Predicting Usual Interstitial Pneumonia Histopathology From Chest CT Imaging With Deep Learning.

BACKGROUND: Idiopathic pulmonary fibrosis (IPF) is a progressive, often fatal form of interstitial lung disease (ILD) characterized by the absence of a known cause and usual interstitial pneumonitis (UIP) pattern on chest CT imaging and/or histopathology. Distinguishing UIP/IPF from other ILD subtypes is essential given different treatments and prognosis. Lung biopsy is necessary when noninvasive data are insufficient to render a confident diagnosis. RESEARCH QUESTION: Can we improve noninvasive diagnosis of UIP be improved by predicting ILD histopathology from CT scans by using deep learning? STUDY DESIGN AND METHODS: This study retrospectively identified a cohort of 1,239 patients in a multicenter database with pathologically proven ILD who had chest CT imaging. Each case was assigned a label based on histopathologic diagnosis (UIP or non-UIP). A custom deep learning model was trained to predict class labels from CT images (training set, n = 894) and was evaluated on a 198-patient test set. Separately, two subspecialty-trained radiologists manually labeled each CT scan in the test set according to the 2018 American Thoracic Society IPF guidelines. The performance of the model in predicting histopathologic class was compared against radiologists' performance by using area under the receiver-operating characteristic curve as the primary metric. Deep learning model reproducibility was compared against intra-rater and inter-rater radiologist reproducibility. RESULTS: For the entire cohort, mean patient age was 62 ± 12 years, and 605 patients were female (49%). Deep learning performance was superior to visual analysis in predicting histopathologic diagnosis (area under the receiver-operating characteristic curve, 0.87 vs 0.80, respectively; P < .05). Deep learning model reproducibility was significantly greater than radiologist inter-rater and intra-rater reproducibility (95% CI for difference in Krippendorff's alpha did not include zero). INTERPRETATION: Deep learning may be superior to visual assessment in predicting UIP/IPF histopathology from CT imaging and may serve as an alternative to invasive lung biopsy.

Observer Performance for Detection of Pulmonary Nodules at Chest CT over a Large Range of Radiation Dose Levels.

Background There is a wide variation in radiation dose levels that can be used with chest CT in order to detect indeterminate pulmonary nodules. Purpose To compare the performance of lower-radiation-dose chest CT with that of routine dose in the detection of indeterminate pulmonary nodules 5 mm or greater. Materials and Methods In this retrospective study, CT projection data from 83 routine-dose chest CT examinations performed in 83 patients (120 kV, 70 quality reference mAs [QRM]) were collected between November 2013 and April 2014. Reference indeterminate pulmonary nodules were identified by two nonreader thoracic radiologists. By using validated noise insertion, five lower-dose data sets were reconstructed with filtered back projection (FBP) or iterative reconstruction (IR; 30 QRM with FBP, 10 QRM with IR, 5 QRM with FBP, 5 QRM with IR, and 2.5 QRM with IR). Three thoracic radiologists circled pulmonary nodules, rating confidence that the nodule was a 5-mm-or-greater indeterminate pulmonary nodule, and graded image quality. Analysis was performed on a per-nodule basis by using jackknife alternative free-response receiver operating characteristic figure of merit (FOM) and noninferiority limit of -0.10. Results There were 66 indeterminate pulmonary nodules (mean size, 8.6 mm ± 3.4 [standard deviation]; 21 part-solid nodules) in 42 patients (mean age, 51 years ± 17; 21 men and 21 women). Compared with the FOM for routine-dose CT (size-specific dose estimate, 6.5 mGy ± 1.8; FOM, 0.86 [95% confidence interval: 0.80, 0.91]), FOM was noninferior for all lower-dose configurations except for 2.5 QRM with IR. The sensitivity for subsolid nodules at 70 QRM was 60% (range, 48%-72%) and was significantly worse at a dose of 5 QRM and lower, whether or not IR was used (P < .05). Diagnostic image quality decreased with decreasing dose (P < .001) and was better with IR at 5 QRM (P < .05). Conclusion CT images reconstructed at dose levels down to 10 quality reference mAs (size-specific dose estimate, 0.9 mGy) had noninferior performance compared with routine dose in depicting pulmonary nodules. Iterative reconstruction improved subjective image quality but not performance at low dose levels. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by White and Kazerooni in this issue.

Prospective Pilot Evaluation of Radiologists and Computer-aided Pulmonary Nodule Detection on Ultra-low-Dose CT With Tin Filtration.

