{"id":2115,"date":"2024-08-25T20:31:35","date_gmt":"2024-08-25T20:31:35","guid":{"rendered":"https:\/\/labs.cs.queensu.ca\/perklab\/?post_type=qsc_member&p=2115"},"modified":"2024-08-25T20:31:35","modified_gmt":"2024-08-25T20:31:35","slug":"olivia-radcliffe","status":"publish","type":"qsc_member","link":"https:\/\/labs.cs.queensu.ca\/perklab\/members\/olivia-radcliffe\/","title":{"rendered":"Olivia Radcliffe"},"content":{"rendered":"
\n\t
\n\t\t\"\"\n\t<\/div>\n\t
\n\t\t

Olivia Radcliffe<\/h1><\/div>\n\t\t
MSc Student<\/div>\n\t\t
School of Computing<\/div>\n\t\t
Queen’s University<\/div>\n\t\t
\n\t\t\t
olivia.radcliffe@queensu.ca<\/a><\/div>\n\t\t\t
\n\t\t\t<\/i><\/a>\n\t\t\t<\/div>\n\t\t<\/div>\n\t<\/div>\n<\/div>\n
\n\t\n

Biography<\/h2>\n\n\n\n

Olivia is a MSc student in the School of Computing under the supervision of Dr. Gabor Fichtinger and Dr. Parvin Mousavi. Her research involves cancer detection through micro-ultrasound and the iKnife in breast-conserving surgery<\/p>\n\n\n\n

Publications<\/h2>\n\n\n
<\/a><\/form>

Radcliffe, Olivia; Lawford-Wickham, Imogen A; Wilson, Paul FR; Abolmaesumi, Purang; Mousavi, Parvin<\/p>

Leveraging SAM for automatic prostate segmentation on micro-ultrasound images<\/a> Proceedings<\/span> <\/p>

SPIE Proceedings Volume 13412, Medical Imaging 2025: Ultrasonic Imaging and Tomography, <\/span>2025<\/span>.<\/p>

Abstract<\/a><\/span> | Links<\/a><\/span> | BibTeX<\/a><\/span><\/p>

@proceedings{nokey,
\r\ntitle = {Leveraging SAM for automatic prostate segmentation on micro-ultrasound images},
\r\nauthor = {Olivia Radcliffe and Imogen A Lawford-Wickham and Paul FR Wilson and Purang Abolmaesumi and Parvin Mousavi},
\r\ndoi = {10.1117\/12.3047368},
\r\nyear  = {2025},
\r\ndate = {2025-04-10},
\r\nurldate = {2025-04-10},
\r\nabstract = {PURPOSE: Deep learning is promising for enabling accurate and automatic prostate segmentation. Existing deep learning segmentation model approaches often rely on large training datasets for good generalization. We aim to adapt pre-trained foundation models to enable accurate, data-efficient, and robust prostate segmentation. METHODS: We adapt the promptable foundation model SAM1 for prostate segmentation on rotational micro-ultrasound scans2 of the prostate. We design novel prompting strategies to provide the model with enhanced 3D context awareness to improve segmentation. RESULTS: Our model, which we call SliceTrack-SAM, outperforms prior state-of-the-art in micro-ultrasound prostate segmentation. We achieve a Dice coefficient of 94.0% and Hausdorff distance 1.76mm on the MicroSegNet dataset. Qualitative analysis and ablation studies further validate the success of our approach. CONCLUSION: Transfer learning from pre-trained foundation models can alleviate the challenge of data scarcity and improve the generalization of medical imaging deep learning systems. Prompts provide a flexible and effective way to leverage auxiliary information, such as 3D context, when using these models.},
\r\nhowpublished = {SPIE Proceedings Volume 13412, Medical Imaging 2025: Ultrasonic Imaging and Tomography},
\r\nkeywords = {},
\r\npubstate = {published},
\r\ntppubtype = {proceedings}
\r\n}
\r\n<\/pre><\/div>
Close<\/a><\/p><\/div>
PURPOSE: Deep learning is promising for enabling accurate and automatic prostate segmentation. Existing deep learning segmentation model approaches often rely on large training datasets for good generalization. We aim to adapt pre-trained foundation models to enable accurate, data-efficient, and robust prostate segmentation. METHODS: We adapt the promptable foundation model SAM1 for prostate segmentation on rotational micro-ultrasound scans2 of the prostate. We design novel prompting strategies to provide the model with enhanced 3D context awareness to improve segmentation. RESULTS: Our model, which we call SliceTrack-SAM, outperforms prior state-of-the-art in micro-ultrasound prostate segmentation. We achieve a Dice coefficient of 94.0% and Hausdorff distance 1.76mm on the MicroSegNet dataset. Qualitative analysis and ablation studies further validate the success of our approach. CONCLUSION: Transfer learning from pre-trained foundation models can alleviate the challenge of data scarcity and improve the generalization of medical imaging deep learning systems. Prompts provide a flexible and effective way to leverage auxiliary information, such as 3D context, when using these models.<\/div>
Close<\/a><\/p><\/div>