Barriers and incentives for choosing to specialise in mammography- A survey of Australian undergraduate diagnostic radiography students

INTRODUCTION

There is a projected shortage of radiographers working in breast screening in Australia. This study aimed to examine the perceived barriers and potential incentives among current undergraduate diagnostic radiography students to train, specialise and work in mammography.

METHODS

A survey was undertaken to assess students' experience of six modalities during both University study and clinical practice. Students were asked to rank their priority of which they would want to specialise and if training and working in mammography was an option.

RESULTS

From 248 students invited to participate, 101 students replied to the survey. Students indicated that they had been taught all modalities, however not all students had experienced them on clinical placement. Most students had observed or performed ultrasound (US) (90%, 90%), Computed Tomography (CT) (97%, 75%) and fluoroscopy (80%, 52%), though very few had experience of mammography (55%, 8%). US, CT, and MRI were the modalities that most students would like to specialise in. Male students reported being discriminated against working in mammography as this is "female only" work.

CONCLUSION

Students have more clinical access to the major modalities of US, CT and MRI, and this appears to be a major factor into why student wish to specialise in these modalities. Role extension and higher wages were given as possible incentives that would make specialising in Mammography more appealing. The gender divide may have to be bridged to enable crucial recruitment as male students feel discriminated against.

Intercountry analysis of breast density classification using visual grading

OBJECTIVE

Disagreement in mammographic breast density (MBD) assessment can impact breast cancer risk stratification, choices of further breast cancer screening intervals and pathways. This study examines whether intercountry MBD expectations and assessment approaches are associated with differences in MBD assessment.

METHODS

20 American Board of Radiology (ABR) examiners and 24 UK practitioners using the 4th edition BI-RADS^® lexicon assessed 40 mammogram cases of 20 females. 26 Royal Australian and New Zealand College of Radiologists (RANZCR) registered radiologists also assessed the same cases. Interobserver correlation and agreement were assessed using Spearman's correlation (ρ) and weighted kappa (κ_w), respectively.

RESULTS

Strong positive correlation was observed between the study cohorts on a binary scale (1-2 vs 3-4) [ABR examiners and RANZCR radiologists (ρ = 0.950); ABR examiners and UK practitioners (ρ = 0.940); and RANZCR radiologists and UK practitioners (ρ = 0.958)]. ABR and RANZCR radiologists demonstrated slight agreement [κ_w = 0.10; 95% confidence interval (CI) = -1.13-0.43], whereas ABR and UK practitioners showed a fair agreement [κ_w = 0.25; 95% CI = -0.42-0.61], and an almost perfect agreement was observed between RANZCR radiologists and UK practitioners [κ_w = 0.95; 95% CI = 0.91-0.97].

CONCLUSION

Findings demonstrate wide international and interobserver variability in MBD assessment. This level of variability underscores the need for automation and standardization of MBD assessment. Advances in knowledge: Intercountry analysis of MBD assessment shows variations, with less variation on the binary scale than on the 4-point scale. With this level of variation, automation and standardization of MBD assessment becomes more appropriate.

Evaluation of radiographers' mammography screen-reading accuracy in Australia

Introduction

This study aimed to evaluate the accuracy of radiographers’ screen-reading mammograms. Currently, radiologist workforce shortages may be compromising the BreastScreen Australia screening program goal to detect early breast cancer. The solution to a similar problem in the United Kingdom has successfully encouraged radiographers to take on the role as one of two screen-readers. Prior to consideration of this strategy in Australia, educational and experiential differences between radiographers in the United Kingdom and Australia emphasise the need for an investigation of Australian radiographers’ screen-reading accuracy.

Methods

Ten radiographers employed by the Westmead Breast Cancer Institute with a range of radiographic (median = 28 years), mammographic (median = 13 years) and BreastScreen (median = 8 years) experience were recruited to blindly and independently screen-read an image test set of 500 mammograms, without formal training. The radiographers indicated the presence of an abnormality using BI-RADS®. Accuracy was determined by comparison with the gold standard of known outcomes of pathology results, interval matching and client 6-year follow-up.

Results

Individual sensitivity and specificity levels ranged between 76.0% and 92.0%, and 74.8% and 96.2% respectively. Pooled screen-reader accuracy across the radiographers estimated sensitivity as 82.2% and specificity as 89.5%. Areas under the reading operating characteristic curve ranged between 0.842 and 0.923.

Conclusions

This sample of radiographers in an Australian setting have adequate accuracy levels when screen-reading mammograms. It is expected that with formal screen-reading training, accuracy levels will improve, and with support, radiographers have the potential to be one of the two screen-readers in the BreastScreen Australia program, contributing to timeliness and improved program outcomes.

Mammogram retrieval through machine learning within BI-RADS standards

A content-based mammogram retrieval system can support usual comparisons made on images by physicians, answering similarity queries over images stored in the database. The importance of searching for similar mammograms lies in the fact that physicians usually try to recall similar cases by seeking images that are pathologically similar to a given image. This paper presents a content-based mammogram retrieval system, which employs a query example to search for similar mammograms in the database. In this system the mammographic lesions are interpreted based on their medical characteristics specified in the Breast Imaging Reporting and Data System (BI-RADS) standards. A hierarchical similarity measurement scheme based on a distance weighting function is proposed to model user's perception and maximizes the effectiveness of each feature in a mammographic descriptor. A machine learning approach based on support vector machines and user's relevance feedback is also proposed to analyze the user's information need in order to retrieve target images more accurately. Experimental results demonstrate that the proposed machine learning approach with Radial Basis Function (RBF) kernel function achieves the best performance among all tested ones. Furthermore, the results also show that the proposed learning approach can improve retrieval performance when applied to retrieve mammograms with similar mass and calcification lesions, respectively.