Lung nodule malignancy classification with associated pulmonary fibrosis using 3D attention-gated convolutional network with CT scans.

BACKGROUND: Chest Computed tomography (CT) scans detect lung nodules and assess pulmonary fibrosis. While pulmonary fibrosis indicates increased lung cancer risk, current clinical practice characterizes nodule risk of malignancy based on nodule size and smoking history; little consideration is given to the fibrotic microenvironment. PURPOSE: To evaluate the effect of incorporating fibrotic microenvironment into classifying malignancy of lung nodules in chest CT images using deep learning techniques. MATERIALS AND METHODS: We developed a visualizable 3D classification model trained with in-house CT dataset for the nodule malignancy classification task. Three slightly-modified datasets were created: (1) nodule alone (microenvironment removed); (2) nodule with surrounding lung microenvironment; and (3) nodule in microenvironment with semantic fibrosis metadata. For each of the models, tenfold cross-validation was performed. Results were evaluated using quantitative measures, such as accuracy, sensitivity, specificity, and area-under-curve (AUC), as well as qualitative assessments, such as attention maps and class activation maps (CAM). RESULTS: The classification model trained with nodule alone achieved 75.61% accuracy, 50.00% sensitivity, 88.46% specificity, and 0.78 AUC; the model trained with nodule and microenvironment achieved 79.03% accuracy, 65.46% sensitivity, 85.86% specificity, and 0.84 AUC. The model trained with additional semantic fibrosis metadata achieved 80.84% accuracy, 74.67% sensitivity, 84.95% specificity, and 0.89 AUC. Our visual evaluation of attention maps and CAM suggested that both the nodules and the microenvironment contributed to the task. CONCLUSION: The nodule malignancy classification performance was found to be improving with microenvironment data. Further improvement was found when incorporating semantic fibrosis information.

Comparison of lung cancer occurring in fibrotic versus non-fibrotic lung on chest CT.

PURPOSE: Evaluate the behavior of lung nodules occurring in areas of pulmonary fibrosis and compare them to pulmonary nodules occurring in the non-fibrotic lung parenchyma. METHODS: This retrospective review of chest CT scans and electronic medical records received expedited IRB approval and a waiver of informed consent. 4500 consecutive patients with a chest CT scan report containing the word fibrosis or a specific type of fibrosis were identified using the system M*Model Catalyst (Maplewood, Minnesota, U.S.). The largest nodule was measured in the longest dimension and re-evaluated, in the same way, on the follow-up exam if multiple time points were available. The nodule doubling time was calculated. If the patient developed cancer, the histologic diagnosis was documented. RESULTS: Six hundred and nine patients were found to have at least one pulmonary nodule on either the first or the second CT scan. 274 of the largest pulmonary nodules were in the fibrotic tissue and 335 were in the non-fibrotic lung parenchyma. Pathology proven cancer was more common in nodules occurring in areas of pulmonary fibrosis compared to nodules occurring in areas of non-fibrotic lung (34% vs 15%, p < 0.01). Adenocarcinoma was the most common cell type in both groups but more frequent in cancers occurring in non-fibrotic tissue. In the non-fibrotic lung, 1 of 126 (0.8%) of nodules measuring 1 to 6 mm were cancer. In contrast, 5 of 49 (10.2%) of nodules in fibrosis measuring 1 to 6 mm represented biopsy-proven cancer (p < 0.01). The doubling time for squamous cell cancer was shorter in the fibrotic lung compared to non-fibrotic lung, however, the difference was not statistically significant (p = 0.24). 15 incident lung nodules on second CT obtained ≤ 18 months after first CT scan was found in fibrotic lung and eight (53%) were diagnosed as cancer. CONCLUSIONS: Nodules occurring in fibrotic lung tissue are more likely to be cancer than nodules in the nonfibrotic lung. Incident pulmonary nodules in pulmonary fibrosis have a high likelihood of being cancer.

Chest CT for Breast Cancer Diagnosis.

Background: We report the results of our retrospective analysis of the ability of standard chest CT scans to correctly diagnose cancer in the breast. Methods: Four hundred and fifty-three consecutive women with chest CT scans (contrast and non-contrast) preceding mammograms within one year comprise the study population. All chest CT images were reviewed by an experienced fellowship-trained chest radiologist and mammograms by an experienced fellowship-trained mammographer without the benefit of prior or ancillary studies; only four mammographic views were included for analysis. The size, location, and shape of breast masses were documented; on CT, the average Hounsfield units were measured. On both imaging modalities, the presence of lymphadenopathy, architectural distortion, skin thickening, and microcalcifications were recorded. Ultimately, the interpreting radiologist was asked to decide if a biopsy was indicated, and these recommendations were correlated with the patient's outcome. Findings: Nineteen of four hundred and fifty-three patients had breast cancer at the time of the mammography. Breast masses were the most common finding on chest CT, leading to the recommendation for biopsy. Hounsfield units were the most important feature for discerning benign from malignant masses. CT sensitivity, specificity, and accuracy of CT for breast cancer detection was 84.21%, 99.3%, and 98.68% compared to 78.95%, 93.78%, and 93.16% for four-view mammography. Chest CT scans with or without contrast had similar outcomes for specificity and accuracy, but sensitivity was slightly less without contrast. Chest CT alone, without the benefit of prior exams and patient recall, correctly diagnosed cancer with a p-value of <0.0001 compared to mammography with the same limitations. Conclusion: Chest CT accurately diagnosed breast cancer with few false positives and negatives and did so without the need for patient recall for additional imaging.

