Radiographic Graft Surveillance in Lung Transplantation: Prognostic Role of Parametric Response Mapping.

Rationale: Chronic lung allograft dysfunction (CLAD) results in significant morbidity after lung transplantation. Potential CLAD occurs when lung function declines to 80-90% of baseline. Better noninvasive tools to prognosticate at potential CLAD are needed. Objectives: To determine whether parametric response mapping (PRM), a computed tomography (CT) voxel-wise methodology applied to high-resolution CT scans, can identify patients at risk of progression to CLAD or death. Methods: Radiographic features and PRM-based CT metrics quantifying functional small airway disease (PRMfSAD) and parenchymal disease (PRMPD) were studied at potential CLAD (n = 61). High PRMfSAD and high PRMPD were defined as ⩾30%. Restricted mean modeling was performed to compare CLAD-free survival among groups. Measurements and Main Results: PRM metrics identified the following three unique signatures: high PRMfSAD (11.5%), high PRMPD (41%), and neither (PRMNormal; 47.5%). Patients with high PRMfSAD or PRMPD had shorter CLAD-free median survival times (0.46 yr and 0.50 yr) compared with patients with predominantly PRMNormal (2.03 yr; P = 0.004 and P = 0.007 compared with PRMfSAD and PRMPD groups, respectively). In multivariate modeling adjusting for single- versus double-lung transplant, age at transplant, body mass index at potential CLAD, and time from transplant to CT scan, PRMfSAD ⩾30% or PRMPD ⩾30% continue to be statistically significant predictors of shorter CLAD-free survival. Air trapping by radiologist interpretation was common (66%), was similar across PRM groups, and was not predictive of CLAD-free survival. Ground-glass opacities by radiologist read occurred in 16% of cases and were associated with decreased CLAD-free survival (P < 0.001). Conclusions: PRM analysis offers valuable prognostic information at potential CLAD, identifying patients most at risk of developing CLAD or death.

Variabilities in Reference Standard by Radiologists and Performance Assessment in Detection of Pulmonary Embolism in CT Pulmonary Angiography.

Annotating lesion locations by radiologists' manual marking is a key step to provide reference standard for the training and testing of a computer-aided detection system by supervised machine learning. Inter-reader variability is not uncommon in readings even by expert radiologists. This study evaluated the variability of the radiologist-identified pulmonary emboli (PEs) to demonstrate the importance of improving the reliability of the reference standard by a multi-step process for performance evaluation. In an initial reading of 40 CTPA PE cases, two experienced thoracic radiologists independently marked the PE locations. For markings from the two radiologists that did not agree, each radiologist re-read the cases independently to assess the discordant markings. Finally, for markings that still disagreed after the second reading, the two radiologists read together to reach a consensus. The variability of radiologists was evaluated by analyzing the agreement between two radiologists. For the 40 cases, 475 and 514 PEs were identified by radiologists R1 and R2 in the initial independent readings, respectively. For a total of 545 marks by the two radiologists, 81.5% (444/545) of the marks agreed but 101 marks in 36 cases differed. After consensus, 65 (64.4%) and 36 (35.6%) of the 101 marks were determined to be true PEs and false positives (FPs), respectively. Of these, 48 and 17 were false negatives (FNs) and 14 and 22 were FPs by R1 and R2, respectively. Our study demonstrated that there is substantial variability in reference standards provided by radiologists, which impacts the performance assessment of a lesion detection system. Combination of multiple radiologists' readings and consensus is needed to improve the reliability of a reference standard.

Development and validation of a radiological diagnosis model for hypersensitivity pneumonitis.

