Deep learning based on co-registered ultrasound and photoacoustic imaging improves assessment of rectal cancer chemoradiotherapy response

Yixiao Lin; Sitai Kou; Haolin Nie; Quing Zhu

doi:10.1117/12.2649328

9 March 2023 Deep learning based on co-registered ultrasound and photoacoustic imaging improves assessment of rectal cancer chemoradiotherapy response

Yixiao Lin, Sitai Kou, Haolin Nie, Quing Zhu

Author Affiliations +

Proceedings Volume PC12379, Photons Plus Ultrasound: Imaging and Sensing 2023; PC123790C (2023) https://doi.org/10.1117/12.2649328
Event: SPIE BiOS, 2023, San Francisco, California, United States

Abstract

We present results of rectal cancer treatment response assessment using co-registered ultrasound and photoacoustic imaging from over 20 in vivo patients. We develop a deep learning model based on co-registered dual-modality images with individualized prior information. Compared to models using only ultrasound images, our model identifies complete treatment responders with significantly higher accuracy. We achieve a 3-class classification accuracy (normal, cancer, and image artifact) of 89.1±0.8%. To facilitate surgeons’ decision-making, we generate localized hotspots to indicate suspicious cancer regions based on model predictions. We conclude that the addition of photoacoustic imaging to conventional ultrasound improves treatment response assessment.

Conference Presentation

Citation Download Citation

Yixiao Lin, Sitai Kou, Haolin Nie, and Quing Zhu "Deep learning based on co-registered ultrasound and photoacoustic imaging improves assessment of rectal cancer chemoradiotherapy response", Proc. SPIE PC12379, Photons Plus Ultrasound: Imaging and Sensing 2023, PC123790C (9 March 2023); https://doi.org/10.1117/12.2649328

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available