PURPOSE: The aim of this study was to evaluate the ability of computer-aided detection (CAD) and human readers to detect pulmonary nodules ≥5 mm using 100 kV ultra-low-dose computed tomography (ULDCT) utilizing a tin filter. MATERIALS AND METHODS: After informed consent, 55 patients prospectively underwent standard-dose chest CT (SDCT) using 120 kV followed by ULDCT using 100 kV/tin. Reference nodules ≥5 mm were identified by a thoracic radiologist using SDCT. Four thoracic radiologists marked detected nodules on SDCT and ULDCT examinations using a dedicated computer workstation. After a 6-month memory extinction, readers were shown the same ULDCT cases with all CAD markings as well as their original detections, and characterized CAD detections as true positive or false positive. RESULTS: Volume CT Dose index (CTDIvol) for SDCT and ULDCT were 5.3±2 and 0.4±0.2 mGy (P<0.0001), respectively. Forty-five reference nodules were detected in 30 patients. Reader sensitivity varied widely but similarly for SDCT (ranging from 45% to 87%) and ULDCT (45% to 83%). CAD sensitivity was 76% (34/45) for SDCT and 71% (32/45) for ULDCT. After CAD, reader sensitivity substantially improved by 19% and 18% for 2 readers, and remained nearly unchanged for the other 2 readers (0% and 2%), despite reader perception that many more nodules were identified with CAD. There was a mean of 2 false-positive CAD detections/case. CONCLUSIONS: ULDCT with 100 kV/tin reduced patient dose by over 90% without compromising pulmonary nodule detection sensitivity. CAD can substantially improve nodule detection sensitivity at ULDCT for some readers, maintaining interobserver performance.

Estimation of Observer Performance for Reduced Radiation Dose Levels in CT: Eliminating Reduced Dose Levels That Are Too Low Is the First Step.

RATIONALE AND OBJECTIVES: This study aims to estimate observer performance for a range of dose levels for common computed tomography (CT) examinations (detection of liver metastases or pulmonary nodules, and cause of neurologic deficit) to prioritize noninferior dose levels for further analysis. MATERIALS AND METHODS: Using CT data from 131 examinations (abdominal CT, 44; chest CT, 44; head CT, 43), CT images corresponding to 4%-100% of the routine clinical dose were reconstructed with filtered back projection or iterative reconstruction. Radiologists evaluated CT images, marking specified targets, providing confidence scores, and grading image quality. Noninferiority was assessed using reference standards, reader agreement rules, and jackknife alternative free-response receiver operating characteristic figures of merit. Reader agreement required that a majority of readers at lower dose identify target lesions seen by the majority of readers at routine dose. RESULTS: Reader agreement identified dose levels lower than 50% and 4% to have inadequate performance for detection of hepatic metastases and pulmonary nodules, respectively, but could not exclude any low dose levels for head CT. Estimated differences in jackknife alternative free-response receiver operating characteristic figures of merit between routine and lower dose configurations found that only the lowest dose configurations tested (ie, 30%, 4%, and 10% of routine dose levels for abdominal, chest, and head CT examinations, respectively) did not meet criteria for noninferiority. At lower doses, subjective image quality declined before observer performance. Iterative reconstruction was only beneficial when filtered back projection did not result in noninferior performance. CONCLUSION: Opportunity exists for substantial radiation dose reduction using existing CT technology for common diagnostic tasks.

Addition of the Fleischner Society Guidelines to Chest CT Examination Interpretive Reports Improves Adherence to Recommended Follow-up Care for Incidental Pulmonary Nodules.

RATIONALE AND OBJECTIVES: The study aimed to determine whether the addition of the Fleischner Society guidelines to chest computed tomography (CT) reports identifying incidental pulmonary nodules affects follow-up care. PATIENTS AND METHODS: Beginning in 2008, a template containing the Fleischner Society guidelines was added at the interpreting radiologist's discretion to chest CT reports describing incidental solid pulmonary nodules at our institution. The records of all medical centers in Olmsted county were used to capture the complete medical history of local patients >35 years old diagnosed with a pulmonary nodule from April 1, 2008 to October 1, 2011. Patients with a history of cancer or previously diagnosed nodule, or who died before follow-up, were excluded. Patients were categorized according to whether they did ("template group") or did not ("control group") have the template added. Nodule size and smoking history were used to determine recommended follow-up care. Differences in follow-up were compared between groups using Pearson's chi-square test. RESULTS: A total of 510 patients (276 in the template group, 234 in the control group) were included in the study. Only 198 patients (39%) received their recommended follow-up care. Template group patients were significantly more likely to receive recommended follow-up care compared to control group patients (45% vs 31%, P = .0014). Most patients whose management did not adhere to Fleischner Society guidelines did not receive a recommended follow-up chest CT (210 out of 312, 67%). CONCLUSIONS: The addition of the Fleischner Society guidelines to chest CT reports significantly increases the likelihood of receiving recommended follow-up care for patients with incidental pulmonary nodules. Additional education is needed to improve appropriate guideline utilization by radiologists and adherence by ordering providers.

