadorizada de tórax (TC) em pacientes com DRET em terapia renal substitutiva (TRS) hospitalizados com suspeita de COVID-19. Métodos: Indivíduos maiores de 18 anos com DRET, encaminhados ao pronto-socorro com suspeita de COVID-19 foram incluídos. Dados clínicos e epidemiológicos foram extraídos de registros eletrônicos de saúde. A TC foi classificada como típica, indeterminada, atípica, negativa. Comparamos achados tomográficos de pacientes com COVID-19 positivos e negativos. Resultados: Recrutamos 109 pacientes (62,3% COVID-19-positivos) entre março e dezembro de 2020, idade média de 60 ± 12,5 anos, 43% mulheres. A etiologia mais comum da DRET foi diabetes. Tempo médio em diálise foi 36 meses, intervalo interquartil = 12-84. A lesão pulmonar mais comum foi opacidades em vidro fosco. O padrão típico de TC foi mais comum em pacientes com COVID-19 (40 (61%) vs. 0 (0%) em pacientes sem COVID-19, p < 0,001). Sensibilidade 60,61% (40/66), especificidade 100% (40/40). Valores preditivos positivos e negativos foram 100% e 62,3%, respectivamente. Padrão atípico de TC foi mais frequente em pacientes COVID-19-negativos (9 (14%) vs. 24 (56%) em COVID-19-positivos, p < 0,001), enquanto padrão indeterminado foi semelhante em ambos os grupos (13 (20%) vs. 6 (14%), p = 0,606), e padrão negativo foi mais comum em pacientes COVID-19-negativos (4 (6%) vs. 12 (28%), p = 0,002). Conclusões: Em pacientes com DRET em TRS hospitalizados, um padrão atípico de TC de tórax não pode excluir adequadamente o diagnóstico de COVID-19.

Risk scores and coronary artery disease in patients with suspected acute coronary syndrome and intermediate cardiac troponin concentrations.

BACKGROUND: Guidelines recommend the use of risk scores to select patients for further investigation after myocardial infarction has been ruled out but their utility to identify those with coronary artery disease is uncertain. METHODS: In a prospective cohort study, patients with intermediate high-sensitivity cardiac troponin I concentrations (5 ng/L to sex-specific 99th percentile) in whom myocardial infarction was ruled out were enrolled and underwent coronary CT angiography (CCTA) after hospital discharge. History, ECG, Age, Risk factors, Troponin (HEART), Emergency Department Assessment of Chest Pain Score (EDACS), Global Registry of Acute Coronary Event (GRACE), Thrombolysis In Myocardial Infarction (TIMI), Systematic COronary Risk Evaluation 2 and Pooled Cohort Equation risk scores were calculated and the odds ratio (OR) and diagnostic performance for obstructive coronary artery disease were determined using established thresholds. RESULTS: Of 167 patients enrolled (64±12 years, 28% female), 29.9% (50/167) had obstructive coronary artery disease. The odds of having obstructive disease were increased for all scores with the lowest and highest increase observed for an EDACS score ≥16 (OR 2.2 (1.1-4.6)) and a TIMI risk score ≥1 (OR 12.9 (3.0-56.0)), respectively. The positive predictive value (PPV) was low for all scores but was highest for a GRACE score >88 identifying 39% as high risk with a PPV of 41.9% (30.4-54.2%). The negative predictive value (NPV) varied from 77.3% to 95.2% but was highest for a TIMI score of 0 identifying 26% as low risk with an NPV of 95.2% (87.2-100%). CONCLUSIONS: In patients with intermediate cardiac troponin concentrations in whom myocardial infarction has been excluded, clinical risk scores can help identify patients with and without coronary artery disease, although the performance of established risk thresholds is suboptimal for utilisation in clinical practice. TRIAL REGISTRATION NUMBER: NCT04549805.

Improving cardiovascular risk prediction with machine learning: a focus on perivascular adipose tissue characteristics.

BACKGROUND: Timely prevention of major adverse cardiovascular events (MACEs) is imperative for reducing cardiovascular diseases-related mortality. Perivascular adipose tissue (PVAT), the adipose tissue surrounding coronary arteries, has attracted increased amounts of attention. Developing a model for predicting the incidence of MACE utilizing machine learning (ML) integrating clinical and PVAT features may facilitate targeted preventive interventions and improve patient outcomes. METHODS: From January 2017 to December 2019, we analyzed a cohort of 1077 individuals who underwent coronary CT scanning at our facility. Clinical features were collected alongside imaging features, such as coronary artery calcium (CAC) scores and perivascular adipose tissue (PVAT) characteristics. Logistic regression (LR), Framingham Risk Score, and ML algorithms were employed for MACE prediction. RESULTS: We screened seven critical features to improve the practicability of the model. MACE patients tended to be older, smokers, and hypertensive. Imaging biomarkers such as CAC scores and PVAT characteristics differed significantly between patients with and without a 3-year MACE risk in a population that did not exhibit disparities in laboratory results. The ensemble model, which leverages multiple ML algorithms, demonstrated superior predictive performance compared with the other models. Finally, the ensemble model was used for risk stratification prediction to explore its clinical application value. CONCLUSIONS: The developed ensemble model effectively predicted MACE incidence based on clinical and imaging features, highlighting the potential of ML algorithms in cardiovascular risk prediction and personalized medicine. Early identification of high-risk patients may facilitate targeted preventive interventions and improve patient outcomes.

