In addition to lung cancer, other thoracic abnormalities, such as emphysema, can be visualized within low-dose CT scans that were initially obtained in cancer screening programs, and thus, ...opportunistic evaluation of these diseases may be highly valuable. However, manual assessment for each scan is tedious and often subjective, thus we have developed an automatic, rapid computer-aided diagnosis system for emphysema using attention-based multiple instance deep learning and 865 LDCTs. In the task of determining if a CT scan presented with emphysema or not, our novel Transfer AMIL approach yielded an area under the ROC curve of 0.94 ± 0.04, which was a statistically significant improvement compared to other methods evaluated in our study following the Delong Test with correction for multiple comparisons. Further, from our novel attention weight curves, we found that the upper lung demonstrated a stronger influence in all scan classes, indicating that the model prioritized upper lobe information. Overall, our novel Transfer AMIL method yielded high performance and provided interpretable information by identifying slices that were most influential to the classification decision, thus demonstrating strong potential for clinical implementation.
Objective Surgical management is a critical component of computed tomography (CT) screening for lung cancer. We report the results for US sites in a large ongoing screening program, the International ...Early Lung Cancer Action Program (I-ELCAP). Methods We identified all patients who underwent surgical resection. We compared the results before (1993-2005) and after (2006-2011) termination of the National Lung Screening Trial to identify emerging trends. Results Among 31,646 baseline and 37,861 annual repeat CT screenings, 492 patients underwent surgical resection; 437 (89%) were diagnosed with lung cancer; 396 (91%) had clinical stage I disease. In the 54 (11%) patients with nonmalignant disease, resection was sublobar in 48 and lobectomy in 6. The estimated cure rate based on the 15-year Kaplan-Meier survival for all 428 patients (excluding 9 typical carcinoids) with lung cancer was 84% (95% confidence interval CI, 80%-88%) and 88% (95% CI, 83%-92%) for clinical stage I disease resected within 1 month of diagnosis. Video-assisted thoracoscopic surgery and sublobar resection increased significantly, from 10% to 34% ( P < .0001) and 22% to 34% ( P = .01) respectively; there were no significant differences in the percentage of malignant diagnoses (90% vs 87%, P = .36), clinical stage I (92% vs 89%, P = .33), pathologic stage I (85% vs 82%, P = .44), tumor size ( P = .61), or cell type ( P = .81). Conclusions The frequency and extent of surgery for nonmalignant disease can be minimized in a CT screening program and provide a high cure rate for those diagnosed with lung cancer and undergoing surgical resection.
The objective of this study is to evaluate measurement variability in volumetric assessment of pulmonary nodules on low-dose CT images with a view toward determining how this variability is ...influenced by nodule size.
A large CT screening database was reviewed to identify solid pulmonary nodules that had remained stable in size on the basis of findings from at least three scans obtained over a 2-year period. Two software packages (Lung VCAR and syngo.via) were used to assess the nodule volume on the two most recent CT scans, which were obtained at a slice thickness of 0.625 mm. The percentage of volume change was calculated for each nodule. The SD of the percentage of volume change was determined for nodules in each of the following nodule diameter size categories: less than 4 mm, 4-5 mm, 6-9 mm, and 10 mm or larger. The diameter was the mean of the length and width in the CT image that represented the largest cross-sectional area of the nodule.
The 171 stable nodules that were identified in 117 CT screening participants (median age, 61 years) ranged in size from 2.2 to 18.7 mm. The time between acquisition of the first and last CT images ranged from 3.7 to 17.8 years (median, 11.5 years). For each of the four categories of diameter size (< 4, 4-5, 6-9, and ≥ 10 mm), the SD of the percentage of volume change was 20.4%, 17.7%, 14.6%, and 3.7%, with the use of Lung VCAR, and 59.5%, 24.3%, 9.1%, and 6.2%, with the use of syngo.via, respectively. The SD decreased with increasing nodule diameter, with the use of both software packages.
Measurement variability decreased with increasing nodule diameter for both software packages and was different between the two software packages.
Objective Our objective was to assess the usefulness of emphysema scores in predicting death from COPD and lung cancer. Methods Emphysema was assessed with low-dose CT scans performed on 9,047 men ...and women for whom age and smoking history were documented. Each scan was scored according to the presence of emphysema as follows: none, mild, moderate, or marked. Follow-up time was calculated from time of CT scan to time of death or December 31, 2007, whichever came first. Cox regression analysis was used to calculate the hazard ratio (HR) of emphysema as a predictor of death. Results Median age was 65 years, 4,433 (49%) were men, and 4,133 (46%) were currently smoking or had quit within 5 years. Emphysema was identified in 2,637 (29%) and was a significant predictor of death from COPD (HR, 9.3; 95% CI, 4.3-20.2; P < .0001) and from lung cancer (HR, 1.7; 95% CI, 1.1-2.5; P = .013), even when adjusted for age and smoking history. Conclusions Visual assessment of emphysema on CT scan is a significant predictor of death from COPD and lung cancer.
