A case of pituitary metastasis discovered when diabetes insipidus developed in a patient 20 years after breast cancer treatment.

The patient was a 52-year-old woman. She had a history of left breast cancer at age 32 years, with no recurrences. She was examined for a feeling of oral dryness and nocturia, and central diabetes insipidus was diagnosed. A mass was seen in the posterior pituitary on magnetic resonance imaging, and multiple pulmonary nodules were seen on computed tomography. Breast cancer metastases were diagnosed in both tissues. Since this patient had no cancer other than the breast cancer treated 20 years earlier, it was difficult to reach a diagnosis of pituitary metastasis with pituitary gland imaging alone. In estrogen receptor-positive breast cancer, there may be recurrences after a long period of time. It may be that recommending a full body examination could be useful in the differential diagnosis of metastasis even in patients who have had a long disease-free period, if they had undergone surgery for breast cancer.

Superiority of MRI for Evaluation of Sacral Insufficiency Fracture.

STUDY DESIGN: Retrospective observational study. BACKGROUND: Sacral insufficiency fractures (SIF) are relatively rare fractures and difficult to diagnose on plain radiographs. The primary objective of the present study was to evaluate the role of lumbar magnetic resonance imaging (MRI) for the diagnosis of SIF. The secondary objective was to identify the classification of SIF by computed tomography (CT). METHODS: A total of 77 (Male 11, female 66, mean 80.3 years) people were included in this study. Inclusion criteria for this study were: age ≥ 60 years and no history of high energy trauma. Exclusion criteria were high energy trauma and a current history of malignancy. Differences in the fracture detection and description in the various radiologic procedures were evaluated. Fracture patterns were evaluated with CT. The detection rates of additional pathologies in the MRI of the pelvis and lumbar spine were also recorded. RESULTS: The sensitivities for SIF were 28.5% in radiographs and 94.2% in CT, and all fractures were detected in MRI. MRI showed a more complex fracture pattern compared with CT in 65% of the cases. We observed 71.4% of single SIFs, 9.1% with other spinal fractures, 13.0% with other pelvic fractures, and 7.8% with other fractures. According to the SIF fracture pattern, the H/U type was 40.2%, transverse type was 33.7%, λ/T type was 24.7%, unilateral vertical type was 1.3%, and bilateral vertical type was 0%. CONCLUSIONS: an MRI of the lumbar spine including the sacrum with a coronal fat-suppressed T2-weighted image is useful for elderly patients with suddenly increasing low back pain at an early stage. This procedure improves an early SIF detection, recognition of concomitant pathologies, and adequate treatment for the patients.

Bronchogenic cyst with atypical imaging findings and repeated ruptures in a short period of time: A case report.

Bronchogenic cysts are the most common primary cysts of the mediastinum. Although most are asymptomatic, some bronchogenic cysts cause symptoms such as chest pain and dyspnea. Here, we report a case of bronchogenic cyst that ruptured twice in a short period of time in a patient who presented with sudden back pain. The lesion was apparent on computed tomography (CT) as a mass lesion with heterogeneous and high attenuation in the posterior mediastinal region. CT-guided puncture performed for diagnostic purposes revealed the contents as bloody fluid. The patient suffered chest pain approximately 3 months after the first presentation, and re-growth and re-rupture of the mass was suspected. The lesion was surgically resected and pathologically diagnosed as a bronchogenic cyst. Spontaneous rupture is a very rare complication of bronchogenic cyst, usually into the trachea, pleural cavity, or pericardial cavity. However, there are no reports of multiple ruptures. This case highlights the importance of recognizing the atypical imaging findings of bronchogenic cyst and the rare complication of rupture.

Asymptomatic cerebral findings on 3-Tesla MRI in patients with severe carotid artery stenoses.