Texture-based speciation of otitis media-related bacterial biofilms from optical coherence tomography images using supervised classification.

Otitis media (OM), a highly prevalent inflammatory middle-ear disease in children worldwide, is commonly caused by an infection, and can lead to antibiotic-resistant bacterial biofilms in recurrent/chronic OM cases. A biofilm related to OM typically contains one or multiple bacterial species. OCT has been used clinically to visualize the presence of bacterial biofilms in the middle ear. This study used OCT to compare microstructural image texture features from bacterial biofilms. The proposed method applied supervised machine-learning-based frameworks (SVM, random forest, and XGBoost) to classify multiple species bacterial biofilms from in vitro cultures and clinically-obtained in vivo images from human subjects. Our findings show that optimized SVM-RBF and XGBoost classifiers achieved more than 95% of AUC, detecting each biofilm class. These results demonstrate the potential for differentiating OM-causing bacterial biofilms through texture analysis of OCT images and a machine-learning framework, offering valuable insights for real-time in vivo characterization of ear infections.

Mucociliary Clearance is Impaired in Small Airways of Cystic Fibrosis Pigs.

Cystic fibrosis is a genetic disorder characterized by recurrent airway infections, inflammation, impaired mucociliary clearance and progressive decline in lung function. The disease may start in the small airways; however, this is difficult to prove due to limited accessibility of the small airways with the current single photon mucociliary clearance assay. Here, we developed a dynamic positron emission tomography assay with high spatial and temporal resolution. We tested that mucociliary clearance is abnormal in the small airways of newborn cystic fibrosis pigs. Clearance of [68Ga] tagged macro-aggregated albumin from small airways started immediately after delivery and continued for the duration of the study. Initial clearance was fast but slowed down few minutes after delivery. Cystic fibrosis pig small airways cleared significantly less than non-CF pig small airways (non-CF 25.1±3.1% vs. CF 14.6±0.1%). Stimulation of the cystic fibrosis airways with the purinergic secretagogue UTP further impaired clearance (non-CF with UTP 20.9±0.3% vs. CF with UTP 13.0±1.8%). None of the cystic fibrosis pig treated with UTP (N = 6) cleared more than 20% of the delivered dose. These data indicate that mucociliary clearance in the small airways is fast and can easily be missed if the assay is not sensitive enough. The data also indicate that mucociliary clearance is impaired in the small airways of cystic fibrosis pigs. This defect is exacerbated by stimulation of mucus secretions with purinergic agonists.

Comparison of diagnostic value of technetium 99m-labeled red blood cell computed single photon emission computed tomography/computed tomography (99mTc-RBC SPECT/CT) and contrast-enhanced multidetector computed tomography (MDCT) for small bowel bleeding: a retrospective study.

Background: Although small bowel bleeding is relatively rare, it is a potentially fatal disease, and its diagnosis still faces challenges. Technetium 99m-labeled red blood cell computed single photon emission computed tomography/computed tomography (99mTc-RBC SPECT/CT) and contrast-enhanced multidetector computed tomography (MDCT) are common imaging methods for diagnosing small bowel bleeding, but there have been no studies comparing their diagnostic efficacy for this purpose. This study aims to compare the diagnostic value of 99mTc-RBC SPECT/CT and contrast-enhanced MDCT for small bowel bleeding. Methods: A total of 44 patients (30 males and 14 females, median age of 64 years) definitively diagnosed with small bowel bleeding and 15 non-small bowel bleeding patients (8 males and 7 females, median age of 66 years) were consecutively included in this study. All patients underwent 99mTc-RBC SPECT/CT and contrast-enhanced MDCT examinations at Beijing Friendship Hospital of Capital Medical University between January 2020 to September 2023. The definitive diagnosis had been made through surgery or colonoscopy, or through patient history, patient management, and clinical follow-up. We collected clinical data of the participants. 99mTc-RBC SPECT/CT and contrast-enhanced MDCT were reviewed in a blinded fashion for accuracy of detection of active bleeding as well as the active small bowel bleeding location. Results: Among the 59 patients, the accuracy, sensitivity, and specificity of 99mTc-RBC SPECT were 27.3%, 93.3%, and 92.3%; for 99mTc-RBC SPECT/CT they were 76.3%, 40.5%, and 93.3%; whereas for contrast-enhanced MDCT they were 45.8%, 27.3%, and 100%, respectively. The diagnostic accuracy of 99mTc-RBC SPECT/CT for jejunal and ileal bleeding was high, at 100% and 86.4%, respectively. Meanwhile, 99mTc-RBC SPECT/CT had a higher accuracy in diagnosing more causes of small bowel bleeding. In 59 patients, the combination of 99mTc-RBC SPECT/CT and contrast-enhanced MDCT accurately diagnosed small bowel bleeding and provided precise localization in 50 patients, resulting in the accuracy, sensitivity, and specificity of 84.7%, 79.5%, and 100.0%, respectively. Conclusions: 99mTc-RBC SPECT/CT has high diagnostic value in diagnosing small bowel bleeding and is superior to 99mTc-RBC SPECT and contrast-enhanced MDCT. The combination of 99mTc-RBC SPECT/CT and contrast-enhanced MDCT can further improve the diagnostic accuracy of diagnosis, and can accurately guide the diagnosis and treatment of small bowel bleeding.