High-resolution computed tomography (HRCT) may be useful for diagnosing hypersensitivity pneumonitis. Here, we develop and validate a radiological diagnosis model and model-based points score.Patients with interstitial lung disease seen at the University of Michigan Health System (derivation cohort) or enrolling in the Lung Tissue Research Consortium (validation cohort) were included. A thin-section, inspiratory HRCT scan was required. Thoracic radiologists documented radiological features.The derivation cohort comprised 356 subjects (33.9% hypersensitivity pneumonitis) and the validation cohort comprised 424 subjects (15.5% hypersensitivity pneumonitis). An age-, sex- and smoking status-adjusted logistic regression model identified extent of mosaic attenuation or air trapping greater than that of reticulation ("MA-AT>Reticulation"; OR 6.20, 95% CI 3.53-10.90; p<0.0001) and diffuse axial disease distribution (OR 2.33, 95% CI 1.31-4.16; p=0.004) as hypersensitivity pneumonitis predictors (area under the receiver operating characteristic curve 0.814). A model-based score >2 (1 point for axial distribution, 2 points for "MA-AT>Reticulation") has specificity 90% and positive predictive value (PPV) 74% in the derivation cohort and specificity 96% and PPV 44% in the validation cohort. Similar model performance is seen with population restriction to those reporting no exposure (score >2: specificity 91%).When radiological mosaic attenuation or air trapping are more extensive than reticulation and disease has diffuse axial distribution, hypersensitivity pneumonitis specificity is high and false diagnosis risk low (<10%), but PPV is diminished in a low-prevalence setting.

How to perform a critical appraisal of diagnostic tests: 7 steps.

The critically appraised topic (CAT) is a format in evidence-based practice for sharing information. A CAT is a standardized way of summarizing the most current research evidence focused on a pertinent clinical question. Its aim is to provide both a critique of the most up-to-date retrieved research and an indication of the clinical relevance of results. A clinical question is initially generated following a patient encounter, which leads to and directs a literature search to answer the clinical question. Studies obtained from the literature search are assigned a level of evidence. This allows the most valid and relevant articles to be selected and to be critically appraised. The results are summarized, and this information is translated into clinically useful procedures and processes.

An investigation of tomosynthesis on the diagnostic efficacy of spot compression mammography.

To explore the diagnostic efficacy of tomosynthesis spot compression (TSC) compared with conventional spot compression (CSC) for ambiguous findings on full-field digital mammography (FFDM). In this retrospective study, 122 patients (including 108 patients with dense breasts) with ambiguous FFDM findings were imaged with both CSC and TSC. Two radiologists independently reviewed the images and evaluated lesions using the Breast Imaging Reporting and Data System. Pathology or at least a 1-year follow-up imaging was used as the reference standard. Diagnostic efficacies of CSC and TSC were compared, including area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). The mean glandular dose was recorded and compared for TSC and CSC. Of the 122 patients, 63 had benign lesions and 59 had malignant lesions. For Reader 1, the following diagnostic efficacies of TSC were significantly higher than those of CSC: AUC (0.988 vs. 0.906, P = 0.001), accuracy (93.4% vs. 77.8%, P = 0.001), specificity (87.3% vs. 63.5%, P = 0.002), PPV (88.1% vs. 70.5%, P = 0.010), and NPV (100% vs. 90.9%, P = 0.029). For Reader 2, TSC showed higher AUC (0.949 vs. 0.909, P = 0.011) and accuracy (83.6% vs. 71.3%, P = 0.022) than CSC. The mean glandular dose of TSC was higher than that of CSC (1.85 ± 0.53 vs. 1.47 ± 0.58 mGy, P < 0.001) but remained within the safety limit. TSC provides better diagnostic efficacy with a slightly higher but tolerable radiation dose than CSC. Therefore, TSC may be a candidate modality for patients with ambiguous findings on FFDM.

Leveraging Serial Low-Dose CT Scans in Radiomics-based Reinforcement Learning to Improve Early Diagnosis of Lung Cancer at Baseline Screening.

