Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning.
Ontology highlight
ABSTRACT:
SUBMITTER: Golla AK
PROVIDER: S-EPMC8621263 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundAbdominal aortic atherosclerotic plaque burden may have clinical significance but manual measurement is time-consuming and impractical.PurposeTo perform external validation on an automated atherosclerotic plaque detector for noncontrast and postcontrast abdominal CT.Materials and methodsThe training data consisted of 114 noncontrast CT scans and 23 postcontrast CT urography scans. The testing data set consisted of 922 CT colonography (CTC) scans, and 1207 paired noncontrast and postcontrast CT scans from renal donors from a second institution. Reference standard data included manual plaque segmentations in the 137 training scans and manual plaque burden measurements in the 922 CTC scans. The total Agatston score and group (0-3) was determined using fully-automated deep learning software. Performance was assessed by measures of agreement, linear regression, and paired evaluations.ResultsOn CTC scans, automated Agatston scoring correlated highly with manual assessment (R2 = 0.94). On paired renal donor CT scans, automated Agatston scoring on postcontrast CT correlated highly with noncontrast CT (R2 = 0.95). When plaque burden was expressed as a group score, there was excellent agreement for both the CTC (weighted kappa 0.80 ± 0.01 [95% confidence interval: 0.78-0.83]) and renal donor (0.83 ± 0.02 [0.79-0.86]) assessments.ConclusionFully automated detection, segmentation, and scoring of abdominal aortic atherosclerotic plaques on both pre- and post-contrast CT was validated and may have application for population-based studies.
| S-EPMC7969468 | biostudies-literature
Project description:BackgroundRuptured abdominal aortic aneurysm (AAA) is responsible for 1-2% of all male deaths over the age of 65 years. Early detection of AAA and elective surgery can reduce the mortality risk associated with AAA. However, many patients will not be diagnosed with AAA and have therefore an increased death risk due to the untreated AAA. It has been suggested that population screening for AAA in elderly males is effective and cost-effective. The purpose of this study was to perform a systematic review of published cost-effectiveness analyses of screening elderly men for AAA.MethodsWe performed a systematic search for economic evaluations in NHSEED, EconLit, Medline, Cochrane, Embase, Cinahl and two Scandinavian HTA data bases (DACEHTA and SBU). All identified studies were read in full and each study was systematically assessed according to international guidelines for critical assessment of economic evaluations in health care.ResultsThe search identified 16 cost-effectiveness studies. Most studies considered only short term cost consequences. The studies seemed to employ a number of "optimistic" assumptions in favour of AAA screening, and included only few sensitivity analyses that assessed less optimistic assumptions.ConclusionFurther analyses of cost-effectiveness of AAA screening are recommended.
| S-EPMC2607250 | biostudies-literature
Project description:BackgroundPopulation-wide ultrasound screening programmes for abdominal aortic aneurysm (AAA) for men have already been established in some countries. Women account for one third of aneurysm-related mortality and are four times more likely to experience an AAA rupture than men. Whole-population screening for AAA in women is unlikely to be clinically or economically effective. The aim of this study was to determine the outcomes of a targeted AAA screening programme for women at high risk of AAA.MethodWomen aged 65-74 years deemed at high risk of having an AAA (current smokers, ex-smokers, or with a history of coronary artery disease) were invited to attend ultrasound screening (July 2016 to March 2019) for AAA in the Female Aneurysm screening STudy (FAST). Primary outcomes were attendance for screening and prevalence of AAA. Biometric data, medical history, quality of life (QoL) and aortic diameter on ultrasound imaging were recorded prospectively.ResultsSome 6037 women were invited and 5200 attended screening (86.7 per cent). Fifteen AAAs larger than 29 mm were detected (prevalence 0.29 (95 per cent c.i. 0.18 to 0.48) per cent). Current smokers had the highest prevalence (0.83 (95 per cent c.i. 0.34 to 1.89) per cent) but lowest attendance (75.2 per cent). Three AAAs greater than 5.5 cm were identified and referred for consideration of surgical repair; one woman underwent repair. There was a significant reduction in patient-reported QoL scores following screening.ConclusionA low prevalence of AAA was detected in high-risk women, with lowest screening uptake in those at highest risk. Screening for AAA in high-risk women may not be beneficial.