Pulmonary artery aneurysm: computed tomography (CT) imaging findings and diagnosis.

Pulmonary artery aneurysm (PAA) is a rare pulmonary vascular disease with nonspecific symptoms and various etiologies. As the disease progresses, in addition to the dilation of the pulmonary arteries, it may be accompanied by remodeling of the cardiac structure and changes in the morphology of the aorta. Recognizing the cause of PAA is therefore a clinically challenging task. In this review article, we provide an overview of various causes of PAA with the support of corresponding imaging findings on computed tomography pulmonary angiography (CTPA) examination. Firstly, from the perspective of hemodynamics, a logical diagnosis is provided according to whether the main pulmonary artery (MPA) is dilated, and whether the PA is dilated locally or diffusely. Secondly, for the imaging examination of vascular wall lesions, due to the limitations of ultrasound examination and interventional procedures, the irreplaceability of dual-phase CTPA examination in disease assessment is especially emphasized. Finally, for highly suspected disorders, it is necessary to comprehensively check with the patient whether there is a family history or past medical history. For patients with PAA, especially those with Marfan syndrome (MFS) or arteritis, adequate preoperative imaging evaluation, regular postoperative radiographic follow-up, and concurrent treatment of the underlying disease (if necessary) are crucial, which are related to the prognosis and long-term quality of life of such patients. Despite the nonspecific features of PAA presentation, a thorough examination of the patient's clinical history and imaging characteristics will play an important role in diagnosing PAA and planning patient management strategies.

Prediction of epidermal growth factor receptor (EGFR) mutation status in lung adenocarcinoma patients on computed tomography (CT) images using 3-dimensional (3D) convolutional neural network.

Background: Noninvasively detecting epidermal growth factor receptor (EGFR) mutation status in lung adenocarcinoma patients before targeted therapy remains a challenge. This study aimed to develop a 3-dimensional (3D) convolutional neural network (CNN)-based deep learning model to predict EGFR mutation status using computed tomography (CT) images. Methods: We retrospectively collected 660 patients from 2 large medical centers. The patients were divided into training (n=528) and external test (n=132) sets according to hospital source. The CNN model was trained in a supervised end-to-end manner, and its performance was evaluated using an external test set. To compare the performance of the CNN model, we constructed 1 clinical and 3 radiomics models. Furthermore, we constructed a comprehensive model combining the highest-performing radiomics and CNN models. The receiver operating characteristic (ROC) curves were used as primary measures of performance for each model. Delong test was used to compare performance differences between different models. Results: Compared with the clinical [training set, area under the curve (AUC) =69.6%, 95% confidence interval (CI), 0.661-0.732; test set, AUC =68.4%, 95% CI, 0.609-0.752] and the highest-performing radiomics models (training set, AUC =84.3%, 95% CI, 0.812-0.873; test set, AUC =72.4%, 95% CI, 0.653-0.794) models, the CNN model (training set, AUC =94.3%, 95% CI, 0.920-0.961; test set, AUC =94.7%, 95% CI, 0.894-0.978) had significantly better predictive performance for predicting EGFR mutation status. In addition, compared with the comprehensive model (training set, AUC =95.7%, 95% CI, 0.942-0.971; test set, AUC =87.4%, 95% CI, 0.820-0.924), the CNN model had better stability. Conclusions: The CNN model has excellent performance in non-invasively predicting EGFR mutation status in patients with lung adenocarcinoma and is expected to become an auxiliary tool for clinicians.

Spatial position changes in the semicircular canals may be the anatomical basis of Meniere's disease: a preliminary study based on ultra-high-resolution computed tomography (CT) and intelligent segmentation.

