Journal of Thoracic Imaging最新文献

Purpose: To explore the CT features in prognostic evaluations for pulmonary mucormycosis in patients with hematological diseases.

Materials and methods: A retrospective analysis of clinical data and chest CT features of 53 HD patients with PM was conducted. Univariate and multivariate logistic regression analyses were used to determine the risk factors for death. The Cox regression model was used to analyze the factors affecting the survival rate.

Results: A total of 30 patients with proven PM and 23 with probable PM were included. All 30 patients with proven PM underwent bronchoscopy-guided biopsy, among which 9 cases underwent surgical resection. Of the 23 patients with probable PM, 5 cases had positive results in sputum smear microscopy, 4 cases in sputum culture, 13 cases in bronchoalveolar lavage fluid (BALF) microscopy, and 1 case in BALF culture. All identification of pathogen genera and partial species was conducted by metagenomic next-generation sequencing (mNGS) testing. In the multivariate regression analysis, the CT feature of multiple lesions (≥2) on the initial CT scan was an independent risk factor for mortality ( P =0.019). Cox survival analysis demonstrated a significantly lower survival rate ( P =0.043) in patients exhibiting the CT feature of multiple lesions on the initial CT scan.

Conclusions: The CT feature of multiple lesions (≥2) on the initial CT may serve as an independent risk factor for mortality in patients with hematologic disorders with pulmonary mucormycosis.

Purpose: To investigate and compare the diagnostic performance of 10 different large language models (LLMs) and 2 board-certified general radiologists in thoracic radiology cases published by The Society of Thoracic Radiology.

Materials and methods: We collected publicly available 124 "Case of the Month" from the Society of Thoracic Radiology website between March 2012 and December 2023. Medical history and imaging findings were input into LLMs for diagnosis and differential diagnosis, while radiologists independently visually provided their assessments. Cases were categorized anatomically (parenchyma, airways, mediastinum-pleura-chest wall, and vascular) and further classified as specific or nonspecific for radiologic diagnosis. Diagnostic accuracy and differential diagnosis scores (DDxScore) were analyzed using the χ 2 , Kruskal-Wallis, Wilcoxon, McNemar, and Mann-Whitney U tests.

Results: Among the 124 cases, Claude 3 Opus showed the highest diagnostic accuracy (70.29%), followed by ChatGPT 4/Google Gemini 1.5 Pro (59.75%), Meta Llama 3 70b (57.3%), ChatGPT 3.5 (53.2%), outperforming radiologists (52.4% and 41.1%) and other LLMs ( P <0.05). Claude 3 Opus DDxScore was significantly better than other LLMs and radiologists, except ChatGPT 3.5 ( P <0.05). All LLMs and radiologists showed greater accuracy in specific cases ( P <0.05), with no DDxScore difference for Perplexity and Google Bard based on specificity ( P >0.05). There were no significant differences between LLMs and radiologists in the diagnostic accuracy of anatomic subgroups ( P >0.05), except for Meta Llama 3 70b in the vascular cases ( P =0.040).

Conclusions: Claude 3 Opus outperformed other LLMs and radiologists in text-based thoracic radiology cases. LLMs hold great promise for clinical decision systems under proper medical supervision.

Purpose: Compare the impact of photon-counting detector computed tomography (PCD-CT) to conventional CT on an interstitial lung disease (ILD) quantitative machine learning (QML) model.

Materials and methods: A QML model analyzed 52 CT exams from patients who underwent same-day conventional and PCD-CT for suspected ILD. Lin's concordance correlation coefficient (CCC) assessed agreement between conventional and PCD-CT QML results. A CCC >0.90 was regarded as excellent, 0.9 to 0.8 as good, and <0.80 as a poor concordance. Spearman rank correlation evaluated the association between pulmonary function test results (PFT) and QML features (reticulation [R], honeycombing [HC], ground glass [GG], interstitial lung disease [ILD], and vessel-related structures [VRS]). Correlations were statistically significant if the 95% CI did not include 0.00 and P value <0.05.