| S-EPMC8545265 | biostudies-literature
Project description:BackgroundScreening for abdominal aortic aneurysm (AAA) is known to reduce AAA-related mortality; however, the psychological impact of population AAA screening is unclear. The aim was to assess the impact of AAA diagnosis on quality of life (QoL) using data from an established AAA screening programme.MethodsMental and physical QoL scores for men diagnosed with AAA through participation in the English and Welsh AAA screening programmes were compared with no-AAA controls. Participants were identified through the United Kingdom Aneurysm Growth Study (UKAGS), a nationwide prospective cohort study of men with an AAA of less than 55 mm diagnosed through voluntary participation in screening. The UKAGS participants completed QoL questionnaires at the time of screening and annually thereafter.ResultsA transient reduction in mental QoL scores was observed following the diagnosis of AAA, returning to baseline levels after 12 months. Physical QoL remained consistently lower in the AAA cohort. Participants thought about their AAA and the AAA growth progressively less 12 months after the initial screening diagnosis. AAA growth rate had no influence over QoL parameters.DiscussionThis study suggests that screening for AAA does reduce mental QoL; however, this effect is transient (less than 12 months). Men diagnosed with AAA have a consistently worse physical QoL compared with controls.
| S-EPMC5817237 | biostudies-literature
Project description:The long-term effects of abdominal aortic aneurysm (AAA) screening were investigated in extended follow-up from the UK Multicentre Aneurysm Screening Study (MASS) randomized trial.A population-based sample of men aged 65-74 years were randomized individually to invitation to ultrasound screening (invited group) or to a control group not offered screening. Patients with an AAA (3·0 cm or larger) detected at screening underwent surveillance and were offered surgery after predefined criteria had been met. Cause-specific mortality data were analysed using Cox regression.Some 67 770 men were enrolled in the study. Over 13 years, there were 224 AAA-related deaths in the invited group and 381 in the control group, a 42 (95 per cent confidence interval 31 to 51) per cent reduction. There was no evidence of effect on other causes of death, but there was an overall reduction in all-cause mortality of 3 (1 to 5) per cent. The degree of benefit seen in earlier years of follow-up was slightly diminished by the occurrence of AAA ruptures in those with an aorta originally screened normal. About half of these ruptures had a baseline aortic diameter in the range 2·5-2·9 cm. It was estimated that 216 men need to be invited to screening to save one death over the next 13 years.Screening resulted in a reduction in all-cause mortality, and the benefit in AAA-related mortality continued to accumulate throughout follow-up.ISRCTN37381646 (http://www.controlled-trials.com).
| S-EPMC3569614 | biostudies-literature
Project description:Modifiable lifestyle-related factors associated with risk of abdominal aortic aneurysm (AAA) are rarely investigated with a prospective design. We aimed to study possible associations among such factors and comorbidities with mean abdominal aortic diameter (AAD) and with risk of AAA among men screened for the disease. Self-reported lifestyle-related exposures were assessed at baseline (January 1, 1998) among 14 249 men from the population-based Cohort of Swedish Men, screened for AAA between 65 and 75 years of age (mean 13 years after baseline). Multivariable prediction of mean AAD was estimated with linear regression, and hazard ratios (HRs) of AAA (AAD ≥30 mm) with Cox proportional hazard regression. The AAA prevalence was 1.2% (n=168). Smoking, body mass index, and cardiovascular disease were associated with a larger mean AAD, whereas consumption of alcohol and diabetes mellitus were associated with a smaller mean AAD. The HR of AAA was increased among participants who were current smokers with ≥25 pack-years smoked compared with never smokers (HR 15.59, 95% CI 8.96-27.15), those with a body mass index ≥25 versus <25 (HR 1.89, 95% CI, 1.22-2.93), and those with cardiovascular disease (HR 1.77, 95% CI, 1.13-2.77), and hypercholesterolemia (HR 1.59, 95% CI 1.08-2.34). Walking or bicycling for >40 minutes/day (versus almost never) was associated with lower AAA hazard (HR 0.59, 95% CI 0.36-0.97) compared with almost never walking or bicycling. This prospective study confirms that modifiable lifestyle-related factors are associated with AAD and with AAA disease.
| S-EPMC5524061 | biostudies-literature
Project description:Abdominal aortic aneurysm (AAA) is a multifactorial disease with a strong genetic component. Since the first candidate gene studies were published 20 years ago, approximately 100 genetic association studies using single nucleotide polymorphisms (SNPs) in biologically relevant genes have been reported on AAA. These studies investigated SNPs in genes of the extracellular matrix, the cardiovascular system, the immune system, and signaling pathways. Very few studies were large enough to draw firm conclusions and very few results could be replicated in another sample set. The more recent unbiased approaches are family-based DNA linkage studies and genome-wide genetic association studies, which have the potential of identifying the genetic basis for AAA, only when appropriately powered and well-characterized large AAA cohorts are used. SNPs associated with AAA have already been identified in these large multicenter studies. One significant association was of a variant in a gene called contactin-3, which is located on chromosome 3p12.3. However, two follow-up studies could not replicate this association. Two other SNPs, which are located on chromosome 9p21 and 9q33, were replicated in other samples. The two genes with the strongest supporting evidence of contribution to the genetic risk for AAA are the CDKN2BAS gene, also known as ANRIL, which encodes an antisense ribonucleic acid that regulates expression of the cyclin-dependent kinase inhibitors CDKN2A and CDKN2B, and DAB2IP, which encodes an inhibitor of cell growth and survival. Functional studies are now needed to establish the mechanisms by which these genes contribute toward AAA pathogenesis.