The rates of resection of nonmalignant lung nodules suspected preoperatively to be lung cancer vary widely and are reported to be as high as 40%. We determined the impact of the frequent use of ...computed tomography (CT)-guided fine needle aspiration (FNA) on the resection rate of nonmalignant nodules and frequency of resections of benign disease among patients undergoing evaluation for lung cancer resection operation in an academic medical center.
Eligible patients underwent CT-guided FNA, surgical resection, or both during the 12-month period between July 2013 and July 2014 for known or suspected first primary resectable stage I-III lung cancer. Patient data were extracted from the electronic medical records.
One hundred ninety-seven patients underwent surgical resection; among them the overall resection rate of nonmalignant lesions was 13.1% (26/197). For those with preoperative FNA, the rate was 7.9% (11/139), and for those with no biopsy, the rate was 25.9% (15/58) (p = 0.001). The sensitivity and specificity of FNA biopsy were 96% and 98%, respectively. The false-negative rate was 3.9% (5/128).
The resection rate of nonmalignant nodules was significantly lower for patients with preoperative CT-guided FNA biopsy than in those without. The diagnostic accuracy of FNA in these patients at moderate to high risk for lung cancer is higher than that of positron emission tomography, with a low rate of adverse events. These findings suggest that the frequent use of preoperative diagnostic confirmation by FNA results in a low rate of nonmalignant resection.
Emphysema phenotypes and lung cancer risk González, Jessica; Henschke, Claudia I; Yankelevitz, David F ...
PloS one,
07/2019, Letnik:
14, Številka:
7
Journal Article
Recenzirano
Odprti dostop
To assess the relationship between lung cancer and emphysema subtypes.
Airflow obstruction and emphysema predispose to lung cancer. Little is known, however, about the lung cancer risk associated ...with different emphysema phenotypes. We assessed the risk of lung cancer based on the presence, type and severity of emphysema, using visual assessment.
Seventy-two consecutive lung cancer cases were selected from a prospective cohort of 3,477 participants enrolled in the Clínica Universidad de Navarra's lung cancer screening program. Each case was matched to three control subjects using age, sex, smoking history and body mass index as key variables. Visual assessment of emphysema and spirometry were performed. Logistic regression and interaction model analysis were used in order to investigate associations between lung cancer and emphysema subtypes.
Airflow obstruction and visual emphysema were significantly associated with lung cancer (OR = 2.8, 95%CI: 1.6 to 5.2; OR = 5.9, 95%CI: 2.9 to 12.2; respectively). Emphysema severity and centrilobular subtype were associated with greater risk when adjusted for confounders (OR = 12.6, 95%CI: 1.6 to 99.9; OR = 34.3, 95%CI: 25.5 to 99.3, respectively). The risk of lung cancer decreases with the added presence of paraseptal emphysema (OR = 4.0, 95%CI: 3.6 to 34.9), losing this increased risk of lung cancer when it occurs alone (OR = 0.7, 95%CI: 0.5 to 2.6).
Visual scoring of emphysema predicts lung cancer risk. The centrilobular phenotype is associated with the greatest risk.
Objectives
To determine the frequency of adrenal enlargement of participants in a CT-screening program for lung cancer and demonstrate the progression during follow-up, separately for baseline and ...annual repeat rounds.
Materials and methods
HIPAA-compliant informed consent was obtained in 4,776 participants. The adrenal gland was defined as enlarged if it measured ≥6 mm at its largest diameter. Logistic regression analyses were performed.
Results
At baseline, 202 (4 %) of 4,776 participants had adrenal enlargement. Significant factors were age (OR = 1.4, 95 % CI: 1.2–1.7) and current smoker (OR = 1.8, 95 % CI: 1.3–2.4). Follow-up 7–18 months after baseline for 133 cases with adrenal enlargement <40 mm showed it decreased or was stable in 85 (64 %), and increased by <10 mm in 48 (36 %). Five (0.04 %) cases of adrenal enlargement were newly identified, none increased beyond 40 mm on follow-up. Adrenal enlargement was a significant predictor of a subsequent diagnosis of lung cancer (OR = 2.0, 95 % CI: 1.2–3.4).
Conclusion
Participants with adrenal enlargement <40 mm identified at baseline and on repeat screening could be reasonably assessed on subsequent annual screening. Adrenal enlargement increased with increasing pack-years of smoking. Adrenal enlargement was an independent predictor of a subsequent diagnosis of lung cancer.
