Artificial Intelligence | News | Insights | AiThority
[bsfp-cryptocurrency style=”widget-18″ align=”marquee” columns=”6″ coins=”selected” coins-count=”6″ coins-selected=”BTC,ETH,XRP,LTC,EOS,ADA,XLM,NEO,LTC,EOS,XEM,DASH,USDT,BNB,QTUM,XVG,ONT,ZEC,STEEM” currency=”USD” title=”Cryptocurrency Widget” show_title=”0″ icon=”” scheme=”light” bs-show-desktop=”1″ bs-show-tablet=”1″ bs-show-phone=”1″ custom-css-class=”” custom-id=”” css=”.vc_custom_1523079266073{margin-bottom: 0px !important;padding-top: 0px !important;padding-bottom: 0px !important;}”]

Study using Volpara Health’s AI-powered image quality scoring wins Magna cum Laude award at ECR 2021

Volpara TruPGMI algorithm provides automatic, objective positioning feedback on every mammogram image

More than 1,000 scientific posters were presented at this year’s European Congress of Radiology (ECR). Only eight were selected to win the top award of Magna cum Laude — including a retrospective study examining the relationship between technical repeats from the UK NHS Breast Screening Program and Volpara’s automated PGMI scoring.

Volpara’s TruPGMI method first identifies positioning deficiencies and then categorizes each image as Perfect (P), Good (G), Moderate (M), or Inadequate (I), resulting in an overall assessment of an image’s quality from a positioning perspective. Images labeled inadequate are considered to be low in diagnostic quality and may be repeated by the technologist.

In the award-winning retrospective study, 2,134 technical repeat cases were automatically assessed by Volpara TruPGMI and compared to a reference set of accepted images — images which were not repeated.

Recommended AI News: Asirom uses Solera’s Artificial Intelligence for the Risk Inspection of Motor Policies

Volpara TruPGMI assigned significantly more inadequate scores to original images in the pool of technical repeats compared to the set of accepted images (41% vs 2%, p<0.001), and significantly less perfect scores (2% vs 17%, p<0.001).

Related Posts
1 of 23,440

Notably, of the technical repeat cases, 53% of them received a higher score than the original image. However, 40% of the technical repeats received the same score as the original image, and 7% of the technical repeats were scored lower than the original image.

The study found the most common positioning deficiency was an inadequate visualization of the pectoral muscle on medio lateral oblique (MLO) mammogram views. The inclusion of the pectoral muscle in a mammogram is key for ensuring all breast tissue is imaged and less cancers are missed.

Recommended AI News: StockSnips Inc Launches ESG News Sentiment Product

When looking at craniocaudal (CC) views, there were significantly more original images where the nipple was exaggerated from the central midline when compared to accepted images (41% vs 21%, p<0.001).

The results of this study highlight the role of artificial intelligence (AI) tools in providing objective image quality feedback independent of a mammography reader’s training or background. “Automated positioning measurement represents a promising means to continuously monitor image quality and radiographer performance in an objective manner,” first author Hannah-Mary Gilroy and her colleagues concluded.

Volpara’s automated approach is based on best practices from around the world (including the UK PGMI standard10) and enables breast imaging centers to achieve a high standard of mammographic image quality, to provide a continuous training program to advance technologist performance, and to more easily prepare for external quality audits such as the FDA Enhancing Quality Using the Inspection Program (EQUIP) initiative.

Recommended AI News: Army Research Lab Awards ICF $53 Million for Cyber Services

Leave A Reply

Your email address will not be published.