AI Article Synopsis
- AI has shown potential in improving breast cancer detection but lacks comprehensive real-world validation.
- A study evaluated a commercially available AI system against established double reading methods in Norway, analyzing over 122,000 screenings and identifying various cancer types based on AI scoring.
- The findings indicated that a significant majority of cancers were assigned high AI scores, suggesting the system's effectiveness, while also highlighting differences in histopathologic characteristics between selected and non-selected cancers.
Article Abstract
Background Artificial intelligence (AI) has shown promising results for cancer detection with mammographic screening. However, evidence related to the use of AI in real screening settings remain sparse. Purpose To compare the performance of a commercially available AI system with routine, independent double reading with consensus as performed in a population-based screening program. Furthermore, the histopathologic characteristics of tumors with different AI scores were explored. Materials and Methods In this retrospective study, 122 969 screening examinations from 47 877 women performed at four screening units in BreastScreen Norway from October 2009 to December 2018 were included. The data set included 752 screen-detected cancers (6.1 per 1000 examinations) and 205 interval cancers (1.7 per 1000 examinations). Each examination had an AI score between 1 and 10, where 1 indicated low risk of breast cancer and 10 indicated high risk. Threshold 1, threshold 2, and threshold 3 were used to assess the performance of the AI system as a binary decision tool (selected vs not selected). Threshold 1 was set at an AI score of 10, threshold 2 was set to yield a selection rate similar to the consensus rate (8.8%), and threshold 3 was set to yield a selection rate similar to an average individual radiologist (5.8%). Descriptive statistics were used to summarize screening outcomes. Results A total of 653 of 752 screen-detected cancers (86.8%) and 92 of 205 interval cancers (44.9%) were given a score of 10 by the AI system (threshold 1). Using threshold 3, 80.1% of the screen-detected cancers (602 of 752) and 30.7% of the interval cancers (63 of 205) were selected. Screen-detected cancer with AI scores not selected using the thresholds had favorable histopathologic characteristics compared to those selected; opposite results were observed for interval cancer. Conclusion The proportion of screen-detected cancers not selected by the artificial intelligence (AI) system at the three evaluated thresholds was less than 20%. The overall performance of the AI system was promising according to cancer detection. © RSNA, 2022.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9131175 | PMC |
| http://dx.doi.org/10.1148/radiol.212381 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Incidence of a Positive Sentinel Lymph Node Biopsy in Screen-Detected Early Breast Cancer.
Cureus
November 2024
General Surgery, Brighton and Sussex Medical School, Brighton, GBR.
Introduction Current guidelines advocate for a sentinel lymph node biopsy (SLNB) in patients with invasive breast cancer with negative axillary ultrasonography. However, emerging evidence has contradicted this, and SLNB omission has been found to be non-inferior in selected low-risk breast cancers. This retrospective study aimed to evaluate the incidence of SLNB in screen-detected invasive breast cancer.
View Article and Find Full Text PDFApplicability of the Scottish screen-detected polyp cancer study (SSPoCS) algorithm in a multicentric cohort in the management of malignant colorectal polyps.
Scand J Gastroenterol
December 2024
Gastroenterology Department, Centro Hospitalar de Leiria, Leiria, Portugal.
Background/objectives: Robust evidence regarding the management after endoscopic resection of malignant colorectal polyps (MCP) is lacking. Inconsistencies in reporting on potential prognostic factors hinder the decision process. To address these issues, the Scottish Screen-detected Polyp Cancer Study (SSPoCS) introduced an algorithm based in two easily obtainable variables: resection margin and lymphovascular invasion.
View Article and Find Full Text PDFSignificant variation in the assessment and management of screen-detected colorectal polyp cancers.
Int J Colorectal Dis
December 2024
University Hospitals Birmingham, Bordesley Green East, Birmingham, B9 5SS, UK.
Purpose: Endoscopic resection is appropriate for selected colorectal polyp cancers, but significant variation exists in treatment. This study aims to investigate variation in management of screen-detected polyp cancers (T1), factors predicting primary endoscopic polypectomy and threshold for subsequent surgical resection.
Method: Patients with polyp cancers (T1) diagnosed by the bowel cancer screening programme (BCSP) were investigated at two screening centres (5 individual sites and 4 MDTs, 2012-2022).
Prospective evaluation of lung cancer screening eligibility criteria and lung cancer detection in the Yorkshire Lung Screening Trial.
J Thorac Oncol
December 2024
Department of Respiratory Medicine, Leeds Teaching Hospitals NHS Trust, Leeds, UK; Leeds Institute of Health Sciences, University of Leeds, Leeds, UK. Electronic address:
Introduction: Low dose CT (LDCT) screening for lung cancer reduces lung cancer mortality, but there is a lack of international consensus regarding the optimal eligibility criteria for screening. The Yorkshire Lung Screening Trial (YLST) was designed to evaluate lung cancer screening (LCS) implementation and a primary objective was prospective evaluation of 3 pre-defined eligibility criteria.
Methods: Individuals who had ever smoked, aged 55-80yrs, who responded to written invitation, underwent telephone risk assessment and if eligible by at least one criteria (PLCO≥1.
Interhospital variations in diagnostic work-up following recall at biennial screening mammography-a population-based study.
Eur Radiol
December 2024
Canisius Wilhelmina Hospital, Weg door Jonkerbos 100, 6532 SZ, Nijmegen, The Netherlands.
Objectives: Quality control in breast cancer screening programmes has been subject of several studies. However, less is known about the clinical diagnostic work-up in recalled women with a suspicious finding at screening mammography. The current study focuses on interhospital differences in diagnostic work-up strategies.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!