scholarly journals Myocardial tissue phenotyping by radiomic features of native T1 maps and machine learning enhances disease detection and classification

2021 ◽  
Vol 42 (Supplement_1) ◽  
Author(s):  
A Antonopoulos ◽  
M Boutsikou ◽  
S Simantiris ◽  
A Angelopoulos ◽  
G Lazaros ◽  
...  
Keyword(s):  
Machine Learning ◽  
Cardiac Disease ◽  
T1 Mapping ◽  
Disease Type ◽  
Disease Classification ◽  
Disease Detection ◽  
Amyloid A ◽  
Myocardial Disease ◽  
Native T1 ◽  
Myocardial T1

Abstract Background Myocardial T1 mapping by cardiac magnetic resonance (CMR) is a useful technique to detect diffuse myocardial fibrosis, but a major limitation of T1 mapping is the significant overlap in native T1 values between health and disease. Purpose We explored whether radiomic features from T1 maps could enhance the diagnostic value of T1 mapping in distinguishing health from disease and classifying cardiac disease phenotypes. Methods In a total of 149 patients (n=30 with no evidence of heart disease, n=30 with LVH of various etiologies, n=61 with hypertrophic cardiomyopathy (HCM) and n=28 with cardiac amyloidosis) undergoing a CMR scan for various indications were included in this study. In addition to measuring native myocardial T1 values from T1 maps, we extracted a total of 843 radiomic features of myocardial texture and explored their value in disease classification. Results We first demonstrated that T1 mapping images are a rich source of extractable, quantifiable data. The first three principal components of the T1 radiomics were significantly and distinctively correlated with cardiac disease type. Unsupervised hierarchical clustering of the population by myocardial T1 radiomics was significantly associated with myocardial disease type (chi2=55.98, p<0.0001). After machine learning for feature selection, training with internal validation and external testing, a model of T1 radiomics had good diagnostic performance (AUC 0.753) for multinomial classification of disease phenotype (normal vs. LVH vs. HCM vs. amyloid). A subset of seven radiomic features outperformed mean native T1 values for classification between myocardial health vs. disease and HCM phenocopies (for normal: T1 AUC 0.549 vs. radiomics AUC 0.888, for LVH: T1 AUC 0.645 vs. radiomics AUC 0.790, for HCM T1 AUC 0.541 vs. radiomics AUC 0.638 and for amyloid T1 AUC 0.769 vs. radiomics AUC 0.840). Conclusions We have shown that specific imaging patterns in myocardial native T1 maps are linked to features of cardiac disease and we have provided for the first-time evidence that radiomic phenotyping can be used to enhance the diagnostic yield of native T1 mapping for myocardial disease detection and classification. FUNDunding Acknowledgement Type of funding sources: None.

scholarly journals Machine learning of native T1 mapping radiomics for classification of hypertrophic cardiomyopathy phenotypes

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Alexios S. Antonopoulos ◽  
Maria Boutsikou ◽  
Spyridon Simantiris ◽  
Andreas Angelopoulos ◽  
George Lazaros ◽  
...  
Keyword(s):  
Machine Learning ◽  
Feature Selection ◽  
Cardiac Disease ◽  
T1 Mapping ◽  
Diagnostic Value ◽  
Disease Type ◽  
Disease Classification ◽  
Cardiac Amyloid ◽  
Native T1