Background: Meniere's disease (MD) is an ear-related vestibular disorder accompanied by vertigo, hearing loss, and tinnitus. The anatomical structure and spatial position of the semicircular canals are important for understanding vestibular function and disease; however, research on MD and the effect of anatomical changes in the semicircular canals is limited. This study explored the relationship between the spatial location of the semicircular canals and MD using ultra-high-resolution computed tomography (U-HRCT) and intelligent segmentation. Methods: Isotropic U-HRCT images obtained from patients with MD and healthy controls (HCs) were retrospectively analyzed. We extracted the semicircular canal structures and extracted their skeleton. The plane of the skeleton of each semicircular canal was fitted separately. The mutual angles between the semicircular canals, and the angles between each semicircular canal and each plane of the coordinate system were measured. Results: Among 45 MD-affected ears (MDAEs), 33 MD-healthy ears (MDHEs), and 45 HC ears, the angle between the superior and lateral semicircular canals (LSCs) and the angle between the superior and posterior semicircular canals (PSCs) were larger in the MDAE and MDHE groups than the HC group (P<0.01), while the angle between the posterior and LSCs was smaller in the MDAE group than the HC group (P<0.001). The angles between the superior and PSCs and coronal plane (CP) of the coordinate system were significantly smaller in the MDAE and MDHE groups than the HC group (P<0.01); however, the angles between the LSC and axial plane and CP were significantly larger in the MDAE and MDHE groups than the HC group (P<0.001). Conclusions: Spatial position changes in the semicircular canals may be the anatomical basis of MD.

Assessment of body compositions in the prediction of metabolic syndrome in adults.

Background: The musculoskeletal system participates in the pathology of metabolic disorders. Several studies have focused on body composition changes; however, the adipose tissue between muscle bundles with different metabolic statuses has rarely been studied. This study sought to explore the association between body compositions and metabolic disorders in Asians, and identify whether these body compositions can be used to detect metabolic disorders with different waist circumferences (WCs) by computed tomography (CT). Methods: A total of 116 subjects were included in the study and categorized into the following four groups according to WC and metabolic syndrome (MS): (I) the healthy control group; (II) the normal WC with metabolic disorder group; (III) the normal WC with MS group; and (IV) the larger WC with MS group. The International Diabetes Federation (IDF) criteria based on WC, laboratory tests, body mass index (BMI), and medical history was used to diagnose MS. Body composition parameters, such as muscle attenuation, the cross-sectional area of subcutaneous adipose tissue (SAT), muscle, extramyocellular lipid (EMCL), visceral adipose tissue (VAT), and the ratios between different compositions [e.g., the SMR (SAT/muscle), EMR (EMCL/muscle), and VMR (VAT/muscle)] were calculated for the thigh and abdomen. The areas under the curve (AUCs) of the receiver operating characteristic (ROC) curves adjusted for multiple comparisons were used to discriminate among metabolic disorders. Results: The groups with metabolic disorders had more SAT (P=0.001) and EMCL (P=0.040) in the thigh, and more VAT (P=0.001) and a higher SMR (P<0.001) in the abdomen. EMCL and muscle attenuation in the thigh (AUCs =0.790 and 0.791), and the VMR and SMR in the abdomen were better able to diagnose metabolic disorders (AUCs =0.752 and 0.746) than other body composition parameters. While SAT and EMCL in the thigh (AUCs =0.768 and 0.760), and VAT and the VMR in the abdomen (AUCs =0.788 and 0.775) were better able to diagnose MS than other parameters. Conclusions: Body composition parameters for the thigh and abdomen could assist in detecting patients with an increased risk of MS.

Fully automated classification of pulmonary nodules in positron emission tomography-computed tomography imaging using a two-stage multimodal learning approach.

Background: Lung cancer is a malignant tumor, for which pulmonary nodules are considered to be significant indicators. Early recognition and timely treatment of pulmonary nodules can contribute to improving the survival rate of patients with cancer. Positron emission tomography-computed tomography (PET/CT) is a noninvasive, fusion imaging technique that can obtain both functional and structural information of lung regions. However, studies of pulmonary nodules based on computer-aided diagnosis have primarily focused on the nodule level due to a reliance on the annotation of nodules, which is superficial and unable to contribute to the actual clinical diagnosis. The aim of this study was thus to develop a fully automated classification framework for a more comprehensive assessment of pulmonary nodules in PET/CT imaging data. Methods: We developed a two-stage multimodal learning framework for the diagnosis of pulmonary nodules in PET/CT imaging. In this framework, Stage I focuses on pulmonary parenchyma segmentation using a pretrained U-Net and PET/CT registration. Stage II aims to extract, integrate, and recognize image-level and feature-level features by employing the three-dimensional (3D) Inception-residual net (ResNet) convolutional block attention module architecture and a dense-voting fusion mechanism. Results: In the experiments, the proposed model's performance was comprehensively validated using a set of real clinical data, achieving mean scores of 89.98%, 89.21%, 84.75%, 93.38%, 86.83%, and 0.9227 for accuracy, precision, recall, specificity, F1 score, and area under curve values, respectively. Conclusions: This paper presents a two-stage multimodal learning approach for the automatic diagnosis of pulmonary nodules. The findings reveal that the main reason for limiting model performance is the nonsolitary property of nodules in pulmonary nodule diagnosis, providing direction for future research.