Results: Conventional and PCD-CT QML results had good to excellent concordance (CCC ≥0.8) except for total HC (CCC <0.8), likely related to better PCD-CT honeycombing delineation. Overall, compared with conventional CT, PCD-CT had consistently more statistically significant correlation with PFT for HC (9 PCD vs. 2 conventional of 28 total and regional associations), similar correlation for R (20 PCD vs. 18 conventional of 28 associations) and VRS (19 PCD vs. 23 conventional of 28 associations), and less correlation for GG extent (12 PCD vs. 20 conventional associations).

Conclusions: There is strong agreement between conventional and PCD-CT QML ILD features except for HC. PCD-CT improved HC but decreased GG extent correlation with PFT. Therefore, even though most quantitative features were not impacted by the newer PCD-CT technology, model adjustment is necessary.

Acute pulmonary injury can occur in response to any number of inciting factors. The body's response to these insults is much less diverse and usually categorizable as one of several patterns of disease defined by histopathology, with corresponding patterns on chest CT. Common patterns of acute injury include diffuse alveolar damage, organizing pneumonia, acute eosinophilic pneumonia, and hypersensitivity pneumonitis. The ultimate clinical diagnosis is multidisciplinary, requiring a detailed history and relevant laboratory investigations from referring clinicians, identification of injury patterns on imaging by radiologists, and sometimes tissue evaluation by pathologists. In this review, several clinical diagnoses will be explored, grouped by imaging pattern, with a representative clinical presentation, a review of the current literature, and a discussion of typical imaging findings. Additional information on terminology and disambiguation will be provided to assist with comprehension and standardization of descriptions. The focus will be on the acute phase of illness from presentation to diagnosis; treatment methods and chronic sequela of acute disease are beyond the scope of this review.

Purpose: To evaluate the accuracy of ultra-low dose (ULD) chest computed tomography (CT), with a radiation exposure equivalent to a 2-view chest x-ray, for pulmonary nodule detection using deep learning image reconstruction (DLIR).

Material and methods: This prospective cross-sectional study included 60 patients referred to our institution for assessment or follow-up of solid pulmonary nodules. All patients underwent low-dose (LD) and ULD chest CT within the same examination session. LD CT data were reconstructed using Adaptive Statistical Iterative Reconstruction-V (ASIR-V), whereas ULD CT data were reconstructed using DLIR and ASIR-V. ULD CT images were reviewed by 2 readers and LD CT images were reviewed by an experienced thoracic radiologist as the reference standard. Quantitative image quality analysis was performed, and the detectability of pulmonary nodules was assessed according to their size and location.

Results: The effective radiation dose for ULD CT and LD CT were 0.13±0.01 and 1.16±0.6 mSv, respectively. Over the whole population, LD CT revealed 733 nodules. At ULD, DLIR images significantly exhibited better image quality than ASIR-V images. The overall sensitivity of DLIR reconstruction for the detection of solid pulmonary nodules from the ULD CT series was 93% and 82% for the 2 readers, with a good to excellent agreement with LD CT (ICC=0.82 and 0.66, respectively). The best sensitivities were observed in the middle lobe (97% and 85%, respectively).

Conclusions: At ULD, DLIR reconstructions, with minimal radiation exposure that could facilitate large-scale screening, allow the detection of pulmonary nodules with high sensitivity in an unrestricted BMI population.

Purpose: Bronchiectasis is associated with loss of lung function, substantial use of health care resources, and increased morbidity and mortality in people with cardiopulmonary diseases. We assessed the frequency of progression or new development of bronchiectasis and predictors of progression in participants in low-dose computed tomography (CT) screening programs.

Materials and methods: We reviewed our prospectively enrolled screening cohort in the Early Lung and Cardiac Action Program cohort of smokers, aged 40 to 90, between 2010 and 2019, and medical records to assess the progression of bronchiectasis after five or more years of follow-up after baseline low-dose CT. Logistic and multivariate-analysis-of-covariance regression analyses were used to examine factors associated with bronchiectasis progression.