AbstractWe explored whether radiomic features from T1 maps by cardiac magnetic resonance (CMR) could enhance the diagnostic value of T1 mapping in distinguishing health from disease and classifying cardiac disease phenotypes. A total of 149 patients (n = 30 with no heart disease, n = 30 with LVH, n = 61 with hypertrophic cardiomyopathy (HCM) and n = 28 with cardiac amyloidosis) undergoing a CMR scan were included in this study. We extracted a total of 850 radiomic features and explored their value in disease classification. We applied principal component analysis and unsupervised clustering in exploratory analysis, and then machine learning for feature selection of the best radiomic features that maximized the diagnostic value for cardiac disease classification. The first three principal components of the T1 radiomics were distinctively correlated with cardiac disease type. Unsupervised hierarchical clustering of the population by myocardial T1 radiomics was significantly associated with myocardial disease type (chi2 = 55.98, p < 0.0001). After feature selection, internal validation and external testing, a model of T1 radiomics had good diagnostic performance (AUC 0.753) for multinomial classification of disease phenotype (normal vs. LVH vs. HCM vs. cardiac amyloid). A subset of six radiomic features outperformed mean native T1 values for classification between myocardial health vs. disease and HCM phenocopies (AUC of T1 vs. radiomics model, for normal: 0.549 vs. 0.888; for LVH: 0.645 vs. 0.790; for HCM 0.541 vs. 0.638; and for cardiac amyloid 0.769 vs. 0.840). We show that myocardial texture assessed by native T1 maps is linked to features of cardiac disease. Myocardial radiomic phenotyping could enhance the diagnostic yield of T1 mapping for myocardial disease detection and classification.

scholarly journals 395 Diagnostic role of native T1 mapping compared to conventional magnetic resonance techniques in cardiac diseases: a real-life study

2021 ◽  
Vol 23 (Supplement_G) ◽  
Author(s):  
Roberto Licordari ◽  
Chrysanthos Grigoratos ◽  
Giancarlo Todiere ◽  
Andrea Barison ◽  
Gianluca Di Bella ◽  
...  
Keyword(s):  
Myocardial Infarction ◽  
Magnetic Resonance ◽  
Cardiac Disease ◽  
T1 Mapping ◽  
Regional Distribution ◽  
Real Life ◽  
Myocardial Damage ◽  
Post Contrast ◽  
Myocardial Disease ◽  
Native T1

Abstract Aims T1 mapping is a validated technique in cardiac magnetic resonance (CMR), however in real-life clinical practice its effectiveness to diagnose myocardial disease is still unclear. To compare native T1 mapping to conventional late gadolinium enhancement (LGE) and T2-STIR techniques for the evaluation of a cohort of consecutive patients undergoing CMR for the suspicion of myocardial disease. Methods and results CMR was performed in 323 patients, 206 males (64%), mean age 54 ± 8 years, and in 27 age- and sex-matched healthy controls. LGE, T2-STIR, and pre- and post-contrast T1 mapping were acquired as suggested by the SCMR position paper. The CMR findings of global and regional T1 mapping were compared to the respective results of LGE and T2-STIR techniques. The main baseline indications for CMR were: suspicion of ARVC in 20%; non-ischaemic DCM in 19%; HCM in 16%; chest pain without obstructive coronary artery in 14% of patients (suspicion of MINOCA, Tako-tsubo or myocarditis); other indications (amyloidosis, scleroderma, previous myocardial infarction, pericarditis, LV non-compaction) in the remaining of cases. At T2-STIR images myocardial hyperintensity suggesting oedema was found in 41 patients (27%). LGE images were positive in 206 patients (64%). Native T1 mapping was abnormal in 171 (49%). In 206 patients (64%) a matching between LGE and native T1 was found (both positive in 132 and negative in 74). T1 was also abnormal in 32 out of 41 (78%) with oedema at T2-STIR. Overall, LGE and/or T2-STIR were abnormal in 209 patients, whereas native T1 in 154(52%). Conventional techniques and T1 mapping were concordant in 208 patients (64%). Conventional techniques were abnormal in 76 (24%) of patients with negative T1 mapping. Finally, in 39 patients T1 mapping was positive despite negative conventional techniques (12%). Among these latter 39 patients, only in 18 T2-STIR were acquired based on clinical decision. Then, the percentage of cases where T1 mapping could have an additive role would range between 6% and 12%. T1 mapping was particularly able in conditions with diffuse myocardial damage as cardiac amyloidosis, scleroderma and fabry disease (additive role in 42%). On contrast, T1 mapping was less effective in cardiac disease with regional distribution of myocardial damage as myocardial infarction, HCM, myocarditis (additive role in 1%). Conclusions T1 mapping may give additive information in 6–12% of patient but is less effective cardiac disease presenting with regional or segmental distribution of myocardial damage. Results of the present study suggest that conventional LGE/T2-STIR and T1 mapping are complementary techniques and should be used together in every CMR examination.