Magnetic Resonance Imaging of Part-solid Nodules: A Pilot Study.

PURPOSE: The aim of the study was to assess whether magnetic resonance imaging (MRI) characteristics can distinguish benign from malignant part-solid pulmonary nodules and predict the aggressiveness of the latter. We also sought to compare MRI-derived parameters with morphologic and physiological values derived from conventional examinations such as computed tomography and positron emission tomography/computed tomography. MATERIALS AND METHODS: This was an institutional review board-approved pilot study of 28 participants (23 women, mean age 73.5±13.8 y) with 32 biopsy-proven lesions. 3-T unenhanced pulmonary MRI examinations were performed with regions of interest drawn around lesions for T1, T2, T2*, and diffusion-weighted sequences. Apparent diffusion coefficient (ADC) and T2* values were calculated. Two weeks later the regions of interest were redrawn. MRI parameters were compared with lesion pathology, maximal standard uptake value (SUVmax), and Hounsfield units (HU). MRI lesion visibility was correlated with solid component size and the percentage of solid component. Intraobserver and interobserver agreements were determined. RESULTS: Only ADC values correlated with malignancy (P<0.05). ADC≥1.28 µm/ms predicted malignancy with 83.3% sensitivity (area under the curve 0.79). ADC and T2* correlated with adenocarcinoma subtypes (P<0.05). No MRI parameters predicted tumor differentiation (P>0.11). SUVmax did not correlate with any MRI parameters (P>0.56). Visibility on T1-weighted images correlated with the percentage of solid components (P<0.03). T1 and T2 values showed significant correlation with HU measurements of the entire nodule (P<0.001 and P<0.024, respectively) and HU measurements of solid components (P=0.031 and 0.008, respectively). CONCLUSIONS: 3 T MRI with quantitative ADC values demonstrated potential for discriminating benign part-solid pulmonary nodules from malignant lesions. ADC and T2* values correlated with adenocarcinoma subtypes. No MRI parameters correlated with SUVmax. T1 and T2 values showed significant correlation with HU measurements of the entire nodule and of the solid components.

Amyloid-associated Cystic Lung Disease.

BACKGROUND: Amyloid-associated cystic lung disease is rare. It can be associated with collagen vascular disease (CVD). We aimed to describe the clinical, radiology, and pathology findings of this entity. METHODS: We reviewed the records of subjects having biopsy-proven pulmonary amyloidosis with cystic lung disease demonstrated at high-resolution computed tomography (HRCT). Demographic characteristics, association with CVD and lymphoproliferative disorders, pulmonary function, and pathology results were reviewed. HRCT appearance was analyzed for number, size, distribution, and morphology of cysts and nodules. RESULTS: Twenty-one subjects (13 female, eight male; median age, 61 years) with cystic pulmonary amyloidosis were identified. The most common pulmonary function patterns were normal (42%) and obstructive (32%). The most common associated CVD was Sjögren syndrome (10 of 12). Nine subjects had no CVD. Cysts tended to be multiple (≥ 10 in 14 of 21, 67%), round (21 of 21, 100%), or lobulated (20 of 21, 95%); thin-walled (< 2 mm in 17 of 21, 81%); and of small (< 1 cm in 21 of 21, 100%) to moderate (1-2 cm in 17 of 21, 81%) size. Peribronchovascular (19 of 21, 90%) and subpleural (19 of 21, 90%) cysts were typically present. Seventeen (81%) subjects had lung nodules, which tended to be numerous (≥ 10 in 10 of 17, 59%; 4-9 in six of 17, 35%). At least one calcified nodule was present in 14 of 17 subjects (82%). Pulmonary mucosa-associated lymphoid tissue lymphoma (MALToma) was diagnosed in seven subjects (33%). CONCLUSIONS: Amyloid-associated cystic lung disease can occur with or without underlying CVD. Cystic lesions in the lung are commonly numerous, often are peribronchovascular or subpleural, and are frequently associated with nodular lesions that are often calcified. MALToma was a relatively frequent association.

Establishing a chest MRI practice and its clinical applications: our insight and protocols.

Despite its nonionizing technique and exquisite soft tissue characterization, noncardiovascular, and nonmusculoskeletal magnetic resonance imaging (MRI) of the chest has been considered impractical due to various challenges such as respiratory motion, cardiac motion, vascular pulsatility, air susceptibility, and paucity of signal in the lung. With advances in MRI, it is now possible to perform diagnostically useful and good quality MRIs of the chest, but literature on subspecialized chest MRI practices is limited. The purpose of this manuscript is to describe the rationale, nuances, and logistics that went into developing such a practice in the Division of Thoracic Radiology at our institution. The topics addressed include technical and clinical considerations, support at administrative and clinical levels, protocol development, and economic considerations compared with conventional practices. Various MRI techniques are also specifically discussed to facilitate chest MRI at other sites. Although chest MRI is used in a relatively small number of patients at this point, in certain patients, chest MRI can provide additional information to optimize medical management. A few clinical cases illustrate the quality and clinical utility of chest MRI. Given recent advances in MRI techniques, it is now an opportune time to develop a chest MRI practice.

