A Computer‐Aided Diagnostic Algorithm Improves the Accuracy of Transesophageal Echocardiography for Left Atrial Thrombi

Sun, Li; Yang, Li; Zhang, Ying Tao; Shen, Jing; Xue, Feng; Cheng, Heng-Da; Qu, Xiu Fen

doi:10.7863/ultra.33.1.83

Cited by 12 publications

(18 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…43 There were several limitations to this study that merit discus- No similar to that reported by other investigators. Interestingly, beyond a direct relationship be- The addition of D-dimer and other risk factors will help to detect these patients.…”

Section: Discussionmentioning

confidence: 61%

“…Identifying these patients is crucial, as it would alter their subsequent therapy from antiplatelet to anticoagulation, which is known to be much more effective in reducing subsequent thromboembolic events. 43 There were several limitations to this study that merit discus- No similar to that reported by other investigators. Third, though our data were essentially complete for our major values five patients with negative clot and the positive D-dimer level did not have a complete hematologic WBC profile, which could have obscured these results.…”

Section: Discussionmentioning

confidence: 61%

See 1 more Smart Citation

Useful indices of thrombogenesis in the exclusion of intra‐cardiac thrombus

et al. 2019

View full text Add to dashboard Cite

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. Abstract Background: Cardioversion in patients with atrial fibrillation (AF) can cause cardioembolic stroke, and effective clinical management is necessary to reduce morbidity and mortality. Currently, transesophageal echocardiography (TEE) is the accepted standard to diagnose cardiogenic thromboemboli; however, a negative TEE does not eliminate the possibility of left atrial thrombus. The objective of this study was to evaluate the diagnostic value of supplementing the TEE with additional noninvasive markers to ensure thrombus absence. Methods: A prospective study was conducted on 59 patients who underwent TEE for suspected intra-cardiac thrombi. The TEE indications included acute ischemic stroke (45.7%) and AF or flutter (59.3%). D-dimer level and white blood cell counts were assessed. Results: A negative D-dimer level (<200 ng/mL) excluded the presence of intra-cardiac thrombi. Groups with either negative (n = 14) or positive (n = 45) D-dimer levels had comparable clinical characteristics. Comparing positive D-dimer-level patients with thrombus (n = 7) and without thrombus (n = 33), patients with thrombus had reduced left atrial appendage (LAA) velocity (P = .0024), reduced left ventricular ejection fraction (LVEF) (P = .0263), increased neutrophil percent (P = .0261), decreased lymphocyte percent (P = .0216), and increased monocyte counts (P = .0220). The area under the receiver operating characteristic (ROC) curve for thrombus diagnostics was larger for combinations of clinical and biochemical data than for each parameter individually. Conclusions: Supplementing the gold standard TEE with the analysis of LAA velocity, noninvasive LVEF, D-dimer, and hemostatic markers provided additional useful diagnostic information. Larger studies are needed to further validate the efficacy of supplementing the TEE to better assess patients for intra-cardiac thrombi. K E Y W O R D S atrial fibrillation, cardiac thrombus, D-dimer assay, thrombosis, transesophageal echocardiography | 87 IBEBUOGU Et al.

show abstract

Section: Discussionmentioning

confidence: 61%

Section: Discussionmentioning

confidence: 61%

Useful indices of thrombogenesis in the exclusion of intra‐cardiac thrombus

et al. 2019

View full text Add to dashboard Cite

show abstract

“…42 Então, aplica-se a radiação estereotáxica, tratando a arritmia, com mínimo de dano possível ao tecido normal 45 e com tempo de tratamento em média menor que 20 minutos. 45 Essa técnica tem como população alvo pacientes com cardiomiopatia, pois é mais difícil identificar com precisão o substrato da taquiarritmia na ablação por cateter para taquicardia ventricular nesse grupo de pacientes, uma vez que esse substrato tem grande extensão não passível de ablação efetiva ou é inacessível pelas tecnologias atuais e, por isso, o procedimento com cateter está associado a altas taxas de recorrência, complicações e pior prognóstico em relação à IC e morte súbita. 46 Desta forma, para o grupo de cardiomiopatas, o fato da ablação extracorpórea apresentar-se como uma possibilidade de tratamento significa grande avanço terapêutico.…”

Section: Ablação Extracorpóreaunclassified

“…44 Similarly, another meta-analysis comparing deep learning tools with human image interpretations for any disease showed that the AI tools were equivalent, if not superior, to human interpretations. 45 However, these results must be interpreted in the context that the AI models were run on complete datasets, where the investigational testing and adverse patient outcomes had already happened. This retrospective validation is not proof of the algorithms' prospective efficacy and ultimate usability in altering patient outcomes.…”

Section: Impact and Intended Usage Domainmentioning

confidence: 99%

“…44 Da mesma forma, outra metanálise que comparou ferramentas de aprendizado profundo com interpretações de imagens humanas para qualquer doença mostrou que as ferramentas de IA eram equivalentes, se não superiores, às interpretações humanas. 45 No entanto, esses resultados devem ser interpretados no contexto de que os modelos de IA foram executados em conjuntos de dados completos, onde os testes de investigação e os resultados adversos dos pacientes já haviam ocorrido. Essa validação retrospectiva não prova a eficiência prospectiva e a usabilidade final dos algoritmos na alteração dos resultados dos pacientes.…”

Section: Impact and Intended Usage Domainunclassified

See 1 more Smart Citation

Untitled

2020

Rev Soc Cardiol Estado de São Paulo

View full text Add to dashboard Cite

show abstract

Association of Clinician Diagnostic Performance With Machine Learning–Based Decision Support Systems

et al. 2021

View full text Add to dashboard Cite

Key Points Question Is clinician diagnostic performance associated with the use of machine learning–based clinical decision support systems? Findings In this systematic review of 37 studies, no robust evidence was found to suggest an association between the use of machine learning–based clinical algorithms to support rather than replace human decision-making and improved clinician diagnostic performance. Meaning Caution should be observed when estimating the current ability of machine learning algorithms to affect patient care, and emphasis on the evaluation of the human-computer interaction is needed.

show abstract

A Computer‐Aided Diagnostic Algorithm Improves the Accuracy of Transesophageal Echocardiography for Left Atrial Thrombi

Cited by 12 publications

References 25 publications

Useful indices of thrombogenesis in the exclusion of intra‐cardiac thrombus

Useful indices of thrombogenesis in the exclusion of intra‐cardiac thrombus

Untitled

Association of Clinician Diagnostic Performance With Machine Learning–Based Decision Support Systems

Contact Info

Product

Resources

About