Back to search

IKTPLUSS-IKT og digital innovasjon

AIforScreening: Robust and trustworthy AI for breast cancer screening with mammography

Alternative title: AIforScreening: Robust og pålitelig AI for brystkreftscreening med mammografi

Awarded: NOK 12.0 mill.

AIforScreening: Robust and trustworthy AI for breast cancer screening This project concerns research on methods and approaches for robust and trustworthy AI for breast cancer screening with mammography. The project is led by the Norwegian Computing centre, with experts on image analysis and AI, collaborating with a team of medical experts and practitioners within mammographic screening from the Cancer Registry of Norway, the Norwegian Breast Centres and Karolinska University Hospital. The aim of the project is to move research, gain experience and arrive at suitable methods and approaches that can lead to robust and trustworthy application of AI in mammographic screening. The use of AI can lead to a larger capacity to do screening and detect cancers, and by this save lives and contribute to improved health services. This does however require solutions that can work over time and are trustworthy, sustainable and accountable. Furthermore, we need to understand the effects of using AI in this context. Artificial intelligence (AI) building on deep neural networks is now showing potential in mammographic screening to increase the sensitivity of detecting cancers. This suggests that AI can have a role in aiding the early detection of breast cancer. Still, the real world is more complicated and diverse than controlled research environments. Hence, there is a need for addressing topics related to AI for breast cancer screening that can bring todays promising results into real world applications. To address the problem of obtaining robust AI screening over time, the project aims at developing methods that can handle changes between laboratories and changes over time, that can exploit data from examinations over time, and which are optimized for use in clinical settings. To obtain trustworthy screening we will also work on the understanding and explanation of predictions and on understanding the AI-predictions’ potential effects on radiologists’ decisions.

This research project will develop AI methods and approaches for robust, sustainable and trustworthy breast cancer screening with mammography. We expect findings will be transferable to other medical screening programs. Breast cancer is the most common cancer and the leading cause of cancer death among women worldwide. Early detection of breast cancer through screening is recommended by international health organizations to reduce this mortality. It is considered to expand the screening program by examining more women. The use of AI can lead to a larger capacity to do screening and detect cancers, reduce overdiagnosis, and by this save lives and give new opportunities for improved health services. Furthermore, we need to understand the effects of using AI in this context. To address this, the project will involve AI experts, medical experts, organizers and practitioners within mammographic screening to create a strong interdisciplinary team. The main novelties of the proposed project are: * Develop methods that can handle domain shifts without costly annotation of new data. Current methods have problems handing images from different equipment, operators and cohorts. * Exploitation of time series in AI breast cancer screening. Radiologists exploit prior mammograms when doing their manual reading, while it is rarely used in current AI-based models. * Interpretation of predictions from AI breast cancer models such that radiologists can understand the predictions. This gives trust and makes it easier to improve the AI method and combine this with manual reading by radiologists. * Find the differences in the prediction between radiologist and AI both statistically and using data collected in the screening program as a basis when we combine the two methods. * Develop AI systems that is usable for radiologists in mammographic screening utilizing the strength of each approach. A new screening program should give better predictions and use less radiology resources.

Publications from Cristin

No publications found

No publications found

No publications found

No publications found

Activity:

IKTPLUSS-IKT og digital innovasjon

Thematic Areas and Topics

Digitalisering og bruk av IKTIKT forskningsområdeBransjer og næringerIKT-næringenLTP2 Innovasjon i stat og kommuneIKT forskningsområdeMenneske, samfunn og teknologiPolitikk- og forvaltningsområderNæring og handelBransjer og næringerTjenesterettet FoULTP2 Muliggjørende og industrielle teknologierLTP2 Fagmiljøer og talenterLTP2 IKT og digital transformasjonHelseLTP2 Styrket konkurransekraft og innovasjonsevnePortefølje LivsvitenskapPolitikk- og forvaltningsområderDigitaliseringIKT forskningsområdeKunstig intelligens, maskinlæring og dataanalyseIKT forskningsområdeVisualisering og brukergrensesnittFornyelse og innovasjon i offentlig sektorPolitikk- og forvaltningsområderDigitalisering og bruk av IKTOffentlig sektorPolitikk- og forvaltningsområderHelse og omsorgFornyelse og innovasjon i offentlig sektorInnovasjonsprosjekter og prosjekter med forpliktende brukermedvirkningKjønnsperspektiver i forskningKjønnsperspektiver i forskningKjønn som perspektiv i problemstillingPortefølje Industri og tjenestenæringerPolitikk- og forvaltningsområderForskningPortefølje Muliggjørende teknologierGrunnforskningPortefølje HelseInternasjonaliseringResponsible Research & InnovationRRI MedvirkningAnvendt forskningInternasjonaliseringInternasjonalt prosjektsamarbeidLTP2 Fornyelse i offentlig sektorLTP2 Utvikle fagmiljøer av fremragende kvalitetLTP2 Helse, forebygging og behandlingBransjer og næringerHelsenæringenResponsible Research & InnovationLTP2 Et kunnskapsintensivt næringsliv i hele landetPortefølje Demokrati, styring og fornyelsePortefølje Naturvitenskap og teknologi