Overview of trials on artificial intelligence algorithms in breast cancer screening - A roadmap for international evaluation and implementation.

Accumulating evidence from retrospective studies demonstrate at least non-inferior performance when using AI algorithms with different strategies versus double-reading in mammography screening. In addition, AI algorithms for mammography screening can reduce work load by moving to single human reading. Prospective trials are essential to avoid unintended adverse consequences before incorporation of AI algorithms into UK's National Health Service (NHS) Breast Screening Programme (BSP). A stakeholders' meeting was organized in Newnham College, Cambridge, UK to undertake a review of the current evidence to enable consensus discussion on next steps required before implementation into a screening programme. It was concluded that a multicentre multivendor testing platform study with opt-out consent is preferred. AI thresholds from different vendors should be determined while maintaining non-inferior screening performance results, particularly ensuring recall rates are not increased. Automatic recall of cases using an agreed high sensitivity AI score versus automatic rule out with a low AI score set at a high sensitivity could be used. A human reader should still be involved in decision making with AI-only recalls requiring human arbitration. Standalone AI algorithms used without prompting maintain unbiased screening reading performance, but reading with prompts should be tested prospectively and ideally provided for arbitration.

An optoacoustic imaging feature set to characterise blood vessels surrounding benign and malignant breast lesions.

Combining optoacoustic (OA) imaging with ultrasound (US) enables visualisation of functional blood vasculature in breast lesions by OA to be overlaid with the morphological information of US. Here, we develop a simple OA feature set to differentiate benign and malignant breast lesions. 94 female patients with benign, indeterminate or suspicious lesions were recruited and underwent OA-US. An OA-US imaging feature set was developed using images from the first 38 patients, which contained 14 malignant and 8 benign solid lesions. Two independent radiologists blindly scored the OA-US images of a further 56 patients, which included 31 malignant and 13 benign solid lesions, with a sensitivity of 96.8% and specificity of 84.6%. Our findings indicate that OA-US can reveal vascular patterns of breast lesions that indicate malignancy using a simple feature set based on single wavelength OA data, which is therefore amenable to application in low resource settings for breast cancer management.

Opportunities in cancer imaging: risk-adapted breast imaging in screening.

In the UK, women between 50-70 years are invited for 3-yearly mammography screening irrespective of their likelihood of developing breast cancer. The only risk adaption is for women with >30% lifetime risk who are offered annual magnetic resonance imaging (MRI) and mammography, and annual mammography for some moderate-risk women. Using questionnaires, breast density, and polygenic risk scores, it is possible to stratify the population into the lowest 20% risk, who will develop <4% of cancers and the top 4%, who will develop 18% of cancers. Mammography is a good screening test but has low sensitivity of 60% in the 9% of women with the highest category of breast density (BIRADS D) who have a 2.5- to fourfold breast cancer risk. There is evidence that adding ultrasound to the screening mammogram can increase the cancer detection rate and reduce advanced stage interval and next round cancers. Similarly, alternative tests such as contrast-enhanced mammography (CESM) or abbreviated MRI (ABB-MRI) are much more effective in detecting cancer in women with dense breasts. Scintimammography has been shown to be a viable alternative for dense breasts or for follow-up in those with a personal history of breast cancer and scarring as result of treatment. For supplemental screening to be worthwhile in these women, new technologies need to reduce the number of stage II cancers and be cost effective when tested in large scale trials. This article reviews the evidence for supplemental imaging and examines whether a risk-stratified approach is feasible.

A meta-analysis comparing the diagnostic performance of abbreviated MRI and a full diagnostic protocol in breast cancer.

