Faculty of Medicine and Health
BREAST Newsletter
|
|
|
|
BREAST at the BreastScreen Conference in Canberra For the first time in 4-5 years, the BREAST team ran a workshop at the BreastScreen conference in Canberra, in March 2024. Our 2.5-day workshop was a great success, and for the first time, we opened up our sessions to Radiographers to participate in our 'priors' study lead by our PhD student Judith. Welcome to all of our new registrants and welcome to our BREAST Newsletter!
It was fabulous to see so many faces at the conference and to meet lovely new ones. Special thanks to the organising committee for such a wonderful conference and everyone's support of BREAST, it means so much to us.
At this recent workshop, BREAST released our latest 2D test set called High Breast Density (containing 60 cases), along with recruiting participants for multiple studies, including our AI study ran by Dr Yun Trieu, whereby AI selects cases to personalise test sets for individual improvement in breast mammography reading. If you have read cases for us previously and you are interested in participating, please email Mel at breastaustralia@sydney.edu.au, or find more information about our current studies at the end of this newsletter.
If you want to know what's coming up with BREAST, feel free to head to our website with all the latest information, news, publications and our previous newsletter editions.
| |
|
Multiple BREAST research abstracts presented at SPIE in San Diego, USA
BREAST was successful in getting a number of abstracts accepted for presentation in San Diego at SPIE Medical Imaging conference in February this year.
Ling (far right), one of our PhD students, presented her work titled 'Exploring varied time intervals on diagnostic performances of radiologists and trainees via test sets', whilst Howard (top, far left), our postdoc researcher, presented our current AI work titled 'Artificial intelligence can improve cancer detection in a double reading screening mammography scenario'. This AI work is part of the funded project by NBCF. If you are a member of SPIE, you can view our published SPIE proceedings for both the test sets as well as the AI work.
| |
Publication success for BREAST The team have recently had some great success with some of our hard work published in peer-reviewed journals.
In addtion to the SPIE proceedings as mentioned above, we have also published the following papers that can be viewed online in full:
- Evaluating recalibrating AI models for breast cancer diagnosis in a new context: Insights from transfer learning, image enhancement and high-quality training data integration; Cancers 2024, 16(2), 322; https://doi.org/10.3390/cancers16020322
- Surgical and radiology trainees' proficiency in reading mammograms: The importance of education for cancer localisation; Journal of Cancer Education (2024), Volume 39, p186-193; https://doi.org/10.1007/s13187-023-02393-7
- Familiarity, confidence and preference of artificial intelligence feedback and prompts by Australian breast cancer screening readers; Australian Health Review 2024, online early version available; https://doi.org/10.1071/AH23275
Please have a read. If you have completed any of our test sets, then this work has come from the deidentified data collected over many years at BREAST to conduct research to improve breast screening and mammography reading. These published papers are a true testimony of hard work by the BREAST team and the collaborations with readers, such as yourselves, to help pave the way for a better future in detecting early breast cancer.
| |
|
Grant success for BREAST - does this mean Ultrasound is coming?
At the end of last year, BREAST, in collaboration with Associate Professor Jillian Clarke, was successfully awarded a small project grant with the Australiasian Society for Ultrasound in Medicine (ASUM). Titled 'ULTRA_BREAST: Enhancing diagnositc proficiency in breast ultrasound', Jill will be leading the investigation.
Part of this research grant requires uploading ultrasounds to the BREAST platform. This means we may be upgrading our platform to incorporate ultrasound images, which could possibly extend to other test sets. Stay tuned!
| |
Opportunity to volunteer for a BREAST study - participants still needed!
Our research is always about conducting studies that aim to improve breast screening capabilities and experience. If you are interested in giving up some of your time, please get in contact with us via email as we are looking for volunteers for these two studies that will continue running into 2024. Both these projects can be done virtually, they do not need to be carried out on-site at Sydney University.
Study 1:
Explore the effectiveness of individualised training test sets when applying intelligent algorithm support to readers who interpret screening mammograms. The AI will harness the weaknesses and strengths of individual readers and assign the readers to targeted training cases to enhance the diagnositc efficacy in breast cancer detection.
Study 2:
A PhD research project that will investigate the influence of prior mammograms on diagnostic performance. The purpose is to explore the impact of access to prior mammograms on the accurate interpretation by BreastScreen readers, and to identify the characteristics of cases and readers that will benefit from accessing previous mammograms. This work involves our BREAST test sets that encompass a new 2D mammography test set, as well as our DBT Sydney and our junior training sets. If you have done either our DBT Sydney or Junior test set, please get in touch so we can asign the 'no priors' version of these test sets to your account.
| |
Want to collaborate with BREAST?
The BREAST team are calling for any interest in collaborating with us. Would you like to build a test set? Ideally, we are looking for radiologists who work for BreastScreen Australia who have access to images/cases with ground truth. Those that work privately but have access to cases that include ground truth are also welcome to contact us. If you're interested, please reach out to Mel, the BREAST program manager, on breastaustralia@sydney.edu.au.
| |
Follow BREAST on X (formally Twitter)
Lastly, for more information on what's happening with BREAST, you are most welcome to view our website, where you will find information on past projects, funding, news, and other related information or follow us on X @syd_BREAST.
| |
Kindest regards from the BREAST team,
Sarah, Yun, Dania, Howard, Jayden and Mel
| |
|
|
|
|
|