Automated and real-time segmentation of suspicious breast masses using convolutional neural network

PloS One
Viksit KumarAzra Alizad

Abstract

In this work, a computer-aided tool for detection was developed to segment breast masses from clinical ultrasound (US) scans. The underlying Multi U-net algorithm is based on convolutional neural networks. Under the Mayo Clinic Institutional Review Board protocol, a prospective study of the automatic segmentation of suspicious breast masses was performed. The cohort consisted of 258 female patients who were clinically identified with suspicious breast masses and underwent clinical US scan and breast biopsy. The computer-aided detection tool effectively segmented the breast masses, achieving a mean Dice coefficient of 0.82, a true positive fraction (TPF) of 0.84, and a false positive fraction (FPF) of 0.01. By avoiding positioning of an initial seed, the algorithm is able to segment images in real time (13-55 ms per image), and can have potential clinical applications. The algorithm is at par with a conventional seeded algorithm, which had a mean Dice coefficient of 0.84 and performs significantly better (P< 0.0001) than the original U-net algorithm.

References

Aug 1, 1993·Radiology·V P JacksonD B Kopans
Jan 10, 1998·AJR. American Journal of Roentgenology·P Skaane, K Engedal
Aug 1, 2002·Medical Physics·Karen DrukkerEllen B Mendelson
Apr 29, 2003·IEEE Transactions on Medical Imaging·Anant Madabhushi, Dimitris N Metaxas
Jul 25, 2003·Ultrasound in Medicine & Biology·Dar-Ren ChenWen-Lin Wu
Feb 5, 2005·Breast Cancer Research and Treatment·Ruey-Feng ChangDar-Ren Chen
Aug 31, 2010·IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society·Chunming LiMartin D Fox
Sep 14, 2010·Radiographics : a Review Publication of the Radiological Society of North America, Inc·Sughra RazaRobyn L Birdwell
Jan 22, 2011·Journal of Oncology·Lawrence N ShulmanFelicia M Knaul
Nov 23, 2011·Ultrasound in Medicine & Biology·Jia-Wei TianXiang-Long Tang
Dec 14, 2011·International Journal of Emergency Medicine·Stephanie SippelSachita Shah
Feb 24, 2012·AJR. American Journal of Roentgenology·Cristina M CheckaHildegard Toth
Jan 7, 2015·CA: a Cancer Journal for Clinicians·Rebecca L SiegelAhmedin Jemal
Oct 27, 2015·JAMA : the Journal of the American Medical Association·Nancy L Keating, Lydia E Pace

❮ Previous
Next ❯

Citations

Dec 10, 2020·Diagnostics·Tomoyuki FujiokaUkihide Tateishi
Jan 13, 2021·Ultrasonography·Jaeil KimWon Hwa Kim
Jul 3, 2021·Biomedicines·Masaaki KomatsuRyuji Hamamoto
Aug 10, 2021·Frontiers in Oncology·Yu-Meng LeiChristoph F Dietrich
Oct 16, 2020·Computers in Biology and Medicine·Wilfrido Gómez-Flores, Wagner Coelho de Albuquerque Pereira

❮ Previous
Next ❯

Methods Mentioned

BETA
imaging technique
biopsy

Software Mentioned

Multi U
Python
Titan xp
Keras Python library
Multi U - net
Multi
net

Related Concepts

Related Feeds

Carcinoma, Ductal

Ductal carcinoma is a malignant neoplasm involving the ductal systems of any of a number of organs, such as the mammary glands, pancreas, prostate or lacrimal gland. Discover the latest research on ductal carcinoma here.

Carcinoma, Lobular

Lobular carcinoma is an invasive type of breast cancer in which lobules, areas of the breast that produce milk, undergo malignant transformation. Here is the latest research.

Breast Invasive Carcinoma

Invasive breast cancers indicate a spread into breast tissues and lymph nodes. Here are the latest discoveries pertaining to breast invasive carcinomas.

Cancer Imaging

Imaging techniques, including CT and MR, have become essential to tumor detection, diagnosis, and monitoring. Here is the latest research on cancer imaging.

Breast Invasive Carcinoma (Keystone)

Invasive breast cancers indicate a spread into breast tissues and lymph nodes. Here are the latest discoveries pertaining to breast invasive carcinomas.