Journal of Thoracic Imaging最新文献

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 χ2, 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: 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: Lesion overlooking and late diagnostic workup can compromise the efficacy of low-dose CT (LDCT) screening of lung cancer (LC), implying more advanced and less curable disease stages. We hypothesized that the azygos esophageal recess (AER) of the right lower lobe (RLL) might be an area prone to lesion overlooking in LC screening.

Materials and methods: Two radiologists reviewed the LDCT examinations of all the screen-detected incident LCs observed in the active arm of 2 randomized clinical trials: ITALUNG and national lung screening trial. Those in the AER were compared with those in the remainder of the RLL for possible differences in diagnostic lag according to the Lung-RADS 1.1 recommendations, size, stage, and mortality.

Results: Six (11.7%) of 51 screen-detected incident LCs of the RLL were located in the AER. The diagnostic lag time was significantly longer (P=0.046) in the AER LC (mean 14±9 mo) than in the LC in the remaining RLL (mean 7.3±1 mo). Size and stage at diagnosis were not significantly different. All 6 subjects with LC in the AER and 16 (35.5%) of 45 subjects with LC in the remaining RLL (P=0.004) died of LC after a median follow-up of 12 years.

Conclusion: Our retrospective study indicates that AER might represent a lung region of the RLL prone to have early LC overlooked due to detection or interpretation errors with possible detrimental consequences for the subject undergoing LC screening.

Purpose: To compare texture-based analysis using convolutional neural networks (CNNs) against lung densitometry in detecting chest computed tomography (CT) image abnormalities.

Material and methods: A U-NET was used for lung segmentation, and an ensemble of 7 CNN architectures was trained for the classification of low-attenuation areas (LAAs; emphysema, cysts), normal-attenuation areas (NAAs; normal parenchyma), and high-attenuation areas (HAAs; ground-glass opacities, crazy paving/linear opacity, consolidation). Lung densitometry also computes (LAAs, ≤-950 HU), NAAs (-949 to -700 HU), and HAAs (-699 to -250 HU). CNN-based and densitometry-based severity indices (CNN and Dens, respectively) were calculated as (LAA+HAA)/(LAA+NAA+HAA) in 812 CT scans from 176 normal subjects, 343 patients with emphysema, and 293 patients with interstitial lung disease (ILD). The correlation between CNN-derived and densitometry-derived indices was analyzed, alongside a comparison of severity indices among patient subgroups with emphysema and ILD, using the Spearman correlation and ANOVA with Bonferroni correction.

Results: CNN-derived and densitometry-derived severity indices (SIs) showed a strong correlation (ρ=0.90) and increased with disease severity. CNN-SIs differed from densitometry SIs, being lower for emphysema and higher for moderate to severe ILD cases. CNN estimations for normal attenuation areas were higher than those from densitometry across all groups, indicating a potential for more accurate characterization of lung abnormalities.

Conclusions: CNN outputs align closely with densitometry in assessing lung abnormalities on CT scans, offering improved estimates of normal areas and better distinguishing similar abnormalities. However, this requires higher computing power.

Purpose: The aim of this study was to find the optimal strength level of QIR for ultra-high-resolution (UHR) PCCT of the lung.

Materials and methods: This retrospective study included 24 patients who had unenhanced chest CT with the novel UHR scan protocol on the PCCT scanner between March 24, 2023 and May 18, 2023. Two sets of reconstructions were made using different slice thicknesses: standard resolution (SR, 1 mm) and ultra-high-resolution (UHR, 0.2 mm), reconstructed with all strength levels of QIR (0 to 4). Attenuation of the lung parenchyma, noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were assessed as objective criteria of image quality. Two fellowship-trained radiologists compared image quality and noise level, sharpness of the images, and the airway details using a 5-point Likert scale. Wilcoxon signed-rank test was used for statistical analysis of reader scores, and one-way repeated measures analysis of variance for comparing the objective image quality scores.

Results: Objective image quality linearly improved with higher strength levels of QIR, reducing image noise by 66% from QIR-0 to QIR-4 (P<0.001). Subjective image noise was best for QIR-4 (P<0.001). Readers rated QIR-1 and QIR-2 best for SR, and QIR-2 and QIR-3 best for UHR in terms of subjective image sharpness and airway detail, without significant differences between them (P=0.48 and 0.56, respectively).

Conclusions: Higher levels of QIR provided excellent objective image quality, but readers' preference was for intermediate levels. Considering all metrics, we recommend QIR-3 for ultra-high-resolution PCCT of the lung.

Purpose: Delays to biopsy and surgery after lung nodule detection can impact survival from lung cancer. The aim of this study was to identify factors associated with delay in a lung cancer screening (LCS) program.