Native T1 Values and Cardiac Involvement in Patients with Thalassemia Major

Blood ◽  
2020 ◽  
Vol 136 (Supplement 1) ◽  
pp. 24-25
Author(s):  
Alessia Pepe ◽  
Nicola Martini ◽  
Antonio De Luca ◽  
Vincenzo Positano ◽  
Laura Pistoia ◽  
...  
Keyword(s):  
Iron Overload ◽  
T1 Mapping ◽  
Thalassemia Major ◽  
Clinical Impact ◽  
Cardiac Complications ◽  
Myocardial Iron ◽  
Native T1 ◽  
Molli Sequence ◽  
History Of ◽  
Myocardial T1

Background.The T2* cardiovascular magnetic resonance (CMR) is the gold standard for the non invasive detection of myocardial iron overload (MIO). The native myocardial T1 mapping has been proposed as a complementary tool, thanks to its higher sensitivity in presence of small amounts of iron, but no data are available in literature about its clinical impact. Objective:To explore the clinical impact of T1 mapping for detecting cardiac complications in thalassemia major (TM). Methods.We considered 146 TM patients (87 females, 38.7±11.1 years) consecutively enrolled in the Extension-Myocardial Iron Overload in Thalassemia Network. Three parallel short-axis slices of the left ventricle (LV) were acquired with the Modified Look-Locker Inversion recovery (MOLLI) sequence. The native T1 values in all 16 myocardial segments were obtained and the global value was the mean. Results.Twenty-one patients had an history of cardiac complications: 11 heart failure, 8 arrhythmias (7 supraventricular and 1 ventricular), and 2 pulmonary hyperthension. Patients with cardiac complications had significantly lower global heart T1 values (879.3±121.9 ms vs 963.2±98.5 ms; P&lt;0.0001) (Figure) but comparable T2* values (33.32±11.66 ms vs 37.17±9.15 ms; P=0.116). Cardiac complications were more frequent in the group of patients with reduced global heart T1 value (&lt;928 ms for males and &lt;989 ms for females) compared to the group with normal global heart T1 value (71.4% vs 39.5%; P=0.009). Odds ratio (OR) for cardiac complications was 3.8 (95%CI=1.3-10.9) for patients with reduced global heart T1 value versus patients with normal global heart T1 value. Conclusion:We found out a significant association between decreased native global heart T1 values and a history of cardiac complications, suggesting that an early detection of myocardial iron burden by native T1 can support the clinicians in modifing chelation therapy earlier. Figure Disclosures Pepe: ApoPharma Inc.:Other: no profit support;Bayer:Other: no profit support;Chiesi Farmaceutici S.p.A.:Other: no profit support and speakers' honoraria.Pistoia:Chiesi Farmaceutici S.p.A.:Other: speakers' honoraria.Meloni:Chiesi Farmaceutici S.p.A.:Other: speakers' honoraria.

scholarly journals Automated myocardial segmentation in native t1-mapping cardiovascular magnetic resonance images based on machine learning: a validation study in the UK biobank"s covid-19 subset

2021 ◽  
Vol 22 (Supplement_2) ◽  
Author(s):  
E Rauseo ◽  
L Lockhart ◽  
JM Paiva ◽  
K Fung ◽  
MY Khanji ◽  
...  
Keyword(s):  
Machine Learning ◽  
Quality Control ◽  
Cardiovascular Magnetic Resonance ◽  
Magnetic Resonance ◽  
T1 Mapping ◽  
Automated Analysis ◽  
Uk Biobank ◽  
Native T1 ◽  
The Mean ◽  
Good Agreement