Purpose To evaluate the feasibility of leveraging serial low-dose CT (LDCT) scans to develop a radiomics-based reinforcement learning (RRL) model for improving early diagnosis of lung cancer at baseline screening. Materials and Methods In this retrospective study, 1951 participants (female patients, 822; median age, 61 years [range, 55-74 years]) (male patients, 1129; median age, 62 years [range, 55-74 years]) were randomly selected from the National Lung Screening Trial between August 2002 and April 2004. An RRL model using serial LDCT scans (S-RRL) was trained and validated using data from 1404 participants (372 with lung cancer) containing 2525 available serial LDCT scans up to 3 years. A baseline RRL (B-RRL) model was trained with only LDCT scans acquired at baseline screening for comparison. The 547 held-out individuals (150 with lung cancer) were used as an independent test set for performance evaluation. The area under the receiver operating characteristic curve (AUC) and the net reclassification index (NRI) were used to assess the performances of the models in the classification of screen-detected nodules. Results Deployment to the held-out baseline scans showed that the S-RRL model achieved a significantly higher test AUC (0.88 [95% CI: 0.85, 0.91]) than both the Brock model (AUC, 0.84 [95% CI: 0.81, 0.88]; P = .02) and the B-RRL model (AUC, 0.86 [95% CI: 0.83, 0.90]; P = .02). Lung cancer risk stratification was significantly improved by the S-RRL model as compared with Lung CT Screening Reporting and Data System (NRI, 0.29; P < .001) and the Brock model (NRI, 0.12; P = .008). Conclusion The S-RRL model demonstrated the potential to improve early diagnosis and risk stratification for lung cancer at baseline screening as compared with the B-RRL model and clinical models. Keywords: Radiomics-based Reinforcement Learning, Lung Cancer Screening, Low-Dose CT, Machine Learning © RSNA, 2024 Supplemental material is available for this article.

Enhancing Early Lung Cancer Diagnosis: Predicting Lung Nodule Progression in Follow-Up Low-Dose CT Scan with Deep Generative Model.

Early diagnosis of lung cancer can significantly improve patient outcomes. We developed a Growth Predictive model based on the Wasserstein Generative Adversarial Network framework (GP-WGAN) to predict the nodule growth patterns in the follow-up LDCT scans. The GP-WGAN was trained with a training set (N = 776) containing 1121 pairs of nodule images with about 1-year intervals and deployed to an independent test set of 450 nodules on baseline LDCT scans to predict nodule images (GP-nodules) in their 1-year follow-up scans. The 450 GP-nodules were finally classified as malignant or benign by a lung cancer risk prediction (LCRP) model, achieving a test AUC of 0.827 ± 0.028, which was comparable to the AUC of 0.862 ± 0.028 achieved by the same LCRP model classifying real follow-up nodule images (p = 0.071). The net reclassification index yielded consistent outcomes (NRI = 0.04; p = 0.62). Other baseline methods, including Lung-RADS and the Brock model, achieved significantly lower performance (p < 0.05). The results demonstrated that the GP-nodules predicted by our GP-WGAN model achieved comparable performance with the nodules in the real follow-up scans for lung cancer diagnosis, indicating the potential to detect lung cancer earlier when coupled with accelerated clinical management versus the current approach of waiting until the next screening exam.

Recursive Training Strategy for a Deep Learning Network for Segmentation of Pathology Nuclei With Incomplete Annotation.

This study developed a recursive training strategy to train a deep learning model for nuclei detection and segmentation using incomplete annotation. A dataset of 141 H&E stained breast cancer pathologic images with incomplete annotation was randomly split into training/validation set and test set of 89 and 52 images, respectively. The positive training samples were extracted at each annotated cell and augmented with affine translation. The negative training samples were selected from the non-cellular regions free of nuclei using a histogram-based semi-automatic method. A U-Net model was initially trained by minimizing a custom loss function. After the first stage of training, the trained U-Net model was applied to the images in the training set in an inference mode. The U-Net segmented objects with high quality were selected by a semi-automated method. Combining the newly selected high quality objects with the annotated nuclei and the previously generated negative samples, the U-Net model was retrained recursively until the stopping criteria were satisfied. For the 52 test images, the U-Net trained with and without using our recursive training method achieved a sensitivity of 90.3% and 85.3% for nuclei detection, respectively. For nuclei segmentation, the average Dice coefficient and average Jaccard index were 0.831±0.213 and 0.750±0.217, 0.780±0.270 and 0.697±0.264, for U-Net with and without recursive training, respectively. The improvement achieved by our proposed method was statistically significant (P < 0.05). In conclusion, our recursive training method effectively enlarged the set of annotated objects for training the deep learning model and further improved the detection and segmentation performance.

Case Report: Mucocele-Like Tumor of the Breast Associated With Ductal Carcinoma In Situ.

