Bridging the Gap between Structured and Free-form Radiology Reporting: A Case-study on Coronary CT Angiography

2022 ◽  
Vol 3 (1) ◽  
pp. 1-20
Author(s):  
Amara Tariq ◽  
Marly Van Assen ◽  
Carlo N. De Cecco ◽  
Imon Banerjee
Keyword(s):  
Coronary Artery ◽  
Language Processing ◽  
Free Form ◽  
Free Text ◽  
Content Type ◽  
Report Writing ◽  
Coronary Ct ◽  
Radiology Reports ◽  
Structured Reports ◽  
Artery Disease

Free-form radiology reports associated with coronary computed tomography angiography (CCTA) include nuanced and complicated linguistics to report cardiovascular disease. Standardization and interpretation of such reports is crucial for clinical use of CCTA. Coronary Artery Disease Reporting and Data System (CAD-RADS) has been proposed to achieve such standardization by implementing a strict template-based report writing and assignment of a score between 0 and 5 indicating the severity of coronary artery lesions. Even after its introduction, free-form unstructured report writing remains popular among radiologists. In this work, we present our attempts at bridging the gap between structured and unstructured reporting by natural language processing. We present machine learning models that while being trained only on structured reports, can predict CAD-RADS scores by analysis of free-text of unstructured radiology reports. The best model achieves 98% accuracy on structured reports and 92% 1-margin accuracy (difference of \le 1 in the predicted and the actual scores) for free-form unstructured reports. Our model also performs well under very difficult circumstances including nuanced and widely varying terminology used for reporting cardiovascular functions and diseases, scarcity of labeled data for training our model, and uneven class label distribution.

scholarly journals Impact of Coronary CT angiography in combination with CAD-RADS on the management of coronary artery disease patients

2021 ◽  
Vol 22 (Supplement_1) ◽  
Author(s):  
F Andre ◽  
S Seitz ◽  
P Fortner ◽  
R Sokiranski ◽  
F Gueckel ◽  
...  
Keyword(s):  
Coronary Artery Disease ◽  
Coronary Artery ◽  
Risk Stratification ◽  
Coronary Ct Angiography ◽  
Patient Management ◽  
Male Gender ◽  
Coronary Ct ◽  
History Of ◽  
Artery Disease ◽  
The Impact

Abstract Funding Acknowledgements Type of funding sources: Private company. Main funding source(s): Siemens Healthineers Introduction Coronary CT angiography (CCTA) plays an increasing role in the detection and risk stratification of patients with coronary artery disease (CAD). The Coronary Artery Disease – Reporting and Data System (CAD-RADS) allows for standardized classification of CCTA results and, thus, may improve patient management. Purpose Aim of this study was to assess the impact of CCTA in combination with CAD-RADS on patient management and to identify the impact of cardiovascular risk factors (CVRF) on CAD severity. Methods CCTA was performed on a third-generation dual-source CT scanner in patients, who were referred to a radiology centre by their attending physicians. In a total of 4801 patients, CVRF were derived from medical reports and anamnesis. Results The study population consisted of 4770 patients (62.0 (54.0-69.0) years, 2841 males) with CAD (CAD-RADS 1-5), while 31 patients showed no CAD and were excluded from further analyses. Age, male gender and the number of CVRF were associated with more severe CAD stages (all p < 0.001). 3040 patients (63.7 %) showed minimal or mild CAD requiring optimization of CVRF i.e. medical therapy but no further assessment at his time. A group of 266 patients (5.6 %) had a severe CAD defined as CAD-RADS 4B/5. In the multivariate regression analysis, age, male gender, history of smoking, diabetes mellitus and hyperlipidaemia were significant predictors for severe CAD, whereas arterial hypertension and family history of CAD did not reach significance. Of note, a subgroup of 28 patients (10.5 %) with a severe CAD (68.5 (65.5-70.0) years, 26 males, both p = n.s.) had no CVRF. Conclusions CCTA in combination with the CAD-RADS allowed for effective risk stratification of CAD patients. The majority of the patients showed non-obstructive CAD and, thus, could be treated conservatively without the need for further CAD assessment. CVRF out of arterial hypertension and family history had an impact on CAD severity reflected in higher CAD-RADs gradings. Of note, a relevant fraction of patients with CAD did not have any CVRF and, thus, may not be covered by risk stratification models. CAD-RADS n Age (years) Males (%) 1 1453 56.0 (50.0-62.0) 623 (42.9 %) 2 1587 62.0 (55.0-69.0) 918 (57.8 %) 3 1067 66.0 (59.0-71.0) 749 (70.2 %) 4A 397 66.0 (59.0-72.0) 317 (79.8 %) 4B 162 67.0 (61.0-74.0) 139 (85.8 %) 5 104 66.0 (58.5.0-77.0) 95 (91.3 %)