Elevated modiolus density on high-resolution computed tomography as an indicator of cochlear neurodysplasia: a retrospective case-control study.

Background: Cochlear neurodysplasia (CND) is recognized as a contributing factor to sensorineural hearing loss in children. This study aimed to investigate the relationship between modiolus density on high-resolution computed tomography (HRCT) and CND, and to evaluate its performance in diagnosing CND. Methods: This retrospective study collected HRCT images of 34 patients diagnosed with unilateral neurological hearing loss in the Children's Hospital of Chongqing Medical University from March 2018 to December 2023, who were also diagnosed with unilateral CND by computed tomography (CT) and magnetic resonance imaging (MRI) hydroimaging. CT values of the modiolus and petrous bone were measured on the affected and healthy sides, in addition to determining the width of cochlear nerve foramina and the width of internal auditory tract. The receiver operator characteristic (ROC) curve was used to evaluate the diagnostic performance of these features. Simultaneously, comparisons were conducted with parameters obtained from normal children. A total of 29 patients without CND were randomly selected as a control group. Results: The unilateral sensorineural hearing loss group had 34 patients, comprising 18 males and 16 females, with a median age of 4.5 years, ranging from 0.7 to 11 years. The normal children group consisted of 20 males and 9 females, with a median age of 5.9 years, ranging from 0.5 to 12.0 years. Statistically significant differences were observed in the CT values of the modiolus, modiolus/petrous bone CT value ratio, width of cochlear nerve foramina, and width of internal auditory tract between the affected and healthy sides in patients with unilateral sensorineural hearing loss (P<0.05). The area under the ROC curve (AUC) of the modiolus CT value and the width of cochlear nerve foramina for the diagnosis of unilateral sensorineural hearing loss was 0.98 [95% confidence interval (CI): 0.95-1.00] and 0.99 (95% CI: 0.98-1.00), respectively. the modiolus density was significantly elevated in the affected sides in patients with unilateral CND. The optimal cut-off value of modiolus CT values was 983 Hounsfield unit (HU). Conclusions: The elevated density of the modiolus on HRCT holds significant value in diagnosing CND.

Fusion of shallow and deep features from 18F-FDG PET/CT for predicting EGFR-sensitizing mutations in non-small cell lung cancer.

Background: Non-small cell lung cancer (NSCLC) patients with epidermal growth factor receptor-sensitizing (EGFR-sensitizing) mutations exhibit a positive response to tyrosine kinase inhibitors (TKIs). Given the limitations of current clinical predictive methods, it is critical to explore radiomics-based approaches. In this study, we leveraged deep-learning technology with multimodal radiomics data to more accurately predict EGFR-sensitizing mutations. Methods: A total of 202 patients who underwent both flourine-18 fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) scans and EGFR sequencing prior to treatment were included in this study. Deep and shallow features were extracted by a residual neural network and the Python package PyRadiomics, respectively. We used least absolute shrinkage and selection operator (LASSO) regression to select predictive features and applied a support vector machine (SVM) to classify the EGFR-sensitive patients. Moreover, we compared predictive performance across different deep models and imaging modalities. Results: In the classification of EGFR-sensitive mutations, the areas under the curve (AUCs) of ResNet-based deep-shallow features and only shallow features from different multidata were as follows: RES_TRAD, PET/CT vs. CT-only vs. PET-only: 0.94 vs. 0.89 vs. 0.92; and ONLY_TRAD, PET/CT vs. CT-only vs. PET-only: 0.68 vs. 0.50 vs. 0.38. Additionally, the receiver operating characteristic (ROC) curves of the model using both deep and shallow features were significantly different from those of the model built using only shallow features (P<0.05). Conclusions: Our findings suggest that deep features significantly enhance the detection of EGFR-sensitizing mutations, especially those extracted with ResNet. Moreover, PET/CT images are more effective than CT-only and PET-only images in producing EGFR-sensitizing mutation-related signatures.

Weakly supervised low-dose computed tomography denoising based on generative adversarial networks.