BACKGROUND AND PURPOSE: Small vessel diseases (SVDs) are often asymptomatic. However, SVDs significantly influence the prognosis in patients with large vessel diseases (LVDs). We investigated asymptomatic cerebral findings on 3-Tesla MRI in patients with severe carotid artery (CA) stenoses, compared to peoples without a past history of neurological disorders, including strokes. METHODS: We retrospectively analyzed the prevalences of various asymptomatic cerebral findings which were intracerebral hemorrhages (ICHs), cortical superficial siderosis, ventricular dilatation (Evans' index) and SVDs including cerebral microbleeds (CMBs), lacunar infarctions (LIs), deep white matter hyperintensities (WMHs), periventricular hyperintensities (PVHs). The prevalence of each finding was compared using multivariate logistic regression models with adjustment for stroke risk factors. RESULTS: We evaluated the findings in 54 patients with severe CA stenosis treated by stenting (CA stenosis group) and 200 adults with health screening tests of the brain and no past history of neurological disorders (control group). Multivariate analyses adjusted for age ≥ 65 years old, female gender, hypertension, hyperlipidemia, diabetes mellitus, alcohol consumption, and smoking index revealed that the prevalences of severe PVHs, severe deep WMHs, asymptomatic deep ICHs, and asymptomatic LIs were significantly higher in the CA stenosis group than the control group. However, there were no significant differences in the prevalences of CMBs, or the remaining asymptomatic findings described above. CONCLUSIONS: With pathological differences between SVD and LVD, asymptomatic SVDs except CMBs and deep ICHs often co-exists severe CA stenosis as a presentative LVD.

Traction Bronchiectasis/Bronchiolectasis on CT Scans in Relationship to Clinical Outcomes and Mortality: The COPDGene Study.

Background The clinical impact of interstitial lung abnormalities (ILAs) on poor prognosis has been reported in many studies, but risk stratification in ILA will contribute to clinical practice. Purpose To investigate the association of traction bronchiectasis/bronchiolectasis index (TBI) with mortality and clinical outcomes in individuals with ILA by using the COPDGene cohort. Materials and Methods This study was a secondary analysis of prospectively collected data. Chest CT scans of participants with ILA for traction bronchiectasis/bronchiolectasis were evaluated and outcomes were compared with participants without ILA from the COPDGene study (January 2008 to June 2011). TBI was classified as follows: TBI-0, ILA without traction bronchiectasis/bronchiolectasis; TBI-1, ILA with bronchiolectasis but without bronchiectasis or architectural distortion; TBI-2, ILA with mild to moderate traction bronchiectasis; and TBI-3, ILA with severe traction bronchiectasis and/or honeycombing. Clinical outcomes and overall survival were compared among the TBI groups and the non-ILA group by using multivariable linear regression model and Cox proportional hazards model, respectively. Results Overall, 5295 participants (median age, 59 years; IQR, 52-66 years; 2779 men) were included, and 582 participants with ILA and 4713 participants without ILA were identified. TBI groups were associated with poorer clinical outcomes such as quality of life scores in the multivariable linear regression model (TBI-0: coefficient, 3.2 [95% CI: 0.6, 5.7; P = .01]; TBI-1: coefficient, 3.3 [95% CI: 1.1, 5.6; P = .003]; TBI-2: coefficient, 7.6 [95% CI: 4.0, 11; P < .001]; TBI-3: coefficient, 32 [95% CI: 17, 48; P < .001]). The multivariable Cox model demonstrated that ILA without traction bronchiectasis (TBI-0-1) and with traction bronchiectasis (TBI-2-3) were associated with shorter overall survival (TBI-0-1: hazard ratio [HR], 1.4 [95% CI: 1.0, 1.9; P = .049]; TBI-2-3: HR, 3.8 [95% CI: 2.6, 5.6; P < .001]). Conclusion Traction bronchiectasis/bronchiolectasis was associated with poorer clinical outcomes compared with the group without interstitial lung abnormalities; TBI-2 and 3 were associated with shorter survival. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Lee and Im in this issue.

Right Epipericardial Fat Necrosis After Thymectomy.