Mucocele-like tumor of the breast is histologically characterized as mucin-containing cysts with mucin leaking to the stroma. It could be associated with atypical ductal hyperplasia (ADH), ductal carcinoma in situ (DCIS), and invasive ductal carcinoma (IDC). We report a case of mucocele-like tumor of the breast associated with DCIS confirmed by paraffin section. We review the literature and discuss the imaging features, pathology, and clinical management of the lesion. These lesions demonstrate characteristic imaging features, and we especially highlight the MR characteristics, as they have not been well documented. Performing a diagnostic fine-needle aspiration cytology (FNAC) of mucocele-like tumor carries a risk of tumor underestimation; therefore, excision for all mucocele-like tumors is suggested to be the best approach. However, some recent reports recommend close follow-up for patients with low-risk factors who have mucocele-like tumor without atypia on FNAC.

Hybrid U-Net-based deep learning model for volume segmentation of lung nodules in CT images.

OBJECTIVE: Accurate segmentation of the lung nodule in computed tomography images is a critical component of a computer-assisted lung cancer detection/diagnosis system. However, lung nodule segmentation is a challenging task due to the heterogeneity of nodules. This study is to develop a hybrid deep learning (H-DL) model for the segmentation of lung nodules with a wide variety of sizes, shapes, margins, and opacities. MATERIALS AND METHODS: A dataset collected from Lung Image Database Consortium image collection containing 847 cases with lung nodules manually annotated by at least two radiologists with nodule diameters greater than 7 mm and less than 45 mm was randomly split into 683 training/validation and 164 independent test cases. The 50% consensus consolidation of radiologists' annotation was used as the reference standard for each nodule. We designed a new H-DL model combining two deep convolutional neural networks (DCNNs) with different structures as encoders to increase the learning capabilities for the segmentation of complex lung nodules. Leveraging the basic symmetric U-shaped architecture of U-Net, we redesigned two new U-shaped deep learning (U-DL) models that were expanded to six levels of convolutional layers. One U-DL model used a shallow DCNN structure containing 16 convolutional layers adapted from the VGG-19 as the encoder, and the other used a deep DCNN structure containing 200 layers adapted from DenseNet-201 as the encoder, while the same decoder with only one convolutional layer at each level was used in both U-DL models, and we referred to them as the shallow and deep U-DL models. Finally, an ensemble layer was used to combine the two U-DL models into the H-DL model. We compared the effectiveness of the H-DL, the shallow U-DL and the deep U-DL models by deploying them separately to the test set. The accuracy of volume segmentation for each nodule was evaluated by the 3D Dice coefficient and Jaccard index (JI) relative to the reference standard. For comparison, we calculated the median and minimum of the 3D Dice and JI over the individual radiologists who segmented each nodule, referred to as M-Dice, min-Dice, M-JI, and min-JI. RESULTS: For the 164 test cases with 327 nodules, our H-DL model achieved an average 3D Dice coefficient of 0.750 ± 0.135 and an average JI of 0.617 ± 0.159. The radiologists' average M-Dice was 0.778 ± 0.102, and the average M-JI was 0.651 ± 0.127; both were significantly higher than those achieved by the H-DL model (p < 0.05). The radiologists' average min-Dice (0.685 ± 0.139) and the average min-JI (0.537 ± 0.153) were significantly lower than those achieved by the H-DL model (p < 0.05). The results indicated that the H-DL model approached the average performance of radiologists and was superior to the radiologist whose manual segmentation had the min-Dice and min-JI. Moreover, the average Dice and average JI achieved by the H-DL model were significantly higher than those achieved by the individual shallow U-DL model (Dice of 0.745 ± 0.139, JI of 0.611 ± 0.161; p < 0.05) or the individual deep U-DL model alone (Dice of 0.739 ± 0.145, JI of 0.604 ± 0.163; p < 0.05). CONCLUSION: Our newly developed H-DL model outperformed the individual shallow or deep U-DL models. The H-DL method combining multilevel features learned by both the shallow and deep DCNNs could achieve segmentation accuracy comparable to radiologists' segmentation for nodules with wide ranges of image characteristics.