Qualitative ultrasound training: defining the learning curve

Mullaney, P. J.

doi:10.1016/j.crad.2018.12.018

Cited by 20 publications

(18 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Abdominal ultrasonography (US) is a non-invasive, highly convenient, and versatile imaging technique that is commonly used for liver tumor diagnosis. However, extensive experience in US is required for accurate diagnoses because of the need to perform real-time recognition of lesions [ 3 ]. From this viewpoint, the lack of experts in US is an urgent issue in the medical sector that needs to be addressed because missing and incorrect diagnoses by non-experts will have serious consequences.…”

Section: Introductionmentioning

confidence: 99%

Artificial intelligence (AI) models for the ultrasonographic diagnosis of liver tumors and comparison of diagnostic accuracies between AI and human experts

et al. 2022

View full text Add to dashboard Cite

Background Ultrasonography (US) is widely used for the diagnosis of liver tumors. However, the accuracy of the diagnosis largely depends on the visual perception of humans. Hence, we aimed to construct artificial intelligence (AI) models for the diagnosis of liver tumors in US. Methods We constructed three AI models based on still B-mode images: model-1 using 24,675 images, model-2 using 57,145 images, and model-3 using 70,950 images. A convolutional neural network was used to train the US images. The four-class liver tumor discrimination by AI, namely, cysts, hemangiomas, hepatocellular carcinoma, and metastatic tumors, was examined. The accuracy of the AI diagnosis was evaluated using tenfold cross-validation. The diagnostic performances of the AI models and human experts were also compared using an independent test cohort of video images. Results The diagnostic accuracies of model-1, model-2, and model-3 in the four tumor types are 86.8%, 91.0%, and 91.1%, whereas those for malignant tumor are 91.3%, 94.3%, and 94.3%, respectively. In the independent comparison of the AIs and physicians, the percentages of correct diagnoses (accuracies) by the AIs are 80.0%, 81.8%, and 89.1% in model-1, model-2, and model-3, respectively. Meanwhile, the median percentages of correct diagnoses are 67.3% (range 63.6%–69.1%) and 47.3% (45.5%–47.3%) by human experts and non-experts, respectively. Conclusion The performance of the AI models surpassed that of human experts in the four-class discrimination and benign and malignant discrimination of liver tumors. Thus, the AI models can help prevent human errors in US diagnosis.

show abstract

Section: Introductionmentioning

confidence: 99%

Artificial intelligence (AI) models for the ultrasonographic diagnosis of liver tumors and comparison of diagnostic accuracies between AI and human experts

et al. 2022

View full text Add to dashboard Cite

show abstract

“…There are few papers about learning curves in the diagnostic radiology literature [ 18 , 24 ], and those that have been published so far have concerned musculoskeletal MRI and sonography, radiotherapy [ 25 ], the prostate [ 19 ], and interventional radiology [ 26 , 27 ]. The models presented in these studies report mathematical models of prediction through the use of logarithmic, exponential and sigmoid curves.…”

Section: Discussionmentioning

confidence: 99%

“…The curves were obtained based on the evolution of Cohen's Kappa compared to an expert in the technique. As in our study, the Kappa was used to assess inter-rater agreement, to reach an ideal value of ĸ = 1 [ 24 ]. Although the absence of the timing used by the individual trainees does not allow a comparison of the respective curves, the conclusions are similar to our study: in about 80 reports, the residents could reach the experienced Radiologist, as our logarithmic prediction of the curves shows.…”

Section: Discussionmentioning

confidence: 99%

“…The logarithmic prediction, used to value the achievement of ĸ = 1 [ 24 ], is esteem with moderate reliability because it is based on the hypothesis that learning grows following a logarithmic pattern as shown in a few articles in the scientific literature [ 26 , 27 ]. Using this mathematical model, it was possible to estimate the number of reporting sessions required (about 80).…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Learning curves in radiological reporting of whole-body MRI in plasma cell disease: a retrospective study

Negroni¹,

Cassarà

Trisoglio³

et al. 2021

Radiol med

View full text Add to dashboard Cite

Background The plasma cell disease is been studying by the whole-body MRI technology. However, the time requested to learn this radiological technique is unknown. Purpose To esteem, quantitatively and qualitatively, the essential time to learn the whole-body MRI diffusion-weighted imaging with background body signal suppression in patients with plasma cell disease. Materials and methods Between January 2015 and February 2017, three readers in-training with different levels of experience examined the anonymised and randomised whole-body MRI images of 52 patients with a diagnosis of plasma cell disease and analysed their morphological (T1w, T2w with and without fat suppression) and functional sequences. Reports of an expert radiologist were considered the standard of reference. Images were analysed in two sessions, during which each reader was timed. Readers reported the number of segments with lesions and staged the disease using the Durie–Salmon PLUS staging system. Weighted Cohen’s ĸ and Z-test were used to compare the trainees’ reports with those of the expert radiologist, and learning curves were drawn up to show changes between the two sessions. Results Weighted Cohen’s ĸ of number of lesioned segments increased from 0.536 ± 0.123 to 0.831 ± 0.129 (Prob > Z under 0.005), thus approaching the goal of ĸ > 0.8. Trainees reached the level of experienced radiologist in terms of time by the 33rd patient. Agreement concerning the Durie–Salmon PLUS increased from 0.536 ± 0.123 to 0.831 ± 0.129 (Prob > Z under 0.005). Conclusions The findings of this study demonstrate that whole-body MRI with DWIBS can be learned in about 80 reports and leads to a high level of inter-observer concordance when using the Durie–Salmon PLUS staging system.

show abstract

“…All MSU examinations were performed by four PTs (3 male and 1 female) with more than 5 years of experience in primary care and at least 2-years MSU experience (mean 3.75 years/SD 0.47) evaluating more than 150 diagnostic ultrasound scans of the shoulder in primary care (Mullaney, 2019). All four PTs had a certificate of 'basic MSU skills' and 'MSU of the shoulder masterclass'.…”

Section: Examinersmentioning

confidence: 99%

Diagnostic ultrasound in patients with shoulder pain: An inter-examiner agreement and reliability study among Dutch physical therapists

Duijn

Pouliart

Verhagen

et al. 2021

Musculoskeletal Science and Practice

View full text Add to dashboard Cite

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal ? Take down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

show abstract

Qualitative ultrasound training: defining the learning curve

Cited by 20 publications

References 21 publications

Artificial intelligence (AI) models for the ultrasonographic diagnosis of liver tumors and comparison of diagnostic accuracies between AI and human experts

Artificial intelligence (AI) models for the ultrasonographic diagnosis of liver tumors and comparison of diagnostic accuracies between AI and human experts

Learning curves in radiological reporting of whole-body MRI in plasma cell disease: a retrospective study

Diagnostic ultrasound in patients with shoulder pain: An inter-examiner agreement and reliability study among Dutch physical therapists

Contact Info

Product

Resources

About