A 62-year-old male patient underwent median sternotomy and thymectomy for thymoma (World Health Organization classification type B1+B3, Masaoka classification stage II). A 10-month follow-up postoperative computed tomography showed a 22-mm nodule in the anterior mediastinal lipid tissue just above the right diaphragm. The patient was followed for 8 years, and the nodule's capsule eventually became calcified. Considering the possibility of postoperative recurrence of thymoma, we excised the nodule. The pathologic diagnosis was epipericardial fat necrosis. In some cases, fat necrosis is difficult to distinguish from malignancy by imaging only; surgical resection has diagnostic significance and should be considered.

Pleural recurrence after transthoracic needle lung biopsy in stage I lung cancer: a systematic review and individual patient-level meta-analysis.

INTRODUCTION: Conflicting results exist regarding whether preoperative transthoracic biopsy increases the risk of pleural recurrence in early lung cancer. We conducted a systematic, patient-level meta-analysis to evaluate the risk of pleural recurrence in stage I lung cancer after percutaneous transthoracic lung biopsy. METHODS: A systematic search of OVID-MEDLINE, Embase and the Cochrane Database of Systematic Reviews was performed through October 2018. Eligible studies were original articles on the risk of pleural recurrence in stage I lung cancer after transthoracic biopsy. We contacted the corresponding authors of eligible studies to obtain individual patient-level data. We used the Fine-Gray model for time to recurrence and lung cancer-specific survival and a Cox proportional hazards model for overall survival. RESULTS: We analysed 2394 individual patient data from 6 out of 10 eligible studies. Compared with other diagnostic procedures, transthoracic biopsy was associated with a higher risk for ipsilateral pleural recurrence, which manifested solely (subdistribution HR (sHR), 2.58; 95% CI 1.15 to 5.78) and concomitantly with other metastases (sHR 1.99; 95% CI 1.14 to 3.48). In the analysis of secondary outcomes considering a significant interaction between diagnostic procedures and age groups, reductions of time to recurrence (sHR, 2.01; 95% CI 1.11 to 3.64), lung cancer-specific survival (sHR 2.53; 95% CI 1.06 to 6.05) and overall survival (HR 2.08; 95% CI 1.12 to 3.87) were observed in patients younger than 55 years, whereas such associations were not observed in other age groups. DISCUSSION: Preoperative transthoracic lung biopsy was associated with increased pleural recurrence in stage I lung cancer and reduced survival in patients younger than 55 years.

Progression of traction bronchiectasis/bronchiolectasis in interstitial lung abnormalities is associated with increased all-cause mortality: Age Gene/Environment Susceptibility-Reykjavik Study.

PURPOSE: The aim of this study is to assess the role of traction bronchiectasis/bronchiolectasis and its progression as a predictor for early fibrosis in interstitial lung abnormalities (ILA). METHODS: Three hundred twenty-seven ILA participants out of 5764 in the Age, Gene/Environment Susceptibility (AGES)-Reykjavik Study who had undergone chest CT twice with an interval of approximately five-years were enrolled in this study. Traction bronchiectasis/bronchiolectasis index (TBI) was classified on a four-point scale: 0, ILA without traction bronchiectasis/bronchiolectasis; 1, ILA with bronchiolectasis but without bronchiectasis or architectural distortion; 2, ILA with mild to moderate traction bronchiectasis; 3, ILA and severe traction bronchiectasis and/or honeycombing. Traction bronchiectasis (TB) progression was classified on a five-point scale: 1, Improved; 2, Probably improved; 3, No change; 4, Probably progressed; 5, Progressed. Overall survival (OS) among participants with different TB Progression Score and between the TB progression group and No TB progression group was also investigated. Hazard radio (HR) was estimated with Cox proportional hazards model. RESULTS: The higher the TBI at baseline, the higher TB Progression Score (P < 0.001). All five participants with TBI = 3 at baseline progressed; 46 (90 %) of 51 participants with TBI = 2 progressed. TB progression was also associated with shorter OS with statistically significant difference (adjusted HR = 1.68, P < 0.001). CONCLUSION: TB progression was visualized on chest CT frequently and clearly. It has the potential to be the predictor for poorer prognosis of ILA.