Materials and methods: We evaluated patients in an LCS program from May 2015 through October 2021 with a malignant lung nodule classified as lung CT screening reporting and data system (Lung-RADS) 4B/4X. A cutoff of more than 30 days between screening computed tomography (CT) and first tissue sampling and a cutoff of more than 60 days between screening CT and surgery were considered delayed. We evaluated the relationship between delays to first tissue sampling and surgery and patient sex, age, race, smoking status, median income by zip code, language, Lung-RADS category, and site of surgery (academic vs community hospital).

Results: A total of 185 lung cancers met the inclusion criteria, of which 150 underwent surgical resection. The median time from LCS CT to first tissue sampling was 42 days, and the median time from CT to surgery was 52 days. 127 (69%) patients experienced a first tissue sampling delay and 60 (40%) had a surgical delay. In multivariable analysis, active smoking status was associated with delay to first tissue sampling (odds ratio: 3.0, CI: 1.4-6.6, P = 0.005). Only performing enhanced diagnostic CT of the chest before surgery was associated with delayed lung cancer surgery (odds ratio: 30, CI: 3.6-252, P = 0.02). There was no statistically significant difference in delays with patients' sex, age, race, language, or Lung-RADS category.

Conclusion: Delays to first tissue sampling and surgery in a LCS program were associated with current smoking and performing diagnostic CT before surgery.

Purpose: To explore and compare the diagnostic values of mono-exponential, bi-exponential, and stretched-exponential diffusion-weighted imaging (DWI) parameters of primary lesions and lymph nodes (LNs) to predict mediastinal LN metastasis in patients with non-small cell lung cancer.

Patients and methods: Sixty-one patients with non-small cell lung cancer underwent preoperative magnetic resonance imaging, including multiple b -value DWI. The DWI parameters, including apparent diffusion coefficient (ADC) from a mono-exponential model, true diffusion (D) coefficient, pseudo-diffusion (D*) coefficient, and perfusion fraction (f) from a bi-exponential model, distributed diffusion coefficient (DDC) and intravoxel diffusion heterogeneity index (α) from a stretched-exponential model of primary tumors and LNs and the size characteristics of LNs, were measured and compared. Multivariate logistic regression analysis was used to establish models for predicting mediastinal LN metastasis. Receiver operating characteristic analysis was applied to evaluate diagnostic performances.

Results: The DWI parameters of primary tumors showed no statistical significance between LN metastasis-positive and LN metastasis-negative groups. Nonmetastatic LNs had significantly higher ADC, D, DDC, and α values compared with metastatic LNs (all P < 0.05). The short-dimension, long-dimension, and short-long dimension ratio of metastatic LNs was significantly larger than those of nonmetastatic ones (all P < 0.05). The D value showed the best diagnostic performance among all DWI-derived single parameters, and the short dimension of LNs performed the same among all the size variables. Furthermore, the combination of DWI parameters (ADC and D) and the short dimension of LNs can significantly improve diagnostic efficiency.

Conclusions: The ADC, D, DDC, and α from the mono-exponential, bi-exponential, and stretched-exponential models were demonstrated efficient in differentiating benign from metastatic LNs, and the combination of ADC, D, and short dimension of LNs may have a better diagnostic performance than DWI or size-derived parameters either in combination or individually.

Purpose: Idiopathic pulmonary fibrosis (IPF) is the most common interstitial lung disease, with a median survival time of 2 to 5 years. The focus of this study is to establish a novel imaging biomarker.

Materials and methods: In this study, 79 patients (19% female) with a median age of 70 years were studied retrospectively. Fully automated body composition analysis (BCA) features (bone, muscle, total adipose tissue, intermuscular, and intramuscular adipose tissue) were combined into Sarcopenia, Fat, and Myosteatosis indices and compared between patients with a survival of more or less than 2 years. In addition, we divided the cohort at the median (high=≥ median, low=

Results: A high Sarcopenia and Fat index and low Myosteatosis index were associated with longer median survival (35 vs. 16 mo for high vs. low Sarcopenia index, P=0.066; 44 vs. 14 mo for high vs. low Fat index, P<0.001; and 33 vs. 14 mo for low vs. high Myosteatosis index, P=0.0056) and better 5-year survival rates (34.0% vs. 23.6% for high vs. low Sarcopenia index; 47.3% vs. 9.2% for high vs. low Fat index; and 11.2% vs. 42.7% for high vs. low Myosteatosis index). Adjusted multivariate Cox regression showed a significant impact of the Fat (HR=0.71, P=0.01) and Myosteatosis (HR=1.12, P=0.005) on overall survival.

Conclusion: The fully automated BCA provides biomarkers with a predictive value for the overall survival in patients with IPF.