Results: Among 2182 baseline screening participants, we identified 534 (mean age: 65±9 y; 53.6% women) with follow-up screening of 5+ years (median follow-up: 103.2 mo). Of the 534 participants, 34 (6.4%) participants had progressed (25/126, 19.8%) or newly developed (9/408, 2.2%) bronchiectasis. Significant predictors of progression (progressed+newly developed) were: age ( P =0.03), pack-years of smoking ( P =0.004), baseline components of the ELCAP Bronchiectasis Score, including the severity of bronchial dilatation ( P =0.01), its extent ( P =0.01), bronchial wall thickening ( P =0.04), and mucoid impaction ( P <0.001).

Conclusions: Assuming similar progression rates, ~136 out of 2182 participants are expected to progress on follow-up screening. This study sheds light on bronchiectasis progression and its significant predictors in a low-dose CT screening program. We recommend reporting bronchiectasis as participants who have smoked are at increased risk, and continued assessment over the entire period of participation in the low-dose CT screening program would allow for the identification of possible causes, early warning, and even early treatment.

Purpose: This study was designed to construct progressive binary classification models based on radiomics and deep learning to predict the presence of epidermal growth factor receptor ( EGFR ) and TP53 mutations and to assess the models' capacities to identify patients who are suitable for TKI-targeted therapy and those with poor prognoses.

Materials and methods: A total of 267 patients with lung adenocarcinomas who underwent genetic testing and noncontrast chest computed tomography from our hospital were retrospectively included. Clinical information and imaging characteristics were gathered, and high-throughput feature acquisition on all defined regions of interest (ROIs) was carried out. We selected features and constructed clinical models, radiomics models, deep learning models, and ensemble models to predict EGFR status with all patients and TP53 status with EGFR-positive patients, respectively. The validity and reliability of each model were expressed as the area under the curve (AUC), sensitivity, specificity, accuracy, precision, and F1 score.

Results: We constructed 7 kinds of models for 2 different dichotomies, namely, the clinical model, the radiomics model, the DL model, the rad-clin model, the DL-clin model, the DL-rad model, and the DL-rad-clin model. For EGFR - and EGFR +, the DL-rad-clin model got the highest AUC value of 0.783 (95% CI: 0.677-0.889), followed by the rad-clin model, the DL-clin model, and the DL-rad model. In the group with an EGFR mutation, for TP53 - and TP53 +, the rad-clin model got the highest AUC value of 0.811 (95% CI: 0.651-0.972), followed by the DL-rad-clin model and the DL-rad model.

Conclusion: Our progressive binary classification models based on radiomics and deep learning may provide a good reference and complement for the clinical identification of TKI responders and those with poor prognoses.

Drug-induced acute lung injury is a significant yet often underrecognized clinical challenge, associated with a wide range of therapeutic agents, including chemotherapy drugs, antibiotics, anti-inflammatory drugs, and immunotherapies. This comprehensive review examines the pathophysiology, clinical manifestations, and radiologic findings of drug-induced acute lung injury across different drug categories. Common imaging findings are highlighted to aid radiologists and clinicians in early recognition and diagnosis. The review emphasizes the importance of immediate cessation of the offending drug and supportive care, which may include corticosteroids. Understanding these patterns is crucial for prompt diagnosis and management, potentially improving patient outcomes.

Acute lung injury (ALI) is acute pulmonary inflammation with underlying pathology of disruption of the pulmonary vasculature endothelial and alveolar epithelial barriers. ALI is not an uncommon diagnosis and has a myriad of causes including pulmonary infection, (including sepsis), drugs, connective tissue disease, and polytrauma. Patients present clinically with hypoxemia with imaging supportive of bilateral pulmonary findings without pulmonary edema. The imaging findings in ALI mirror pathologic changes, with a transition from an early ("exudative") phase to a later fibroblast-rich ("organizing" or "proliferative") phase to, in some cases, a fibrotic phase. The diagnosis of ALI is separate from, but can clinically overlap in presentation with, acute respiratory distress syndrome and is characterized by diffuse alveolar damage and organizing pneumonia patterns on pathology. Clinical management is most often supportive and can include corticosteroids, mechanical ventilation, and careful fluid management, with the goal of preserving and recovering lung function.