A stakeholder analysis to prepare for real-world evaluation of integrating artificial intelligent algorithms into breast screening (PREP-AIR study): a qualitative study using the WHO guide.

BACKGROUND: The national breast screening programme in the United Kingdom is under pressure due to workforce shortages and having been paused during the COVID-19 pandemic. Artificial intelligence has the potential to transform how healthcare is delivered by improving care processes and patient outcomes. Research on the clinical and organisational benefits of artificial intelligence is still at an early stage, and numerous concerns have been raised around its implications, including patient safety, acceptance, and accountability for decisions. Reforming the breast screening programme to include artificial intelligence is a complex endeavour because numerous stakeholders influence it. Therefore, a stakeholder analysis was conducted to identify relevant stakeholders, explore their views on the proposed reform (i.e., integrating artificial intelligence algorithms into the Scottish National Breast Screening Service for breast cancer detection) and develop strategies for managing 'important' stakeholders. METHODS: A qualitative study (i.e., focus groups and interviews, March-November 2021) was conducted using the stakeholder analysis guide provided by the World Health Organisation and involving three Scottish health boards: NHS Greater Glasgow & Clyde, NHS Grampian and NHS Lothian. The objectives included: (A) Identify possible stakeholders (B) Explore stakeholders' perspectives and describe their characteristics (C) Prioritise stakeholders in terms of importance and (D) Develop strategies to manage 'important' stakeholders. Seven stakeholder characteristics were assessed: their knowledge of the targeted reform, position, interest, alliances, resources, power and leadership. RESULTS: Thirty-two participants took part from 14 (out of 17 identified) sub-groups of stakeholders. While they were generally supportive of using artificial intelligence in breast screening programmes, some concerns were raised. Stakeholder knowledge, influence and interests in the reform varied. Key advantages mentioned include service efficiency, quicker results and reduced work pressure. Disadvantages included overdiagnosis or misdiagnosis of cancer, inequalities in detection and the self-learning capacity of the algorithms. Five strategies (with considerations suggested by stakeholders) were developed to maintain and improve the support of 'important' stakeholders. CONCLUSIONS: Health services worldwide face similar challenges of workforce issues to provide patient care. The findings of this study will help others to learn from Scottish experiences and provide guidance to conduct similar studies targeting healthcare reform. STUDY REGISTRATION: researchregistry6579, date of registration: 16/02/2021.

Service and clinical impacts of reader bias in breast cancer screening: a retrospective study.

OBJECTIVES: To determine factors influencing reader agreement in breast screening and investigate the relationship between agreement level and patient outcomes. METHODS: Reader pair agreement for 83 265 sets of mammograms from the Scottish Breast Screening service (2015-2020) was evaluated using Cohen's kappa statistic. Each mammography examination was read by two readers, per routine screening practice, with the second initially blinded but able to choose to view the first reader's opinion. If the two readers disagreed, a third reader arbitrated. Variation in reader agreement was examined by: whether the reader acted as the first or second reader, reader experience, and recall, cancer detection and arbitration recall rate. RESULTS: Readers' opinions varied by whether they acted as the first or second reader. Furthermore, reader 2 was more likely to agree with reader 1 if reader 1 was more experienced than they were, and less likely to agree if they themselves were more experienced than reader 1 (P < .001). Agreement was not significantly associated with cancer detection rate, overall recall rate or arbitration recall rates (P > .05). Lower agreement between readers led to a higher arbiter workload (P < .001). CONCLUSIONS: In mammography screening, the second reader's opinion is influenced by the first reader's opinion, with the degree of influence dependent on the readers' relative experience levels. ADVANCES IN KNOWLEDGE: While less-experienced readers relied on their more experienced reading partner, no adverse impact on service outcomes was observed. Allowing access to the first reader's opinion may benefit newly qualified readers, but reduces independent evaluation, which may lower cancer detection rates.

Impact of Different Mammography Systems on Artificial Intelligence Performance in Breast Cancer Screening.