FRI0250 Coronary CT-Angiography Reveals the Extent of Coronary Artery Disease in Takayasu Arteritis

2015 ◽  
Vol 74 (Suppl 2) ◽  
pp. 515.1-515 ◽  
Author(s):  
E. Tombetti ◽  
R. Khamis ◽  
D. Gopalan ◽  
A. Kiprianos ◽  
B. Ariff ◽  
...  
Keyword(s):  
Coronary Artery Disease ◽  
Coronary Artery ◽  
Ct Angiography ◽  
Takayasu Arteritis ◽  
Coronary Ct Angiography ◽  
Coronary Ct ◽  
Artery Disease

scholarly journals P360 Site specific phenotype of atherosclerotic lesions according to plaque location within the coronary tree, a CCTA based study

2020 ◽  
Vol 21 (Supplement_1) ◽  
Author(s):  
D Opincariu ◽  
N Rat ◽  
A Mester ◽  
R Hodas ◽  
D Cernica ◽  
...  
Keyword(s):  
Coronary Artery ◽  
Stable Coronary Artery Disease ◽  
Coronary Plaque ◽  
The European Union ◽  
Plaque Composition ◽  
Cross Sectional ◽  
Coronary Ct ◽  
Atherosclerotic Lesions ◽  
The Government ◽  
Artery Disease

Abstract Funding Acknowledgements Research grant PlaqueImage, contract number 26/01.09.2016, SMIS code 103544, Project funded by the European Union and the Government of Romania Background The coronary CT angiography (CCTA)-based differences in composition, morphology and vulnerability of coronary plaques (CPs), according to their location within the coronary tree, have not been investigated so far. Purpose We sought to perform a comparative analysis between plaques located at different levels within the coronary tree, to identify the differences in plaque composition, morphology, and vulnerability between the three major coronary branches. Methods We conducted a cross-sectional, observational study on 75 patients with stable coronary artery disease who underwent CCTA for assessment of coronary lesions that exhibited at least one vulnerable plaque (VP) in the coronary tree. After image acquisition, coronary plaque analysis was performed with the use of the Syngo.via Frontier (Siemens) software. Plaque analysis also included evaluation of presence of VM: low attenuation plaque – LAP; napkin ring sign – NRS; spotty calcifications – SC; positive remodeling – PR. VP were defined as lesions that presented at least 1 vulnerability marker (VM). In total, 90 coronary VPs located at the level of the left anterior descending (LAD; n = 30), circumflex (CXA; n = 30) and right coronary artery respectively (RCA; n = 30) were identified and analyzed. Results Lesions located in the RCA presented a significantly higher length (LAD - 18.67± 5.49 vs. CXA - 15.48 ±3.73 vs. RCA - 20.47 ± 5.97 mm, p = 0.001), a higher degree of stenosis (LAD - 57.77 ± 8.62 vs. CXA - 54.50 ± 11.25 vs. RCA - 59.63 ± 10.42 mm, p = 0.022), and were more voluminous (LAD - 187.9 ± 86.03 vs. CXA - 146.9 ± 102.4 vs. RCA - 248.1 ± 11.4 mm3, p = 0.0007) compared to those located in the LAD and CXA, but no difference was observed regarding the remodeling (p = 0.180) or eccentricity indexes (p = 0.423). Plaque composition was also significantly different according to plaque location: calcified volume (LAD - 44.07 ± 63.90 vs. CXA - 12.40 ± 19.65 vs. RCA - 33.69 ± 34.38 mm3, p = 0.002), non-calcified volume (LAD - 143.8 ± 76.02 vs. CXA - 134.5 ± 102.2 vs. RCA - 214.4 ± 99.67 mm3, p = 0.002), lipid rich volume (LAD - 14.95 ± 22.69 vs. CXA - 6.44 ± 13.42 vs. RCA -16.07 ± 15.74 mm3, p = 0.0005), fibrotic volume (LAD - 128.9 ± 66.10 vs. CXA - 128.1 ± 91.56 vs. RCA - 198.3 ± 92.34 mm3, p = 0.003). The highest number of VM per plaque was present in the LAD (LAD - 2.2 ± 0.8 vs. CXA - 1.6 ± 0.7 vs. RCA - 1.8 ± 0.6, p = 0.01), as well as highest rate of VPs (LAD – 80%, CXA – 46.6%, RCA – 70%, p = 0.01). No difference was registered between coronary arteries on the presence of SCs (p = 0.670), NRS (p = 0.455), PR (p = 0.833), but LAPs were more frequently located in the LAD (p = 0.0009). Conclusions Coronary plaques located in the RCA were more voluminous and exhibited a higher volume of lipid rich and non-calcified atheroma. However, compared to the RCA and CXA, the left anterior descending artery presented CPs with a more expressed degree of vulnerability, a higher number of vulnerability markers per plaque, and a higher incidence of LAP.