Background: Low-dose computed tomography (LDCT) is a diagnostic imaging technique designed to minimize radiation exposure to the patient. However, this reduction in radiation may compromise computed tomography (CT) image quality, adversely impacting clinical diagnoses. Various advanced LDCT methods have emerged to mitigate this challenge, relying on well-matched LDCT and normal-dose CT (NDCT) image pairs for training. Nevertheless, these methods often face difficulties in distinguishing image details from nonuniformly distributed noise, limiting their denoising efficacy. Additionally, acquiring suitably paired datasets in the medical domain poses challenges, further constraining their applicability. Hence, the objective of this study was to develop an innovative denoising framework for LDCT images employing unpaired data. Methods: In this paper, we propose a LDCT denoising network (DNCNN) that alleviates the need for aligning LDCT and NDCT images. Our approach employs generative adversarial networks (GANs) to learn and model the noise present in LDCT images, establishing a mapping from the pseudo-LDCT to the actual NDCT domain without the need for paired CT images. Results: Within the domain of weakly supervised methods, our proposed model exhibited superior objective metrics on the simulated dataset when compared to CycleGAN and selective kernel-based cycle-consistent GAN (SKFCycleGAN): the peak signal-to-noise ratio (PSNR) was 43.9441, the structural similarity index measure (SSIM) was 0.9660, and the visual information fidelity (VIF) was 0.7707. In the clinical dataset, we conducted a visual effect analysis by observing various tissues through different observation windows. Our proposed method achieved a no-reference structural sharpness (NRSS) value of 0.6171, which was closest to that of the NDCT images (NRSS =0.6049), demonstrating its superiority over other denoising techniques in preserving details, maintaining structural integrity, and enhancing edge contrast. Conclusions: Through extensive experiments on both simulated and clinical datasets, we demonstrated the superior efficacy of our proposed method in terms of denoising quality and quantity. Our method exhibits superiority over both supervised techniques, including block-matching and 3D filtering (BM3D), residual encoder-decoder convolutional neural network (RED-CNN), and Wasserstein generative adversarial network-VGG (WGAN-VGG), and over weakly supervised approaches, including CycleGAN and SKFCycleGAN.

Integrating myocardial metabolic imaging and stress myocardial contrast echocardiography to improve the diagnosis of coronary microvascular diseases in rabbits.

Background: Persistent challenges associated with misdiagnosis and underdiagnosis of coronary microvascular disease (CMVD) necessitate the exploration of noninvasive imaging techniques to enhance diagnostic accuracy. Therefore, we aimed to integrate multimodal imaging approaches to achieve a higher diagnostic rate for CMVD using high-quality myocardial metabolism imaging (MMI) and myocardial contrast echocardiography (MCE). This combination diagnostic strategy may help address the urgent need for improved CMVD diagnosis. Methods: In this study, we established five distinct pretreatment groups, each consisting of nine male rabbit: a fasted group, a nonfasted group, a sugar load group, an acipimox group, and a combination group of nonfasted rabbits administered insulin. Moreover, positron emission tomography-computed tomography (PET/CT) scan windows were established at 30-, 60-, and 90-minute intervals. We developed 10 CMVD models and conducted a diagnosis of CMVD through an integrated analysis of MMI and MCE, including image acquisition and processing. For each heart segment, we calculated the standardized uptake value (SUV) based on body weight (SUVbw), as well as certain ratios of SUV including SUV of the heart (SUVheart) to that of the liver (SUVliver) and SUVheart to SUV of the lung (SUVlung). Additionally, we obtained three coronary SUVbw uptake values. To clarify the relationship between SUVbw uptake values and echocardiographic parameters of the myocardial contrast agent more thoroughly, we conducted a comprehensive analysis across different pretreatment protocols. Receiver operating characteristic (ROC) curve analysis was employed to evaluate the diagnostic accuracy of each parameter in the context of CMVD. Results: In the context of MMI, the nonfasted-plus-insulin group, as observed during the 60-minute examination, exhibited a noteworthy total 18F-fluorodeoxyglucose (18F-FDG) uptake of 47.44±6.53 g/mL, which was found to be statistically different from the other groups. To ascertain the reliability of the results, two double-blind investigators independently assessed the data and achieved a good level of agreement, according to the intraclass correlation coefficient (ICC) (0.957). The SUVbw of the nonfasted-plus-insulin group exhibited a moderate correlation with the microvascular blood flow reserve (MBFR) parameters derived from the MCE examination, as evidenced by a r value of 0.686. For the diagnosis of CMVD disease, the diagnostic accuracy of the combined diagnostic method [area under the curve (AUC) =0.789; 95% confidence interval (CI): 0.705-0.873] was significantly higher than that of the MBFR (AUC =0.697; 95% CI: 0.597-0.797) and SUVbw (AUC =0.715; 95% CI: 0.622-0.807) methods (P<0.05). Conclusions: Our study demonstrated the feasibility of a simple premedication approach involving free feeding and intravenous insulin in producing high-quality gated heart 18F-FDG PET/CT images in adult male New Zealand white rabbits. This technique holds considerable potential for ischemic heart disease research in rabbits and can enhance CMVD diagnosis via the comprehensive assessment of myocardial metabolism and perfusion.

Predictive value of spectral dual-detector computed tomography for PD-L1 expression in stage I lung adenocarcinoma: development and validation of a novel nomogram.