Endobronchial ultrasound and lymphoproliferative disorders: a retrospective study.

BACKGROUND: Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) has been shown to have excellent diagnostic performance for mediastinal staging of lung cancer. The utility of EBUS-TBNA for the diagnosis of lymphoproliferative disorders involving the mediastinum or hila, or both, is unclear. METHODS: A retrospective analysis was completed of all patients diagnosed with a lymphoproliferative disorder involving the mediastinum or hila, or both, who underwent an EBUS-TBNA within 3 months of the diagnosis. RESULTS: Sixty-five patients with mediastinal or hilar lymph node, or both, involvement of their lymphoproliferative disorder underwent EBUS-TBNA within 3 months of their diagnosis. The initial EBUS-TBNA was nondiagnostic in 34 (52%), 11 were subsequently diagnosed by mediastinoscopy, and the remaining 23 were diagnosed by biopsy of a distant site, with involvement of the mediastinum or hilum assumed from preestablished radiographic criteria. A EBUS-TBNA specimen in 31 patients (48%) was interpreted as consistent with or suspicious for a lymphoproliferative disorder. The overall sensitivity of EBUS-TBNA for establishing a definitive diagnosis was 25 of 65 (38%). The sensitivity was lower for new patients, at 7 of 32 (22%), and better for patients with recurrence, at 18 of 33 (55%). CONCLUSIONS: Contrary to previous studies, our findings suggest that EBUS-TBNA does not provide sufficient diagnostic material for accurate lymphoproliferative disorder subtyping in a significant number of patients and performs especially poorly when evaluating new patients. Mediastinoscopy should still be considered as the initial diagnostic procedure of choice when the clinical suspicion for a lymphoproliferative disorder is high, unless the patient is being evaluated for a recurrence of prior disorder.

Cystic lung disease is not uncommon in men with tuberous sclerosis complex.

OBJECTIVE: To assess the frequency of cystic lung disease suggestive of pulmonary lymphangiomyomatosis in men with tuberous sclerosis complex. PATIENTS AND METHODS: Retrospective review of CT scans of the chest and abdomen on 29 men with tuberous sclerosis complex encountered during a 13-year period, 1998 to 2010. RESULTS: Cystic lung disease (defined as ≥4 cysts) was seen in 11 of 29 men (38%) with tuberous sclerosis complex. The mean age of those with cystic lung disease was 46.3 ± 19.1 years. None of the patients had experienced pneumothorax or chylothorax and none had undergone a lung biopsy for evaluation of cystic lung disease. Lymphangiomyomatosis had been diagnosed in two of 11 patients with cystic lung disease. Renal angiomyolipomas were demonstrated in 14 of 25 patients (56%) with renal imaging available; all 14 had multiple bilateral renal involvement. In this limited cohort of patients, the presence of cystic lung disease did not correlate with age, smoking exposure or the presence of renal AMLs. CONCLUSION: We conclude that tuberous sclerosis complex -related lymphangiomyomatosis may not be rare in men but is milder in severity.

Computed tomography of benign intrapulmonary lymph nodes: retrospective comparison with sarcoma metastases.

OBJECTIVE: To determine whether the computed tomographic (CT) characteristics of benign intrapulmonary lymph nodes and small sarcoma metastases are sufficiently characteristic to allow specific prospective identification. PATIENTS AND METHODS: Preoperative chest CT scans of 41 patients with benign intrapulmonary lymph nodes and 33 patients with sarcoma metastases seen from 1991 through 1996 were retrospectively reviewed and correlated with pathologic findings. RESULTS: Fifty-seven benign intrapulmonary lymph nodes were found. Twenty-six (46%) were subpleural, 38 (67%) were oval, and 46 (81%) were located in the lower portions of the lungs; 43 (75%) had a lymphatic distribution on CT and 54 (95%) at pathologic review. Ninety-eight sarcoma metastases were found. Thirteen (13%) were subpleural, 15 (15%) were oval, and 56 (57%) were in the lower portions of the lungs; 29 (30%) had a lymphatic distribution on CT and 45 (46%) at pathologic review. CONCLUSION: Benign intrapulmonary lymph nodes were more likely than sarcoma metastases to be oval, to occur in a lymphatic distribution, and to be located subpleurally.