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Innovate UK Background  Regional assessment of septal native T1 values with cardiovascular magnetic resonance (CMR) is used to characterise diffuse myocardial diseases. Previous studies suggest its potential role in detecting early pathological alterations, which may help identify high-risk subjects at early disease stages. Automated analysis of myocardial native T1 images may enable faster CMR analysis and reduce inter-observer variability of manual analysis. However, the technical performance of such methodologies has not been previously reported. Purpose  We tested, in a subset of UK Biobank participants, the degree of agreement between CMR septal myocardial T1 values obtained from our machine learning (ML) algorithm and septal native T1 values computed from manual segmentations. Methods  We analysed the first 292 participants who were tested for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and had CMR imaging (1.5 Tesla, Siemens MAGNETOM Aera). T1 mapping was performed in a single mid-ventricular short axis (SAX) slice using ShMOLLI (WIP780B) sequences. Three experienced CMR readers independently measured native T1 values by manually placing a single region of interest (ROI) covering half of the anteroseptal and half of the inferoseptal wall using cvi42 post-processing software (version 5.11). A mean T1 value for each participant was then calculated. A ML algorithm developed by Circle Cardiovascular Imaging Inc. was then applied to the same images to derive the myocardium T1 values automatically. The algorithm was previously trained to segment myocardium from SAX T1 and non-T1 mapping images on two external CMR datasets. We compared the mean septal ROI T1 values to the mean myocardium T1 values predicted by the ML algorithm. Results  Two studies were excluded after quality control. The ML-derived and the manually calculated mean T1 values were significantly correlated (r = 0.82, p &lt; 0.001). The Bland-Altman analysis between the two methods showed a mean bias of 3.64 ms, with 95% limits of agreement of −38.88 to 53.46 ms, indicating good agreement (figure 1). Conclusions  We demonstrated strong correlation and good agreement between native T1 values obtained from our automated analysis method and manual T1 septal analysis in a subset of UK Biobank participants. This algorithm may represent a valuable tool for clinicians allowing for fast and potentially less operator-dependent myocardial tissue characterisation. However, validation of more extensive datasets and quality control processes are needed.

scholarly journals Native T1 mapping and extracellular volume fraction for differentiation of myocardial diseases from normal CMR controls in routine clinical practice

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Rawiwan Thongsongsang ◽  
Thammarak Songsangjinda ◽  
Prajak Tanapibunpon ◽  
Rungroj Krittayaphong
Keyword(s):  
T1 Mapping ◽  
Volume Fraction ◽  
Diagnostic Yield ◽  
Interventricular Septum ◽  
Extracellular Volume ◽  
Extracellular Volume Fraction ◽  
Left Ventricular ◽  
Control Group ◽  
Myocardial Disease ◽  
Native T1

Abstract Background This study aimed to determine native T1 and extracellular volume fraction (ECV) in distinct types of myocardial disease, including amyloidosis, dilated cardiomyopathy (DCM), hypertrophic cardiomyopathy (HCM), myocarditis and coronary artery disease (CAD), compared to controls. Methods We retrospectively enrolled patients with distinct types of myocardial disease, CAD patients, and control group (no known heart disease and negative CMR study) who underwent 3.0 Tesla CMR with routine T1 mapping. The region of interest (ROI) was drawn in the myocardium of the mid left ventricular (LV) short axis slice and at the interventricular septum of mid LV slice. ECV was calculated by actual hematocrit (Hct) and synthetic Hct. T1 mapping and ECV was compared between myocardial disease and controls, and between CAD and controls. Diagnostic yield and cut-off values were assessed. Results A total of 1188 patients were enrolled. The average T1 values in the control group were 1304 ± 42 ms at septum, and 1294 ± 37 ms at mid LV slice. The average T1 values in patients with myocardial disease and CAD were significantly higher than in controls (1441 ± 72, 1349 ± 59, 1345 ± 59, 1355 ± 56, and 1328 ± 54 ms for septum of amyloidosis, DCM, HCM, myocarditis, and CAD). Native T1 of the mid LV level and ECV at septum and mid LV with actual and synthetic Hct of patients with myocardial disease or CAD were significantly higher than in controls. Conclusions Although native T1 and ECV of patients with cardiomyopathy and CAD were significantly higher than controls, the values overlapped. The greatest clinical utilization was found for the amyloidosis group.