AIM: To undertake a meta-analysis of the diagnostic performance of abbreviated (ABB) magnetic resonance imaging (MRI) and full diagnostic protocol MRI (FDP-MRI) in breast cancer. MATERIALS AND METHODS: This meta-analysis was performed using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis for Diagnostic Test Accuracy (PRISMA-DTA) guidelines. The PubMed and EMBASE databases were searched through August 2019 for studies comparing the diagnostic performance of ABB-MRI and FDP-MRI in the breast. Studies were reviewed by two authors independently according to eligibility and exclusion criteria and split into two subgroups (screening population studies and studies using cohorts enriched with known cancers) to avoid bias. Quality assessment and bias for diagnostic accuracy was determined with Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2). The diagnostic accuracy for each subgroup was pooled using a bivariate random effects model and summary receiver operating characteristic (sROC) curves produced. Sensitivities and specificities were compared using a paired t-test. RESULTS: Five screening (62/2,588 cancers/patients) and eight enriched cohort (540/1,432 cancers/patients) studies were included in the meta-analysis. QUADAS-2 assessment showed a low risk of bias in most studies. The pooled sensitivity/specificity/area under the receiver operating characteristic curve (AUC) for screening studies was 0.90/0.92/0.94 for ABB-MRI and 0.92/0.95/0.97 for FDP-MRI. The pooled sensitivity/specificity/AUC for enriched cohort studies was 0.93/0.83/0.94 for ABB-MRI and 0.93/0.84/0.95 for FDP-MRI. There was no significant difference in sensitivity or specificity using ABB-MRI or FDP-MRI (p=0.18 and 0.27, p=0.18 and 0.93, respectively). CONCLUSION: The diagnostic performances of the ABB-MRI and FDP-MRI protocols used in either screening or enriched cohorts were comparable. There was a large variation in patient population, study methodology, and abbreviated protocols reported.

Ten lessons of leadership: reflections of a female academic.

The defining qualities and characteristics of a good leader - visionary, decisive, clear communicator, honesty and integrity, commitment and passion, being inspirational, accountable, well organised with the ability to delegate and empower-have been well documented. There are many articles with advice on how to lead, and there are excellent courses on leadership. In this article, I would like to explore how the general advice pertains to academia (and an academic radiologist), and whether this advice is relevant and whether the principles of leadership are the same. I have reflected on my own career and I have come up with my own 10 lessons of leadership that I have learnt over my working life. There are elements of leadership that are common across different sectors of business and healthcare, and examining how these are relevant to academia can be illuminating.

Artificial intelligence in clinical imaging: a health system approach.

The development and application of artificial intelligence (AI) to radiology requires an approach that encompasses a health system. The UK government and National Health Service (NHS) are creating an ecosystem to facilitate academic/industrial partnerships aimed at accelerating the creation of relevant and robust AI tools, which will improve the development and delivery of healthcare imaging. A series of recent initiatives are described, which will drive the development and adoption of AI in clinical imaging.

Artificial intelligence in breast imaging.

This article reviews current limitations and future opportunities for the application of computer-aided detection (CAD) systems and artificial intelligence in breast imaging. Traditional CAD systems in mammography screening have followed a rules-based approach, incorporating domain knowledge into hand-crafted features before using classical machine learning techniques as a classifier. The first commercial CAD system, ImageChecker M1000, relies on computer vision techniques for pattern recognition. Unfortunately, CAD systems have been shown to adversely affect some radiologists' performance and increase recall rates. The Digital Mammography DREAM Challenge was a multidisciplinary collaboration that provided 640,000 mammography images for teams to help decrease false-positive rates in breast cancer screening. Winning solutions leveraged deep learning's (DL) automatic hierarchical feature learning capabilities and used convolutional neural networks. Start-ups Therapixel and Kheiron Medical Technologies are using DL for breast cancer screening. With increasing use of digital breast tomosynthesis, specific artificial intelligence (AI)-CAD systems are emerging to include iCAD's PowerLook Tomo Detection and ScreenPoint Medical's Transpara. Other AI-CAD systems are focusing on breast diagnostic techniques such as ultrasound and magnetic resonance imaging (MRI). There is a gap in the market for contrast-enhanced spectral mammography AI-CAD tools. Clinical implementation of AI-CAD tools requires testing in scenarios mimicking real life to prove its usefulness in the clinical environment. This requires a large and representative dataset for testing and assessment of the reader's interaction with the tools. A cost-effectiveness assessment should be undertaken, with a large feasibility study carried out to ensure there are no unintended consequences. AI-CAD systems should incorporate explainable AI in accordance with the European Union General Data Protection Regulation (GDPR).