Artificial intelligence (AI) tools may assist breast screening mammography programs, but limited evidence supports their generalizability to new settings. This retrospective study used a 3-year dataset (April 1, 2016-March 31, 2019) from a U.K. regional screening program. The performance of a commercially available breast screening AI algorithm was assessed with a prespecified and site-specific decision threshold to evaluate whether its performance was transferable to a new clinical site. The dataset consisted of women (aged approximately 50-70 years) who attended routine screening, excluding self-referrals, those with complex physical requirements, those who had undergone a previous mastectomy, and those who underwent screening that had technical recalls or did not have the four standard image views. In total, 55 916 screening attendees (mean age, 60 years ± 6 [SD]) met the inclusion criteria. The prespecified threshold resulted in high recall rates (48.3%, 21 929 of 45 444), which reduced to 13.0% (5896 of 45 444) following threshold calibration, closer to the observed service level (5.0%, 2774 of 55 916). Recall rates also increased approximately threefold following a software upgrade on the mammography equipment, requiring per-software version thresholds. Using software-specific thresholds, the AI algorithm would have recalled 277 of 303 (91.4%) screen-detected cancers and 47 of 138 (34.1%) interval cancers. AI performance and thresholds should be validated for new clinical settings before deployment, while quality assurance systems should monitor AI performance for consistency. Keywords: Breast, Screening, Mammography, Computer Applications-Detection/Diagnosis, Neoplasms-Primary, Technology Assessment Supplemental material is available for this article. © RSNA, 2023.

AI in breast screening mammography: breast screening readers' perspectives.

OBJECTIVES: This study surveyed the views of breast screening readers in the UK on how to incorporate Artificial Intelligence (AI) technology into breast screening mammography. METHODS: An online questionnaire was circulated to the UK breast screening readers. Questions included their degree of approval of four AI implementation scenarios: AI as triage, AI as a companion reader/reader aid, AI replacing one of the initial two readers, and AI replacing all readers. They were also asked to rank five AI representation options (discrete opinion; mammographic scoring; percentage score with 100% indicating malignancy; region of suspicion; heat map) and indicate which evidence they considered necessary to support the implementation of AI into their practice among six options offered. RESULTS: The survey had 87 nationally accredited respondents across the UK; 73 completed the survey in full. Respondents approved of AI replacing one of the initial two human readers and objected to AI replacing all human readers. Participants were divided on AI as triage and AI as a reader companion. A region of suspicion superimposed on the image was the preferred AI representation option. Most screen readers considered national guidelines (77%), studies using a nationally representative dataset (65%) and independent prospective studies (60%) as essential evidence. Participants' free-text comments highlighted concerns and the need for additional validation. CONCLUSIONS: Overall, screen readers supported the introduction of AI as a partial replacement of human readers and preferred a graphical indication of the suspected tumour area, with further evidence and national guidelines considered crucial prior to implementation.

Exploring the influence of rural residence on uptake of organized cancer screening - A systematic review of international literature.

Lower screening uptake could impact cancer survival in rural areas. This systematic review sought studies comparing rural/urban uptake of colorectal, cervical and breast cancer screening in high income countries. Relevant studies (n = 50) were identified systematically by searching Medline, EMBASE and CINAHL. Narrative synthesis found that screening uptake for all three cancers was generally lower in rural areas. In meta-analysis, colorectal cancer screening uptake (OR 0.66, 95 % CI = 0.50-0.87, I2 = 85 %) was significantly lower for rural dwellers than their urban counterparts. The meta-analysis found no relationship between uptake of breast cancer screening and rural versus urban residency (OR 0.93, 95 % CI = 0.80-1.09, I2 = 86 %). However, it is important to note the limitation of the significant statistical heterogeneity found which demonstrates the lack of consistency between the few studies eligible for inclusion in the meta-analyses. Cancer screening uptake is apparently lower for rural dwellers which may contribute to poorer survival. National screening programmes should consider geography in planning.