Usefulness of Preoperative Simulation Using a Stereolithographic 3D Printer in Cerebral Aneurysm Coil Embolization.

Objective: We present a preoperative simulation of cerebral aneurysm coil embolization using a hollow model of cerebral blood vessels created by a stereolithography (SLA) 3D printer. Case Presentation: The patient was a 66-year-old woman. During follow-up, coil embolization was planned for an expanding paraclinoid aneurysm. A hollow cerebral vascular model was created preoperatively using an SLA 3D printer. The catheter was malleable and inserted into the hollow model, which enabled the surgeons to confirm its movement, stability, and ease of insertion. In the surgical procedure, the catheter was easily inserted into the aneurysm without reshaping. The procedure was completed without stability problems. Conclusion: The use of a hollow model of cerebral blood vessels was useful as a preoperative simulation and improved the safety of the procedure.

Very delayed liver metastasis from small bowel gastrointestinal stromal tumor (32 years after resection of the small bowel GIST): Report of a case.

INTRODUCTION: Recurrent forms of gastrointestinal stromal tumor (GIST) include liver metastases and peritoneal dissemination. Recurrence often occurs within 2 years. We report a case of liver metastasis, which was detected 30 years after resection of a primary lesion in the small intestine and was resected 32 years later. PRESENTATION OF CASE: The patient was a 72-year-old woman and was being followed up for ureteral stones at the department of urology of our hospital. Computed tomography (CT) showed a small mass in segment 7 of the liver, 2 years ago. As the tumor gradually increased, a biopsy was performed, and a mesenchymal tumor was diagnosed. The tumor continued to increase in size and partial hepatectomy was performed. GIST was suspected from the sample extracted during hepatectomy. The patient had undergone a resection of the small intestine for a tumor 32 years ago. On tracing her medical records, it was confirmed that resection of the small intestine was performed for the diagnosis of leiomyosarcoma. DISCUSSION: Based on the block specimen from 32 years ago, the tumor of the intestine was confirmed to be GIST, and the liver mass was finally diagnosed as liver metastasis of the GIST that had occurred 32 years ago. CONCLUSION: We experienced a case of liver metastasis 32 years after surgery for the first small intestinal GIST. To the best of our knowledge, this case had the longest disease-free interval before metastasis to the liver.

Combination of Deep Learning-Based Denoising and Iterative Reconstruction for Ultra-Low-Dose CT of the Chest: Image Quality and Lung-RADS Evaluation.

OBJECTIVE. The objective of our study was to assess the effect of the combination of deep learning-based denoising (DLD) and iterative reconstruction (IR) on image quality and Lung Imaging Reporting and Data System (Lung-RADS) evaluation on chest ultra-low-dose CT (ULDCT). MATERIALS AND METHODS. Forty-one patients with 252 nodules were evaluated retrospectively. All patients underwent ULDCT (mean ± SD, 0.19 ± 0.01 mSv) and standard-dose CT (SDCT) (6.46 ± 2.28 mSv). ULDCT images were reconstructed using hybrid iterative reconstruction (HIR) and model-based iterative reconstruction (MBIR), and they were postprocessed using DLD (i.e., HIR-DLD and MBIR-DLD). SDCT images were reconstructed using filtered back projection. Three independent radiologists subjectively evaluated HIR, HIR-DLD, MBIR, and MBIR-DLD images on a 5-point scale in terms of noise, streak artifact, nodule edge, clarity of small vessels, homogeneity of the normal lung parenchyma, and overall image quality. Two radiologists independently evaluated the nodules according to Lung-RADS using HIR, MBIR, HIR-DLD, and MBIR-DLD ULDCT images and SDCT images. The median scores for subjective analysis were analyzed using Wilcoxon signed rank test with Bonferroni correction. Intraobserver agreement for Lung-RADS category between ULDCT and SDCT was evaluated using the weighted kappa coefficient. RESULTS. In the subjective analysis, ULDCT with DLD showed significantly better scores than did ULDCT without DLD (p < 0.001), and MBIR-DLD showed the best scores among the ULDCT images (p < 0.001) for all items. In the Lung-RADS evaluation, HIR showed fair or moderate agreement (reader 1 and reader 2: κw = 0.46 and 0.32, respectively); MBIR, moderate or good agreement (κw = 0.68 and 0.57); HIR-DLD, moderate agreement (κw = 0.53 and 0.48); and MBIR-DLD, good agreement (κw = 0.70 and 0.72). CONCLUSION. DLD improved the image quality of both HIR and MBIR on ULDCT. MBIR-DLD was superior to HIR_DLD for image quality and for Lung-RADS evaluation.