| S-EPMC3058859 | biostudies-literature
Project description:We determined the feasibility of abdominal aortic aneurysm (AAA) screening program led by family physicians in public primary healthcare setting using hand-held ultrasound device. The potential study population was 11,214 men aged ≥ 60 years attended by three urban, public primary healthcare centers. Participants were recruited by randomly-selected telephone calls. Ultrasound examinations were performed by four trained family physicians with a hand-held ultrasound device (Vscan®). AAA observed were verified by confirmatory imaging using standard ultrasound or computed tomography. Cardiovascular risk factors were determined. The prevalence of AAA was computed as the sum of previously-known aneurysms, aneurysms detected by the screening program and model-based estimated undiagnosed aneurysms. We screened 1,010 men, with mean age of 71.3 (SD 6.9) years; 995 (98.5%) men had normal aortas and 15 (1.5%) had AAA on Vscan®. Eleven out of 14 AAA-cases (78.6%) had AAA on confirmatory imaging (one patient died). The total prevalence of AAA was 2.49% (95%CI 2.20 to 2.78). The median aortic diameter at diagnosis was 3.5 cm in screened patients and 4.7 cm (p<0.001) in patients in whom AAA was diagnosed incidentally. Multivariate logistic regression analysis identified coronary heart disease (OR = 4.6, 95%CI 1.3 to 15.9) as the independent factor with the highest odds ratio. A screening program led by trained family physicians using hand-held ultrasound was a feasible, safe and reliable tool for the early detection of AAA.
| S-EPMC5409053 | biostudies-literature
Project description:Our objective was to develop and evaluate the clinical feasibility of deep-learning-based synthetic contrast-enhanced computed tomography (DL-SynCCT) in patients designated for nonenhanced CT (NECT). We proposed a weakly supervised learning with the utilization of virtual non-contrast CT (VNC) for the development of DL-SynCCT. Training and internal validations were performed with 2202 pairs of retrospectively collected contrast-enhanced CT (CECT) images with the corresponding VNC images acquired from dual-energy CT. Clinical validation was performed using an external validation set including 398 patients designated for true nonenhanced CT (NECT), from multiple vendors at three institutes. Detection of lesions was performed by three radiologists with only NECT in the first session and an additionally provided DL-SynCCT in the second session. The mean peak signal-to-noise ratio (PSNR) and structural similarity index map (SSIM) of the DL-SynCCT compared to CECT were 43.25 ± 0.41 and 0.92 ± 0.01, respectively. With DL-SynCCT, the pooled sensitivity for lesion detection (72.0% to 76.4%, P < 0.001) and level of diagnostic confidence (3.0 to 3.6, P < 0.001) significantly increased. In conclusion, DL-SynCCT generated by weakly supervised learning showed significant benefit in terms of sensitivity in detecting abnormal findings when added to NECT in patients designated for nonenhanced CT scans.
| S-EPMC11291756 | biostudies-literature
Project description:Lung cancer is the leading cause of cancer related deaths worldwide. The development of orthotopic mouse models of lung cancer, which recapitulates the disease more realistically compared to the widely used subcutaneous tumor models, is expected to critically aid the development of novel therapies to battle lung cancer or related comorbidities such as cachexia. However, follow-up of tumor take, tumor growth and detection of therapeutic effects is difficult, time consuming and requires a vast number of animals in orthotopic models. Here, we describe a solution for the fully automatic segmentation and quantification of orthotopic lung tumor volume and mass in whole-body mouse computed tomography (CT) scans. The goal is to drastically enhance the efficiency of the research process by replacing time-consuming manual procedures with fast, automated ones. A deep learning algorithm was trained on 60 unique manually delineated lung tumors and evaluated by four-fold cross validation. Quantitative performance metrics demonstrated high accuracy and robustness of the deep learning algorithm for automated tumor volume analyses (mean dice similarity coefficient of 0.80), and superior processing time (69 times faster) compared to manual segmentation. Moreover, manual delineations of the tumor volume by three independent annotators was sensitive to bias in human interpretation while the algorithm was less vulnerable to bias. In addition, we showed that besides longitudinal quantification of tumor development, the deep learning algorithm can also be used in parallel with the previously published method for muscle mass quantification and to optimize the experimental design reducing the number of animals needed in preclinical studies. In conclusion, we implemented a method for fast and highly accurate tumor quantification with minimal operator involvement in data analysis. This deep learning algorithm provides a helpful tool for the noninvasive detection and analysis of tumor take, tumor growth and therapeutic effects in mouse orthotopic lung cancer models.
| S-EPMC8471805 | biostudies-literature