Early breast cancer detection with AI + mobile ultrasound
Saving women's lives in underserved communities
Breast cancer is the most common cancer and kills unproportionally in LMICs
Women diagnosed globally 2022
Global deaths 2022
Deaths in Low and Middle Income Countries
Mobile ultrasound hardware is 25x more affordable than mammography.
We use readily available mobile ultrasound probes. Multiple vendors and growing competition drive prices down and quality up. By running on smartphones instead of bulky, hard-to-upgrade consoles, our solution ensures flexibility and ease of use. Open APIs have further enabled a thriving ecosystem of both commercial and research applications.
We’ve built a bespoke app to capture breast cancer
Training is often a serious bottleneck in ultrasound workflows. We strip away everything not essential to breast cancer scanning, less features means it's easier to use. Our goal is to train non-radiologist health care workers up to 10-100x faster.
On device AI analysis
in less than a second
Our breast cancer AI is grounded in research from the Matemathical and Radiology departements at Lund University. It runs without network connection, on any mobile device, and analyses images in less than 100 milliseconds.
- Central in building $100M Ava Women FemTech startup
- Led the development and brought 2 AI-based medical devices to market.
- FemTech trailblazer: author of "Go Figure! The astonishing science of the female. body.", "10 women in Tech in Switzerland to follow 2022", Ted speaker and Forbes contributor.
- Published the largest ever randomized control trial for AI in medicine (Lancet Oncology).
- Nature Medicine 2023:
Top Ten Advancements in Medicine - Breast cancer trailblazer: Chair swedish breast cancer screening guidelines, most cited breast cancer imaging award.
- Modcam co-founder, acquired by Cisco.
- AI & IoT Expert: Builds smart infrastructure with deep AI and edge computing expertise.
- TAT Key employee (acquired by BlackBerry, 100 Mio USD). Pioneered mobile UI design, powering billions of devices globally.
- Design Leader & Founder: 20+ years UX/UI for IKEA, Samsung, fintechs and scaleups.
- Research focusing on deep learning for medical applications, including breast cancer detection and heart disease prediction.
- 5+ years experience in breast cancer detection using ultrasound imaging.
- Led research in machine learning and computer vision focusing on medical applications, including detection of cancer, heart disease and dementia, within radiology and pathology.
Ultrasound + AI outperforms mammography
The performance of our AI + ultrasound can distinguish a suspicious breast tumor from a non-suspicious with a higher accuracy than mammography.
Ultrasound AI AUC1
Mammography AUC2
* Cancer vs non-cancer
Commodity HW
Equip
€100,000+
€4,000+
Higher accessibility through lower price and portability.
Our innovation
Capture
Years of education
10 minutes training
Task-shifting towards community health workers.
Our innovation
Analyze
Decades experience
< 100 ms
Early breast cancer detection.
Avoid unnecessary biopsies.
Reduction of false positives.
Patient flow
Awareness
->
Clinical
breast exam
->
N23 AI mobile
ultrasound triage
->
Biopsy
->
Treatment
Publications
- Karlsson, et al. "Breast Cancer Classification in Point-of-Care Ultrasound Imaging — the Impact of Training Data." Journal of Medical Imaging (2025)
- Wodrich, et al. "Trustworthiness for Deep Learning Based Breast Cancer Detection Using Point-of-Care Ultrasound Imaging in Low-Resource Settings." MICCAI meets Africa Workshop, Springer Nature (2024)
- Karlsson, et al. "Towards Out-of-Distribution Detection for Breast Cancer Classification in Point-of-Care Ultrasound Imaging." International Conference on Pattern Recognition, Springer Nature (2024)
- Karlsson, et al. "A. Classification of point-of-care ultrasound in breast imaging using deep learning" Medical Imaging 2023: Computer-Aided Diagnosis. (2023)
- Karlsson, et al. "Machine learning algorithm for classification of breast ultrasound images." In Medical Imaging 2022: Computer-Aided Diagnosis. (2022)
Kristina Lång
2023 Top ten advancements in medicine
Artificial intelligence-supported screen reading versus standard double reading in the Mammography Screening with Artificial Intelligence trial (MASAI): a clinical safety analysis of a randomised, controlled, non-inferiority, single-blinded, screening accuracy study