Background: Programmed death ligand-1 (PD-L1) expression serves a predictive biomarker for the efficacy of immune checkpoint inhibitors (ICIs) in the treatment of patients with early-stage lung adenocarcinoma (LA). However, only a limited number of studies have explored the relationship between PD-L1 expression and spectral dual-layer detector-based computed tomography (SDCT) quantification, qualitative parameters, and clinical biomarkers. Therefore, this study was conducted to clarify this relationship in stage I LA and to develop a nomogram to assist in preoperative individualized identification of PD-L1-positive expression. Methods: We analyzed SDCT parameters and PD-L1 expression in patients diagnosed with invasive nonmucinous LA through postoperative pathology. Patients were categorized into PD-L1-positive and PD-L1-negative expression groups based on a threshold of 1%. A retrospective set (N=356) was used to develop and internally validate the radiological and biomarker features collected from predictive models. Univariate analysis was employed to reduce dimensionality, and logistic regression was used to establish a nomogram for predicting PD-L1 expression. The predictive performance of the model was evaluated using receiver operating characteristic (ROC) curves, and external validation was performed in an independent set (N=80). Results: The proportions of solid components and pleural indentations were higher in the PD-L1-positive group, as indicated by the computed tomography (CT) value, CT at 40 keV (CT40keV; a/v), electron density (ED; a/v), and thymidine kinase 1 (TK1) exhibiting a positive correlation with PD-L1 expression. In contrast, the effective atomic number (Zeff; a/v) showed a negative correlation with PD-L1 expression [r=-0.4266 (Zeff.a), -0.1131 (Zeff.v); P<0.05]. After univariate analysis, 18 parameters were found to be associated with PD-L1 expression. Multiple regression analysis was performed on significant parameters with an area under the curve (AUC) >0.6, and CT value [AUC =0.627; odds ratio (OR) =0.993; P=0.033], CT40keV.a (AUC =0.642; OR =1.006; P=0.025), arterial Zeff (Zeff.a) (AUC =0.756; OR =0.102; P<0.001), arterial ED (ED.a) (AUC =0.641; OR =1.158, P<0.001), venous ED (ED.v) (AUC =0.607; OR =0.864; P<0.001), TK1 (AUC =0.601; OR =1.245; P=0.026), and diameter of solid components (Dsolid) (AUC =0.632; OR =1.058; P=0.04) were found to be independent risk factors for PD-L1 expression in stage I LA. These seven predictive factors were integrated into the development of an SDCT parameter-clinical nomogram, which demonstrated satisfactory discrimination ability in the training set [AUC =0.853; 95% confidence interval (CI): 0.76-0.947], internal validation set (AUC =0.824; 95% CI: 0.775-0.874), and external validation set (AUC =0.825; 95% CI: 0.733-0.918). Decision curve analyses also revealed the highest net benefit for the nomogram across a broad threshold probability range (20-80%), with a clinical impact curve (CIC) indicating its clinical validity. Comparisons with other models demonstrated the superior discriminatory accuracy of the nomogram over any individual variable (all P values <0.05). Conclusions: Quantitative parameters derived from SDCT demonstrated the ability to predict for PD-L1 expression in early-stage LA, with Zeff.a being notably effective. The nomogram established in combination with TK1 showed excellent predictive performance and good calibration. This approach may facilitate the improved noninvasive prediction of PD-L1 expression.

Performance assessment of an artificial intelligence-based coronary artery calcium scoring algorithm in non-gated chest CT scans of different slice thickness.

Background: The coronary artery calcium score (CACS) has been shown to be an independent predictor of cardiovascular events. The traditional coronary artery calcium scoring algorithm has been optimized for electrocardiogram (ECG)-gated images, which are acquired with specific settings and timing. Therefore, if the artificial intelligence-based coronary artery calcium score (AI-CACS) could be calculated from a chest low-dose computed tomography (LDCT) examination, it could be valuable in assessing the risk of coronary artery disease (CAD) in advance, and it could potentially reduce the occurrence of cardiovascular events in patients. This study aimed to assess the performance of an AI-CACS algorithm in non-gated chest scans with three different slice thicknesses (1, 3, and 5 mm). Methods: A total of 135 patients who underwent both LDCT of the chest and ECG-gated non-contrast enhanced cardiac CT were prospectively included in this study. The Agatston scores were automatically derived from chest CT images reconstructed at slice thicknesses of 1, 3, and 5 mm using the AI-CACS software. These scores were then compared to those obtained from the ECG-gated cardiac CT data using a conventional semi-automatic method that served as the reference. The correlations between the AI-CACS and electrocardiogram-gated coronary artery calcium score (ECG-CACS) were analyzed, and Bland-Altman plots were used to assess agreement. Risk stratification was based on the calculated CACS, and the concordance rate was determined. Results: A total of 112 patients were included in the final analysis. The correlations between the AI-CACS at three different thicknesses (1, 3, and 5 mm) and the ECG-CACS were 0.973, 0.941, and 0.834 (all P<0.01), respectively. The Bland-Altman plots showed mean differences in the AI-CACS for the three thicknesses of -6.5, 15.4, and 53.1, respectively. The risk category agreement for the three AI-CACS groups was 0.868, 0.772, and 0.412 (all P<0.01), respectively. While the concordance rates were 91%, 84.8%, and 62.5%, respectively. Conclusions: The AI-based algorithm successfully calculated the CACS from LDCT scans of the chest, demonstrating its utility in risk categorization. Furthermore, the CACS derived from images with a slice thickness of 1 mm was more accurate than those obtained from images with slice thicknesses of 3 and 5 mm.

