Microwave imaging for breast cancer screening: protocol for an open, multicentric, interventional, prospective, non-randomised clinical investigation to evaluate cancer detection capabilities of MammoWave system on an asymptomatic population across multiple European countries

• Published in: BMJ open 2024-11, Vol.14 (11), p.e088431
• Main Authors: Álvarez Sánchez-Bayuela, Daniel, Fernández Martín, Juan, Tiberi, Gianluigi, Ghavami, Navid, Giovanetti González, Rubén, Cruz Hernánez, Lina Marcela, Aguilar Angulo, Paul Martín, Martínez Gómez, Aarón Darío, Rodríguez Sánchez, Ana, Bigotti, Alessandra, Khalesi, Banafsheh, Pontoriero, Letizia, Calabrese, Massimo, Tagliafico, Alberto Stefano, Romero Castellano, Cristina
• Format: Article
• Language: English
• Subjects: Algorithms; Antennas; Artificial Intelligence; Asymptomatic; Breast cancer; Breast imaging; Breast Neoplasms - diagnosis; Breast Neoplasms - diagnostic imaging; Classification; Clinical decision making; Decision support systems; Dielectric properties; Early Detection of Cancer - methods; Electromagnetism; Europe; Female; Humans; Machine learning; Mammography; Mammography - methods; Mass Screening; Medical screening; Microwave Imaging; Microwaves; Multicenter Studies as Topic; Prospective Studies; Protocol; Radiation; Radiology and Imaging; Sensitivity and Specificity; Software; Support vector machines; Tissues; Women; X-rays
• ISSN: 2044-6055; 2044-6055
• DOI: 10.1136/bmjopen-2024-088431

IntroductionMicrowave imaging presents several potential advantages including its non-ionising and harmless nature. This open, multicentric, interventional, prospective, non-randomised trial aims to validate MammoWave’s artificial intelligence (AI)-based classification algorithm, leveraging microwave imaging, to achieve a sensitivity exceeding 75% and a specificity exceeding 90% in breast screening.Methods and analysis10 000 volunteers undergoing regular mammographic breast cancer screening will be recruited across 9 European centres and invited to participate in the clinical study, involving MammoWave testing on both breasts. MammoWave results will be checked against the reference standard, to be intended as the output of conventional breast examination path (with histological confirmation of cancer cases) with 2 years follow-up. Anonymised clinical and MammoWave’s results, including microwave images, associated features and a label provided by the AI-based classification algorithm, will be collected and stored in a dedicated electronic case report form. The prospective study will involve a comparative analysis between the output of the conventional breast examination path (control intervention) and the labels provided by MammoWave’s AI system (experimental intervention). These labels will categorise breasts into two groups: breast With Suspicious Finding, indicating the presence of a suspicious lesion or No Suspicious Finding, indicating the absence of a lesion or the presence of a low-suspicion lesion. This trial aims to provide evidence regarding the novel MammoWave’s AI system for detecting breast cancer in asymptomatic populations during screening.Ethics and disseminationThis study was approved by the Research Ethics Committee of the Liguria Region (CET), Italy (CET-Liguria: 524/2023—DB id 13399), the Research Ethics Committee of Complejo Hospitalario de Toledo (CEIC), Spain (CEIC-1094), the National Ethics Committee for Clinical Research (CEIC), Portugal (CEIC-2311KC814), the Bioethical Committee of Pomeranian Medical University in Szczecin, Poland (KB-006/23/2024) and the Zurich Cantonal Ethics Commission, Switzerland (BASEC 2023-D0101). The findings of this study will be disseminated through academic and scientific conferences as well as peer-reviewed journals.Trial registration number NCT06291896.