scholarly journals Comparison of coronary CT angiography with conventional coronary angiography in the diagnosis of coronary artery disease

2014 ◽  
Vol 40 (1) ◽  
pp. 31-35
Author(s):  
N Mannan ◽  
MA Basher ◽  
J Mohammad ◽  
MU Jahan ◽  
NAM Momenuzzaman ◽  
...  
Keyword(s):  
Coronary Artery ◽  
Coronary Angiography ◽  
High Sensitivity ◽  
Therapeutic Interventions ◽  
Conventional Coronary Angiography ◽  
Promising Alternative ◽  
Non Invasive ◽  
Coronary Ct ◽  
Coronary Imaging ◽  
Artery Disease

Noninvasive CT coronary angiography is a promising coronary imaging technique. In spite of the unprecedented temporal and spatial resolution and the inability to perform therapeutic interventions in the same session multi-detector computed tomography (MDCT) has been considering a promising alternative, non invasive tool for coronary artery imaging due to its high sensitivity and specificity for the detection of significant coronary artery stenosis. To evaluate the diagnostic accuracy of 64-slice MDCT for assessing haemodynamically significant stenoses of the coronary arteries in comparison with the conventional standard cardiac angiography. Fifty patients scheduled for conventional coronary angiography at the department of Radiology and Imaging, United Hospital, Dhaka were enrolled between July 2007 and June 2008. All patients underwent both conventional and MDCT angiography within mean 10.70 days. Overall sensitivity of 64-slice MDCT for the detection of stenosis ?50%, stenosis >50%, and stenosis >75% was 90.0%, 83.8%, and 80.7%, respectively, and specificity was 96.5%, 98.4%, and 98.3% respectively and accuracy was 96.0 %, 96.5%, and 96.6% respectively. Contrast-enhanced 64-slice MDCT allows the identification of coronary stenosis with excellent accuracy. Measurements of stenosis derived by MDCT correlated well with conventional angiogram. A major limitation is the insufficient ability of CT to exactly quantify the degree of stenosis. DOI: http://dx.doi.org/10.3329/bmrcb.v40i1.20334 Bangladesh Med Res Counc Bull 2014; 40: 31-35

Sign in / Sign up

Export Citation Format

Share Document