Assessing the relationship of agger nasi pneumatization to the lacrimal sac: a dynamic computed tomography-dacryocystography analysis.

Background: An understanding of the anatomical structure is crucial for completing successful endoscopic dacryocystorhinostomy (DCR) surgery. This study aimed to precisely delineate the spatial relationship between the lacrimal sac and the agger nasi cell (ANC) and evaluate the impact of ANC on surgical strategies in endoscopic DCR. Methods: This retrospective cross-sectional study included 110 Han Chinese patients diagnosed with unilateral primary acquired nasolacrimal duct obstruction (PANDO) from January 2021 to June 2023. This study was conducted in Eye, Ear, Nose, and Throat Hospital of Fudan University and involved inpatient participants who were scheduled for DCR surgery under general anesthesia. Patients were consecutively enrolled. The patients underwent preoperative computed tomography-dacryocystography (CT-DCG), and contrast-enhanced images were used to locate the positions of the lacrimal sac and the common canaliculus. A dynamic approach was adopted to analyze the multiplanar CT imaging, facilitating a detailed assessment of the morphology of the lacrimal drainage system and potential overlap of the lacrimal sac. Patient ages and measured values are presented as the mean ± standard deviation, which were measured three times by the same observer and averaged for statistical analysis. Results: The prevalence of ANC in this study was 90.9% (100/110). Dynamic examination revealed that only 42.7% (47/110) of ANCs appeared as discrete cells, while the majority were connected to nearby sinus openings. Spatial analysis showed that in 57 out of 110 cases, ANCs were situated below the common canaliculus and not posterior to the lacrimal sac, indicating an overlap rate of 51.8%. Notably, our dynamic approach identified five critical cases of overlap below the level of the common canaliculus, which might have been missed by prior studies that used different methodologies. Conclusions: More than half of the ANCs exhibited overlap with the lacrimal sac, suggesting a significant proportion may necessitate opening during endoscopic DCR procedures. ANCs are often interconnected with adjacent nasal sinuses, necessitating careful consideration in the decision to open the ANCs during surgery. The dynamic evaluation employed in CT-DCG effectively assessed the extent of ANC coverage over the lacrimal sac.

Repeatability and reproducibility of deep learning features for lung adenocarcinoma subtypes with nodules less than 10 mm in size: a multicenter thin-slice computed tomography phantom and clinical validation study.

Background: Deep learning features (DLFs) derived from radiomics features (RFs) fused with deep learning have shown potential in enhancing diagnostic capability. However, the limited repeatability and reproducibility of DLFs across multiple centers represents a challenge in the clinically validation of these features. This study thus aimed to evaluate the repeatability and reproducibility of DLFs and their potential efficiency in differentiating subtypes of lung adenocarcinoma less than 10 mm in size and manifesting as ground-glass nodules (GGNs). Methods: A chest phantom with nodules was scanned repeatedly using different thin-slice computed tomography (TSCT) scanners with varying acquisition and reconstruction parameters. The robustness of the DLFs was measured using the concordance correlation coefficient (CCC) and intraclass correlation coefficient (ICC). A deep learning approach was used for visualizing the DLFs. To assess the clinical effectiveness and generalizability of the stable and informative DLFs, three hospitals were used to source 275 patients, in whom 405 nodules were pathologically differentially diagnosed as GGN lung adenocarcinoma less than 10 mm in size and were retrospectively reviewed for clinical validation. Results: A total of 64 DLFs were analyzed, which revealed that the variables of slice thickness and slice interval (ICC, 0.79±0.18) and reconstruction kernel (ICC, 0.82±0.07) were significantly associated with the robustness of DLFs. Feature visualization showed that the DLFs were mainly focused around the nodule areas. In the external validation, a subset of 28 robust DLFs identified as stable under all sources of variability achieved the highest area under curve [AUC =0.65, 95% confidence interval (CI): 0.53-0.76] compared to other DLF models and the radiomics model. Conclusions: Although different manufacturers and scanning schemes affect the reproducibility of DLFs, certain DLFs demonstrated excellent stability and effectively improved diagnostic the efficacy for identifying subtypes of lung adenocarcinoma. Therefore, as the first step, screening stable DLFs in multicenter DLFs research may improve diagnostic efficacy and promote the application of these features.