Key Points
•
Adrenal enlargement was seen in 4 % of participants at baseline screening.
•
Age and currently smoking were significantly associated with adrenal enlargement.
•
0.04 % of participants were newly identified with adrenal enlargement.
•
Annual follow-up for adrenal enlargement <40 mm was appropriate.
•
Adrenal enlargement was an independent predictor of a diagnosis of lung cancer.
Abstract Background CT screening has been shown to increase lung cancer curability and we now assess the corresponding reduction in lung cancer mortality. Methods Lung-cancer mortality in a cohort of ...7995 smokers who underwent CT screening for lung cancer in New York State (NYS) was compared with two unscreened cohorts (CPS-II and CARET). The standardized mortality ratio (SMR) of observed to expected lung cancer deaths for NYS was jointly adjusted for age, sex, and smoking history. As more current NYS smokers might have quit as a result of the screening, thus reducing deaths from lung cancer, another analysis was restricted to those participants smoking at entry and still smoking 6 years later. Results The SMR was 64/99.8 = 0.64 ( P = 0.84 × 10−4 ) and 28/77.6 = 0.36 ( P = 0.83 × 10−10 ), showing a significant reduction in deaths from lung cancer of 36% and 64% for CPS-II and CARET, respectively. Considering participants who were smoking at entry and still smoking 6 years later, the SMR using CPS-II rates was 29/49.1 = 0.59 ( P = 0.001) and using CARET rates it was 21/57.4 = 0.37 ( P = 0.31 × 10−7 ). Conclusions CT screening significantly reduces lung-cancer mortality.
The aim of this study was to understand the effect of rounding nodule size measurements on the frequency of positive results in the setting of lung cancer screening.
Four methods for determining ...nodule size were compared, including rounding each individual length and width measurement and also rounding the overall average. These were applied to the International Early Lung Cancer Action Program database, in which we determined the frequency of a positive result by using standard size thresholds of 6.0 mm on baseline screening and 3.0 mm on repeat scanning. We also explored how rounding influences the ability to measure growth according to a predefined cutoff value of 1.5 mm as required in the Lung Computed Tomography Screening Reporting and Data System (Lung-RADS) (version 1.0).
Each method for rounding increased the frequency of positive results compared with that with no rounding. The largest increases—28.9% and 22.3% for the baseline and repeat round, respectively—occurred when rounding was used for both the individual length and width measurements and again for the final diameter. If the 1.5-mm increase in size were used for determining growth, a 4-mm nodule would need to have a volume doubling time of 130.6 days or less to demonstrate growth in 6 months (Lung-RADS category 3) whereas a 6-mm nodule would need volume doubling time of 93.2 days to demonstrate growth in 3 months (Lung-RADS category 4A). In addition, rounding can have the effect of having nodules that appear to not be growing meet the criteria for growth and vice versa.
Rounding influences the frequency of positive results and growth assessment, substantially decreasing the efficiency of screening.
High venous thromboembolism (VTE) rates have been described in critically ill patients with COVID-19. We hypothesized that specific clinical characteristics may help differentiate hypoxic COVID-19 ...patients with and without a diagnosed pulmonary embolism (PE).
We performed a retrospective observational case-control study of 158 consecutive patients hospitalized in one of four Mount Sinai Hospitals with COVID-19 between March 1 and May 8, 2020, who received a Chest CT Pulmonary Angiogram (CTA) to diagnose a PE. We analyzed demographic, clinical, laboratory, radiological, treatment characteristics, and outcomes in COVID-19 patients with and without PE.
92 patients were negative (CTA-), and 66 patients were positive for PE (CTA+). CTA + had a longer time from symptom onset to admission (7 days vs. 4 days, p = 0.05), higher admission biomarkers, notably D-dimer (6.87 vs. 1.59, p < 0.0001), troponin (0.015 vs. 0.01, p = 0.01), and peak D-dimer (9.26 vs. 3.8, p = 0.0008). Predictors of PE included time from symptom onset to admission (OR = 1.11, 95% CI 1.03-1.20, p = 0.008), and PESI score at the time of CTA (OR = 1.02, 95% CI 1.01-1.04, p = 0.008). Predictors of mortality included age (HR 1.13, 95% CI 1.04-1.22, p = 0.006), chronic anticoagulation (13.81, 95% CI 1.24-154, p = 0.03), and admission ferritin (1.001, 95% CI 1-1.001, p = 0.01).
In 158 hospitalized COVID-19 patients with respiratory failure evaluated for suspected PE, 40.8% patients had a positive CTA. We identified clinical predictors of PE and mortality from PE, which may help with early identification and reduction of PE-related mortality in patients with COVID-19.