Cardiac-Specific Conversion Factors to Estimate Radiation Effective Dose From Dose-Length Product in Computed Tomography.

OBJECTIVES: This study sought to determine updated conversion factors (k-factors) that would enable accurate estimation of radiation effective dose (ED) for coronary computed tomography angiography (CTA) and calcium scoring performed on 12 contemporary scanner models and current clinical cardiac protocols and to compare these methods to the standard chest k-factor of 0.014 mSv·mGy-1cm-1. BACKGROUND: Accurate estimation of ED from cardiac CT scans is essential to meaningfully compare the benefits and risks of different cardiac imaging strategies and optimize test and protocol selection. Presently, ED from cardiac CT is generally estimated by multiplying a scanner-reported parameter, the dose-length product, by a k-factor which was determined for noncardiac chest CT, using single-slice scanners and a superseded definition of ED. METHODS: Metal-oxide-semiconductor field-effect transistor radiation detectors were positioned in organs of anthropomorphic phantoms, which were scanned using all cardiac protocols, 120 clinical protocols in total, on 12 CT scanners representing the spectrum of scanners from 5 manufacturers (GE, Hitachi, Philips, Siemens, Toshiba). Organ doses were determined for each protocol, and ED was calculated as defined in International Commission on Radiological Protection Publication 103. Effective doses and scanner-reported dose-length products were used to determine k-factors for each scanner model and protocol. RESULTS: k-Factors averaged 0.026 mSv·mGy-1cm-1 (95% confidence interval: 0.0258 to 0.0266) and ranged between 0.020 and 0.035 mSv·mGy-1cm-1. The standard chest k-factor underestimates ED by an average of 46%, ranging from 30% to 60%, depending on scanner, mode, and tube potential. Factors were higher for prospective axial versus retrospective helical scan modes, calcium scoring versus coronary CTA, and higher (100 to 120 kV) versus lower (80 kV) tube potential and varied among scanner models (range of average k-factors: 0.0229 to 0.0277 mSv·mGy-1cm-1). CONCLUSIONS: Cardiac k-factors for all scanners and protocols are considerably higher than the k-factor currently used to estimate ED of cardiac CT studies, suggesting that radiation doses from cardiac CT have been significantly and systematically underestimated. Using cardiac-specific factors can more accurately inform the benefit-risk calculus of cardiac-imaging strategies.

A scatter correction method for contrast-enhanced dual-energy digital breast tomosynthesis.

Contrast-enhanced dual energy digital breast tomosynthesis (CE-DE-DBT) is designed to image iodinated masses while suppressing breast anatomical background. Scatter is a problem, especially for high energy acquisition, in that it causes severe cupping artifact and iodine quantitation errors. We propose a patient specific scatter correction (SC) algorithm for CE-DE-DBT. The empirical algorithm works by interpolating scatter data outside the breast shadow into an estimate within the breast shadow. The interpolated estimate is further improved by operations that use an easily obtainable (from phantoms) table of scatter-to-primary-ratios (SPR)--a single SPR value for each breast thickness and acquisition angle. We validated our SC algorithm for two breast emulating phantoms by comparing SPR from our SC algorithm to that measured using a beam-passing pinhole array plate. The error in our SC computed SPR, averaged over acquisition angle and image location, was about 5%, with slightly worse errors for thicker phantoms. The SC projection data, reconstructed using OS-SART, showed a large degree of decupping. We also observed that SC removed the dependence of iodine quantitation on phantom thickness. We applied the SC algorithm to a CE-DE-mammographic patient image with a biopsy confirmed tumor at the breast periphery. In the image without SC, the contrast enhanced tumor was masked by the cupping artifact. With our SC, the tumor was easily visible. An interpolation-based SC was proposed by (Siewerdsen et al 2006 Med. Phys. 33 187-97) for cone-beam CT (CBCT), but our algorithm and application differ in several respects. Other relevant SC techniques include Monte-Carlo and convolution-based methods for CBCT, storage of a precomputed library of scatter maps for DBT, and patient acquisition with a beam-passing pinhole array for breast CT. Our SC algorithm can be accomplished in clinically acceptable times, requires no additional imaging hardware or extra patient dose and is easily transportable between sites.