Prognostic impact of mucin spread, tumor cell spread, and invasive size in invasive mucinous adenocarcinoma of the lung.

OBJECTIVES: The pathological T descriptor of lung invasive mucinous adenocarcinoma (IMA) is currently defined according to mucin spread, whereas that of lung non-mucinous adenocarcinoma is defined according to invasive lesion. This study aimed to evaluate and compare the prognostic impact of mucin spread, tumor cell spread, and invasive lesion in patients with lung IMA. MATERIALS AND METHODS: Twenty-seven patients with completely resected pT1-4N0M0 IMA were evaluated. The radiological size (RS), mucin spread size (MS), tumor cell spread size (TS), and invasive size (IS) of the primary tumors were determined. Cox proportional hazards models were used to estimate recurrence-free survival (RFS). Because the MS, TS, and IS may be mutually confounding factors, they were evaluated using separate multivariate models including potential prognostic factors identified as significant on univariate analyses. RESULTS: The median postoperative follow-up time was 4.9 years. TS and IS were significantly smaller than RS by a median of 0.3 cm (p = 0.027) and 1.4 cm (p < 0.0001), whereas MS and RS were not significantly different (p > 0.999). Univariate analyses identified T descriptors defined by MS, TS, and IS as potentially negative prognostic factors, in addition to age >75 years and carcinoembryonic antigen >5 ng/mL. Multivariate analyses revealed that T factors defined by MS, TS, and IS were significant predictors of RFS (p < 0.0001, p = 0.0002, and p = 0.0067, respectively). CONCLUSION: MS is a reasonable determinant of the pathological T descriptor of lung IMA. TS and IS are potential candidates, although they remain discordant with RS. If the TS or IS is to be considered a candidate for the pathological T descriptor of lung IMA, the discordance with RS should first be resolved. If IS is used to define pathological T factor, clear criteria for mucinous AIS/MIA with IMA features should be developed.

The effect of the reconstruction algorithm for the pulmonary nodule detection under the metal artifact caused by a pacemaker.

The aim was to compare the effects of metal artifacts from a pacemaker on pulmonary nodule detection among computed tomography (CT) images reconstructed using filtered back projection (FBP), single-energy metal artifact reduction (SEMAR), and forward-projected model-based iterative reconstruction solution (FIRST).Nine simulated nodules were placed inside a chest phantom with a pacemaker. CT images reconstructed using FBP, SEMAR, and FIRST were acquired at low and standard dose, and were evaluated by 2 independent radiologists.FIRST demonstrated the most significantly improved metal artifact and nodule detection on low dose CT (P < .0032), except at 10 mA and 5-mm thickness. At standard-dose CT, SEMAR showed the most significant metal artifact reduction (P < .00001). In terms of nodule detection, no significant differences were observed between FIRST and SEMAR (P = .161).With a pacemaker present, FIRST showed the best nodule detection ability at low-dose CT and SEMAR is comparable to FIRST at standard dose CT.

Traction Bronchiectasis/Bronchiolectasis is Associated with Interstitial Lung Abnormality Mortality.