Usefulness of acute phase proteins in differentiating between feline infectious peritonitis and other diseases in cats with body cavity effusions

2016 ◽  
Vol 19 (8) ◽  
pp. 809-816 ◽  
Author(s):  
Katarina Hazuchova ◽  
Susanne Held ◽  
Reto Neiger
Keyword(s):  
Machine Learning ◽  
Game Theory ◽  
Acute Phase ◽  
Cardiac Disease ◽  
Serum Amyloid A ◽  
Acute Phase Proteins ◽  
Body Cavity ◽  
Feline Infectious Peritonitis ◽  
Acid Glycoprotein ◽  
Amyloid A

Objectives The aim of this study was to evaluate the measurement of acute phase proteins (APPs) as a diagnostic tool to differentiate between feline infectious peritonitis (FIP) and other diseases in cats with body cavity effusions. Methods Cats with pleural, abdominal or pericardial effusion were prospectively enrolled. Cats were classified as having or not having FIP based on immunohistochemistry (if available) or a sophisticated statistical method using machine learning methodology with concepts from game theory. Cats without FIP were further subdivided into three subgroups: cardiac disease, neoplasia and other diseases. Serum amyloid A (SAA), haptoglobin (Hp) and α1-acid glycoprotein (AGP) were measured in serum and effusion, using assays previously validated in cats. Results Serum and effusion samples were available for the measurement of APPs from 88 and 67 cats, respectively. Concentrations of the APPs in serum and effusion were significantly different in cats with and without FIP ( P <0.001 for all three APPs). The best APP to distinguish between cats with and without FIP was AGP in the effusion; a cut-off value of 1550 µg/ml had a sensitivity and specificity of 93% each for diagnosing FIP. Conclusions and relevance AGP, particularly if measured in effusion, was found to be useful in differentiating between FIP and other diseases, while SAA and Hp were not. The concentration of all three APPs in some diseases (eg, septic processes, disseminated neoplasia) was as high as in cats with FIP; therefore, none of these can be recommended as a single diagnostic test for FIP.

Application of Machine Learning Techniques in Crop Disease Classification: A Comprehensive Review

2021 ◽  
Vol 9 (1) ◽  
pp. 89-93
Author(s):  
Khwairakpam Amitab ◽  
◽  
Lal Hmingliana ◽  
Amitabha Nath ◽  
◽  
...  
Keyword(s):  
Machine Learning ◽  
Classification Systems ◽  
Disease Classification ◽  
Machine Learning Techniques ◽  
Future Research ◽  
Disease Detection ◽  
Comprehensive Review ◽  
Road Map ◽  
Learning Techniques ◽  
Crop Disease

Crop diseases are the main threat to agricultural products. Fast, accurate, and automatic detection of diseases can help to overcome this problem. Literature suggests, machine learning techniques are capable of achieving these goals in near real-time. This article presents a comprehensive review of machine learning techniques for crop disease detection and classification. Existing plant disease classification systems are explored and commonly used processing steps are investigated. Analysis of machine learning techniques, accuracy factor, and current state-of-the-art in this domain have been reviewed and presented through this manuscript. The survey resulted in the identification of the strengths and limitations of existing techniques and provides a road map for future research works. These would help researchers to understand and pursue machine learning applications in the field of disease detection and classification

Sign in / Sign up

Export Citation Format

Share Document