Does Reader Performance with Digital Breast Tomosynthesis Vary according to Experience with Two-dimensional Mammography?

Purpose To assess whether individual reader performance with digital breast tomosynthesis (DBT) and two-dimensional (2D) mammography varies with number of years of experience or volume of 2D mammograms read. Materials and Methods After written informed consent was obtained, 8869 women (age range, 29-85 years; mean age, 56 years) were recruited into the TOMMY trial (A Comparison of Tomosynthesis with Digital Mammography in the UK National Health Service Breast Screening Program), an ethically approved, multicenter, multireader, retrospective reading study, between July 2011 and March 2013. Each case was read prospectively for clinical assessment and to establish ground truth. A retrospective reading data set of 7060 cases was created and randomly allocated for independent blinded review of (a) 2D mammograms, (b) DBT images and 2D mammograms, and (c) synthetic 2D mammograms and DBT images, without access to previous examinations. Readers (19 radiologists, three advanced practitioner radiographers, and two breast clinicians) who had 3-25 (median, 10) years of experience in the U.K. National Health Service Breast Screening Program and read 5000-13 000 (median, 8000) cases per annum were included in this study. Specificity was analyzed according to reader type and years and volume of experience, and then both specificity and sensitivity were analyzed by matched inference. The median duration of experience (10 years) was used as the cutoff point for comparison of reader performance. Results Specificity improved with the addition of DBT for all readers. This was significant for all staff groups (56% vs 68% and 49% vs 67% [P < .0001] for radiologists and advanced practitioner radiographers, respectively; 46% vs 55% [P = .02] for breast clinicians). Sensitivity was improved for 19 of 24 (79%) readers and was significantly higher for those with less than 10 years of experience (91% vs 86%; P = .03) and those with total mammographic experience of fewer than 80 000 cases (88% vs 86%; P = .03). Conclusion The addition of DBT to conventional 2D screening mammography improved specificity for all readers, but the gain in sensitivity was greater for readers with less than 10 years of experience.

Idiopathic orbital inflammation syndrome with retro-orbital involvement: a retrospective study of eight patients.

BACKGROUND: The aim of this retrospective study was to document the clinical findings and radiological features of idiopathic orbital inflammation syndrome with retro-orbital involvement. METHODS: We searched for ophthalmological patients who received orbital imaging at Zhejiang Provincial People's Hospital between October 2003 and April 2010. Seventy-three patients were diagnosed with idiopathic orbital inflammation syndrome based on clinicoradiological features, with pathological confirmation of nonspecific inflammatory conditions in 47 patients. Eight patients (11%) had MRI or CT evidence of retro-orbital involvement. All 8 patients were diagnosed with idiopathic orbital inflammation syndrome after biopsy of the orbital lesion. MR images were obtained for all 8 patients; 3 patients also had a contrast-enhanced CT scan. RESULTS: Seven out of 8 patients with retro-orbital involvement also had orbital apex lesions. Of the 65 patients without retro-orbital involvement, 19 had orbital apex lesions. The difference in the number of patients with orbital apex lesions between the two populations was significant (Fisher exact test P = .002). In all 8 patients with retro-orbital involvement, the inflammation spread through the superior orbital fissure. The retro-orbital lesions were isointense to grey matter on T1-weighted images, hypointense on T2-weighted images, and displayed uniform contrast enhancement; on contrast-enhanced CT scans, they were hyperdense relative to the contralateral mirror area and had radiological contours that were similar to those seen on MR images. The diffuse inflammation with marked sclerosis and hyalinization that we observed in the patients with retro-orbital involvement is consistent with the diagnosis of the sclerosing subtype of idiopathic orbital inflammation syndrome. All 8 patients also complained of mild to moderate periorbital pain (headache). CONCLUSIONS: In patients with idiopathic orbital inflammation syndrome, it is important to perform MRI and CT scans to identify possible retro-orbital involvement. Retro-orbital involvement is more frequent when the lesion is present in the orbital apex.