PURPOSE: To investigate if the presence and severity of traction bronchiectasis/bronchiolectasis are associated with poorer survival in subjects with ILA. METHOD: The study included 3,594 subjects (378 subjects with ILA and 3,216 subjects without ILA) in AGES-Reykjavik Study. Chest CT scans of 378 subjects with ILA were evaluated for traction bronchiectasis/bronchiolectasis, defined as dilatation of bronchi/bronchioles within areas demonstrating ILA. Traction bronchiectasis/bronchiolectasis Index (TBI) was assigned as: TBI = 0, ILA without traction bronchiectasis/bronchiolectasis: TBI = 1, ILA with bronchiolectasis but without bronchiectasis or architectural distortion: TBI = 2, ILA with mild to moderate traction bronchiectasis: TBI = 3, ILA and severe traction bronchiectasis and/or honeycombing. Overall survival (OS) was compared among the subjects in different TBI groups and those without ILA. RESULTS: The median OS was 12.93 years (95%CI; 12.67 - 13.43) in the subjects without ILA; 11.95 years (10.03 - not reached) in TBI-0 group; 8.52 years (7.57 - 9.30) in TBI-1 group; 7.63 years (6.09 - 9.10) in TBI-2 group; 5.40 years (1.85 - 5.98) in TBI-3 group. The multivariable Cox models demonstrated significantly shorter OS of TBI-1, TBI-2, and TBI-3 groups compared to subjects without ILA (P < 0.0001), whereas TBI-0 group had no significant OS difference compared to subjects without ILA, after adjusting for age, sex, and smoking status. CONCLUSIONS: The presence and severity of traction bronchiectasis/bronchiolectasis are associated with shorter survival. The traction bronchiectasis/bronchiolectasis is an important contributor to increased mortality among subjects with ILA.

Ultra high-resolution computed tomography with 1024-matrix: Comparison with 512-matrix for the evaluation of pulmonary nodules.

PURPOSE: To determine whether a 1024-matrix provides superior image quality for the evaluation of pulmonary nodules. MATERIALS AND METHODS: Prospective evaluation conducted between December 2017 and April 2018, during which CT images showing lung nodules of more than 6 mm and less than 30 mmm were reconstructed with 2 different protocols: 0.5-mm thickness, 512 × 512 matrix, 34.5-cm field of view (FOV) (0.5-512 protocol); and 2-mm thickness, 1024 × 1024 matrix, 34.5-cm FOV (2-1024 protocol). Lung nodule characteristics such as margin, lobulation, pleural indentation, spiculation as well as peripheral vessels and bronchioles visibility and overall image quality were evaluated by three chest radiologists, using a 5-point scale. Image noise was evaluated by measuring the standard deviation in the region of interest for each image. RESULTS: A total of 89 nodules were evaluated. The 2-1024 protocol performed significantly better for the subjective evaluation of pulmonary nodules (p = 0.006 ∼ p < 0.0001). However, image noise was significantly higher both subjectively and objectively (p = 0.036, p < 0.0001). CONCLUSION: The use of a 2-1024 protocol does not increase the amount of images and allows better assessment of pulmonary nodules, despite noise increase.

Influence of field of view size on image quality: ultra-high-resolution CT vs. conventional high-resolution CT.

