Computer-aided assessment of breast density: Comparison of supervised deep learning and feature based statistical learning.

In this study, our purpose was to develop a supervised deep learning approach for automated estimation of percentage density (PD) on digital mammograms (DM). The input "for processing" DM was first log-transformed, enhanced by a multi-resolution preprocessing scheme, and subsampled to a pixel size of 800µm×800µm from 100µm×100µm. A deep convolutional neural network (DCNN) was trained to estimate a probability map of breast density (PMD) by using a domain adaptation resampling method. The PD was estimated as the ratio of the dense area to the breast area based on the PMD. The DCNN approach was compared to a feature-based statistical learning approach. Gray level, texture and morphological features were extracted and least absolute shrinkage and selection operator (LASSO) was used to combine the features into a feature-based PMD. With IRB approval, we retrospectively collected a training set of 478 DMs and an independent test set of 183 DMs from patient files in our institution. Two experienced MQSA radiologists interactively segmented PD as the reference standard. Ten-fold cross-validation was used for model selection and evaluation with the training set. With cross-validation, DCNN obtained a Dice's coefficient (DC) of 0.79±0.13 and Pearson's correlation (r) of 0.97 whereas feature-based learning obtained DC=0.72±0.18 and r=0.85. For the independent test set, DCNN achieved DC=0.76±0.09 and r=0.94 while feature...
Source: Physics in Medicine and Biology - Category: Physics Authors: Tags: Phys Med Biol Source Type: research

Related Links:

Publication date: Available online 17 October 2019Source: European Journal of RadiologyAuthor(s): Said Pertuz, Antti Sassi, Kirsi Holli-Helenius, Joni Kamarainen, Irina Rinta-Kiikka, Anna-Leena Laaperi, Otso ArponenAbstractPurposeTo assess the association between breast cancer risk and mammographic parenchymal measures obtained using a fully-automated, publicly available software, OpenBreast.MethodsThis retrospective case-control study involved screening mammograms of asymptomatic women diagnosed with breast cancer between 2016 and 2017. The 114 cases were matched with corresponding healthy controls by birth and screening ...
Source: European Journal of Radiology - Category: Radiology Source Type: research
AbstractPurposeThe future of non-operative management of DCIS relies on distinguishing lesions requiring treatment from those needing only active surveillance. More accurate preoperative staging and grading of DCIS would be helpful. We identified determinants of upstaging preoperative breast biopsies showing ductal carcinoma in situ (DCIS) to invasive breast cancer (IBC), or of upgrading them to higher-grade DCIS, following examination of the surgically excised specimen.MethodsWe studied all women with DCIS at preoperative biopsy in a large specialist cancer centre during 2000 –2014. Information from clinical records...
Source: Breast Cancer Research and Treatment - Category: Cancer & Oncology Source Type: research
An artificial intelligence (AI) tool - trained on roughly a million screening mammography images - identified breast cancer with approximately 90 percent accuracy when combined with analysis by radiologists, a new study finds. Led by researchers from NYU School of Medicine and the NYU Center for Data Science, the study examined the ability of a type of AI, a machine learning computer program, to add value to
Source: eHealth News EU - Category: Information Technology Tags: Featured Research Research and Development Source Type: news
Digital breast tomosynthesis (DBT) has a higher cancer detection rate compared...Read more on AuntMinnie.comRelated Reading: Synthesized 2D mammography identifies smaller lesions Ultrasound after DBT helpful in women with dense tissue Is screening with DBT worth the additional cost? DBT detects same tumor subtypes as digital mammography DBT finds more cancers in dense breast tissue
Source: Headlines - Category: Radiology Source Type: news
ConclusionsThis work has demonstrated a significant correlation between the degree of lesion enhancement in CESM and malignancy. Quantitative analysis of enhancement levels in CESM can distinguish between invasive cancers and benign or in situ lesions.Key Points• There is a possibility of quantitative assessment of contrast enhancement in CESM.• Correlation between quantitative assessment of contrast enhancement in CESM and histopathology was observed.
Source: European Radiology - Category: Radiology Source Type: research
Women must be offered lunchtime mammograms or smear tests to halt a collapse in screening rates, a report commissioned by the NHS has stated.
Source: the Mail online | Health - Category: Consumer Health News Source Type: news
Abstract In this article, we explore the evidence around the relative benefits and harms of breast cancer screening using a single radiologist to examine each female's mammograms for signs of cancer (single reading), or two radiologists (double reading). First, we briefly explore the historical evidence using film-screen mammography, before providing an in-depth description of evidence using digital mammography. We classify studies according to which exact version of double reading they use, because the evidence suggests that effectiveness of double reading is contingent on whether the two radiologists are blinded...
Source: The British Journal of Radiology - Category: Radiology Authors: Tags: Br J Radiol Source Type: research
AbstractBreast cancer (BC) is the most common cancer among females with more than 2 million new cases diagnosed worldwide in 2018. Although the prognosis in the majority of cases in the early stages combined with appropriate treatment is positive, there are still about 30% of patients who will develop locoregional diseases and distant metastases. Molecular imaging is very important in the diagnosis, staging, follow-up, and radiotherapy planning. Additionally, it is useful in characterizing lesions, prognosis, and therapy response in BC patients. Nuclear medicine imaging modalities (SPECT and PET) are of indispensable impor...
Source: Nuclear Medicine and Molecular Imaging - Category: Nuclear Medicine Source Type: research
American Journal of Roentgenology, Ahead of Print.
Source: American Journal of Roentgenology - Category: Radiology Authors: Source Type: research
Mammography screening for breast cancer saves lives. But another test may be better for women with dense breasts. Dr. Deborah Rhodes, an internal medicine specialist from the Breast Diagnostic Clinic at Mayo Clinic, says that data from a study being done at Mayo Clinic and other centers across the country suggest that MBI may also [...]
Source: News from Mayo Clinic - Category: Databases & Libraries Source Type: news
More News: Biology | Cancer | Cancer & Oncology | Computers | Learning | Mammography | Physics | Radiology | Statistics | Study | Training | Universities & Medical Training