首页|Data on Breast Cancer Reported by Jianrong Jiang and Colleagues (A novel approach for segmentation and quantitative analysis of breast calcification in mammograms)
Data on Breast Cancer Reported by Jianrong Jiang and Colleagues (A novel approach for segmentation and quantitative analysis of breast calcification in mammograms)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning DailyNews Daily News - New research on Oncology - Breast Canc er is the subject of a report. According tonews reporting originating in Fujian , People’s Republic of China, by NewsRx journalists, research stated,“Breast ca ncer is a major threat to women’s health globally. Early detection of breast can cer is crucialfor saving lives.”The news reporters obtained a quote from the research, “One important early sign is the appearanceof breast calcification in mammograms. Accurate segmentation and analysis of calcification can improvediagnosis and prognosis. However, smal l size and diffuse distribution make calcification prone to oversight.This stud y aims to develop an efficient approach for segmenting and quantitatively analyz ing breastcalcification from mammograms. The goal is to assist radiologists in discerning benign versus malignantlesions to guide patient management. This stu dy develops a framework for breast calcification segmentationand analysis using mammograms. A Pro_UNeXt algorithm is proposed to accurately segmen t calcificationlesions by enhancing the UNeXt architecture with a microcalcific ation detection block, fused-MBConvmodules, multiple-loss-function training, an d data augmentation. Quantitative features are then extractedfrom the segmented calcification, including morphology, size, density, and spatial distribution. T hesefeatures are used to train machine learning classifiers to categorize lesio ns as malignant or benign. Theproposed Pro_UNeXt algorithm achieve d superior segmentation performance versus UNet and UNeXtmodels on both public and private mammogram datasets. It attained a Dice score of 0.823 for microcalcification detection on the public dataset, demonstrating its accuracy for small l esions. For quantitativeanalysis, the extracted calcification features enabled high malignant/benign classification, with AdaBoostreaching an AUC of 0.97 on t he private dataset. The consistent results across datasets validate therepresen tative and discerning capabilities of the proposed features. This study develops an efficientframework integrating customized segmentation and quantitative ana lysis of breast calcification. Pro_UNeXt offers precise localizatio n of calcification lesions. Subsequent feature quantification and machinelearni ng classification provide comprehensive malignant/benign assessment.”
FujianPeople’s Republic of ChinaAsiaBreast CancerBreast Cancer ScreeningCancerCyborgsDiagnostics and Scree ningEmerging TechnologiesHealth and MedicineMachine LearningMammogramMammographyOncologyRisk and PreventionWomen’s Health
NETL
NSTL