OBJECTIVES: This study was conducted in order to compare the effect of field of view (FOV) size on image quality between ultra-high-resolution CT (U-HRCT) and conventional high-resolution CT (HRCT). METHODS: Eleven cadaveric lungs were scanned with U-HRCT and conventional HRCT and reconstructed with five FOVs (40, 80, 160, 240, and 320 mm). Three radiologists evaluated and scored the images. Three image evaluations were performed, comparing the image quality with the five FOVs with respect to the 160-mm FOV. The first evaluation was performed on conventional HRCT images, and the second evaluation on U-HRCT images. Images were scored on normal structure, abnormal findings, and overall image quality. The third evaluation was a comparison of the images obtained with conventional HRCT and U-HRCT, with scoring performed on overall image quality. Quantitative evaluation of noise was performed by setting ROIs. RESULTS: In conventional HRCT, image quality was improved when the FOV was reduced to 160 mm. In U-HRCT, image quality, except for noise, improved when the FOV was reduced to 80 mm. In the third evaluation, overall image quality was improved in U-HRCT over conventional HRCT at all FOVs. Noise of U-HRCT increased with respect to conventional HRCT when the FOV was reduced from 160 to 40 mm. However, at 240- and 320-mm FOVs, the noise of U-HRCT and conventional HRCT showed no differences. CONCLUSIONS: In conventional HRCT, image quality did not improve when the FOV was reduced below 160 mm. However, in U-HRCT, image quality improved even when the FOV was reduced to 80 mm. KEY POINTS: • Reducing the size of the field of view to 160 mm improves diagnostic imaging quality in high-resolution CT. • In ultra-high-resolution CT, improvements in image quality can be obtained by reducing the size of the field of view to 80 mm. • Ultra-high-resolution CT produces images of higher quality compared with conventional HRCT irrespective of the size of the field of view.

Pulmonary Textures Classification via a Multi-Scale Attention Network.

Precise classification of pulmonary textures is crucial to develop a computer aided diagnosis (CAD) system of diffuse lung diseases (DLDs). Although deep learning techniques have been applied to this task, the classification performance is not satisfied for clinical requirements, since commonly-used deep networks built by stacking convolutional blocks are not able to learn discriminative feature representation to distinguish complex pulmonary textures. For addressing this problem, we design a multi-scale attention network (MSAN) architecture comprised by several stacked residual attention modules followed by a multi-scale fusion module. Our deep network can not only exploit powerful information on different scales but also automatically select optimal features for more discriminative feature representation. Besides, we develop visualization techniques to make the proposed deep model transparent for humans. The proposed method is evaluated by using a large dataset. Experimental results show that our method has achieved the average classification accuracy of 94.78% and the average f-value of 0.9475 in the classification of 7 categories of pulmonary textures. Besides, visualization results intuitively explain the working behavior of the deep network. The proposed method has achieved the state-of-the-art performance to classify pulmonary textures on high resolution CT images.

Application of deep learning (3-dimensional convolutional neural network) for the prediction of pathological invasiveness in lung adenocarcinoma: A preliminary study.

To compare results for radiological prediction of pathological invasiveness in lung adenocarcinoma between radiologists and a deep learning (DL) system.Ninety patients (50 men, 40 women; mean age, 66 years; range, 40-88 years) who underwent pre-operative chest computed tomography (CT) with 0.625-mm slice thickness were included in this retrospective study. Twenty-four cases of adenocarcinoma in situ (AIS), 20 cases of minimally invasive adenocarcinoma (MIA), and 46 cases of invasive adenocarcinoma (IVA) were pathologically diagnosed. Three radiologists of different levels of experience diagnosed each nodule by using previously documented CT findings to predict pathological invasiveness. DL was structured using a 3-dimensional (3D) convolutional neural network (3D-CNN) constructed with 2 successive pairs of convolution and max-pooling layers, and 2 fully connected layers. The output layer comprises 3 nodes to recognize the 3 conditions of adenocarcinoma (AIS, MIA, and IVA) or 2 nodes for 2 conditions (AIS and MIA/IVA). Results from DL and the 3 radiologists were statistically compared.No significant differences in pathological diagnostic accuracy rates were seen between DL and the 3 radiologists (P >.11). Receiver operating characteristic analysis demonstrated that area under the curve for DL (0.712) was almost the same as that for the radiologist with extensive experience (0.714; P = .98). Compared with the consensus results from radiologists, DL offered significantly inferior sensitivity (P = .0005), but significantly superior specificity (P = .02).Despite the small training data set, diagnostic performance of DL was almost the same as the radiologist with extensive experience. In particular, DL provided higher specificity than radiologists.

Ultra-low-dose chest computed tomography for interstitial lung disease using model-based iterative reconstruction with or without the lung setting.

The aim of this study was to assess the effects of reconstruction on the image quality and quantitative analysis for interstitial lung disease (ILD) using filtered back projection (FBP) and model-based iterative reconstruction (MBIR) with the lung setting and the conventional setting on ultra-low-dose computed tomography (CT).Fifty-two patients with known ILD were prospectively enrolled and underwent CT at an ultra-low dose (0.18 ±â€Š0.02 mSv) and a standard dose (7.01 ±â€Š2.66 mSv). Ultra-low-dose CT was reconstructed using FBP (uFBP) and MBIR with the lung setting (uMBIR-Lung) and the conventional setting (uMBIR-Stnd). Standard-dose CT was reconstructed using FBP (sFBP). Three radiologists subjectively evaluated the images on a 3-point scale (1 = worst, 3 = best). For objective image quality analysis, regions of interest were placed in the lung parenchyma and the axillary fat, and standard deviation (SD), signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were evaluated. For 32 patients with clinically diagnosed idiopathic interstitial pneumonia, quantitative measurements including total lung volume (TLV) and the percentage of ILD volume (%ILDV) were obtained. The medians of 3 radiologists' scores were analyzed using the Wilcoxon signed-rank test and the objective noise was analyzed using the paired t test. The Bonferroni correction was used for multiple comparisons. The quantitative measurements were analyzed using the Bland-Altman method.uMBIR-Lung scored better than uMBIR-Stnd and worse than sFBP (P < .001), except for noise and streak artifact in subjective analysis. The SD decreased significantly in the order of uMBIR-Stnd, uMBIR-Lung, sFBP, and uFBP (P < .001). The SNR and CNR increased significantly in the order of uMBIR-Stnd, uMBIR-Lung, sFBP, and uFBP (P < .001). For TLV, there was no significant bias between ultra-low-dose MBIRs and sFBP (P > .3). For %ILDV, there was no significant bias between uMBIR-Lung and sFBP (p = 0.8), but uMBIR-Stnd showed significantly lower %ILDV than sFBP (P = .013).uMBIR-Lung provided more appropriate image quality than uMBIR-Stnd. Although inferior to standard-dose CT for image quality, uMBIR-Lung showed equivalent CT quantitative measurements to standard-dose CT.

Imaging Patterns Are Associated with Interstitial Lung Abnormality Progression and Mortality.

Rationale: Interstitial lung abnormalities (ILA) are radiologic abnormalities on chest computed tomography scans that have been associated with an early or mild form of pulmonary fibrosis. Although ILA have been associated with radiologic progression, it is not known if specific imaging patterns are associated with progression or risk of mortality. Objectives: To determine the role of imaging patterns on the risk of death and ILA progression. Methods: ILA (and imaging pattern) were assessed in 5,320 participants from the AGES-Reykjavik Study, and ILA progression was assessed in 3,167 participants. Multivariable logistic regression was used to assess factors associated with ILA progression, and Cox proportional hazards models were used to assess time to mortality. Measurements and Main Results: Over 5 years, 327 (10%) had ILA on at least one computed tomography, and 1,435 (45%) did not have ILA on either computed tomography. Of those with ILA, 238 (73%) had imaging progression, whereas 89 (27%) had stable to improved imaging; increasing age and copies of MUC5B genotype were associated with imaging progression. The definite fibrosis pattern was associated with the highest risk of progression (odds ratio, 8.4; 95% confidence interval, 2.7-25; P = 0.0003). Specific imaging patterns were also associated with an increased risk of death. After adjustment, both a probable usual interstitial pneumonia and usual interstitial pneumonia pattern were associated with an increased risk of death when compared with those indeterminate for usual interstitial pneumonia (hazard ratio, 1.7; 95% confidence interval, 1.2-2.4; P = 0.001; hazard ratio, 3.9; 95% confidence interval, 2.3-6.8;P < 0.0001), respectively. Conclusions: In those with ILA, imaging patterns can be used to help predict who is at the greatest risk of progression and early death.