Startseite


Cancer screening challenges for radiology

Mammography screenings are the key to early cancer detection and higher survival rates for the patients. Unfortunately, the number of diagnostic images generated through screening programs is increasing beyond the coping capacity of the healthcare system. 


Ever-growing workload

The number of radiologists is insufficient for the ever-growing workload they face every year. The workflow for examining a case involves many time-consuming, repetitive steps, causing radiologists to have very little time to search for cancer lesions. 

In 2005

0
Number of radiologists per capita
0
Number of radiological examinations

In 2017

0
Number of radiologists per capita
0
Number of radiological examinations
Reference: Physician statistics. German Medical Association & OECD (2021)

2,9 million cases to be examined in the German mammography screening program

  • 97 % healthy
  • 3 % suspicious

Only 3% of the cases require additional investigation

  • 81% healthy – false positive
  • 14% cancer positive
  • 5% missed – false negative

Asymmetry of the screening cases

97% of the breast screening cases turn out negative for cancer lesions. Radiologists need to further examine only approximately 3% (87,000) of the cases in order to determine whether they can diagnose them as cancer cases. Only 14% of the suspicious cases are diagnosed positive with cancer, while 81% turn out negative results and unfortunately 5% (6000) remain undetected.

Reference: Annual Evaluation Report 2018. German Mammography Screening Program

The NOVU Solution

AI technologies and machine learning can help doctors to detect and quantify cancer lesions faster and more precisely. NOVU develops web-based breast cancer detection software powered by artificial intelligence, ensuring faster workflow for the radiologist. The breast cancer diagnosis AI solution provides accuracy and efficiency, positively contributing to a problem that has perturbed the healthcare industry for decades.

How does our solution benefit radiologists

Optimized workflow

The NOVU solution aggregates information and displays a complete anamnesis overview with a single click, allowing doctors to have an overview of the case and eliminating any distraction to their workflow.

Prioritization of suspicious scans

Over 90% of the yearly scans generated by the mammography screening program turn out negative. The AI cancer detection algorithms detect the suspicious lesions and sort out the scans. Sensitive cases are presented first to the radiologists, allowing them to distribute their concentration and efficiency optimally.

Guided characterization of detected lesions

Within the embedded medical viewer, the artificial intelligence core displays the suspicious lesions and suggests all associated features. Radiologists confirm, deny or add any findings. The segmentation and classification of the lesions are done in compliance with the medical standards, following the standard cancer screening system for radiology.

Automatized reporting

While the user performs the guided analysis, the AI software generates the report in line with the international standards and exports it to the healthcare IT system automatically.

Modular plug-and-play solution

The software consists of modules that allow it to adapt its capabilities and features according to the user’s needs. The solution is web-based, platform-independent and accessible easily through any Internet browser.

Simple integration with healthcare IT systems

Thanks to the dedicated IT integration module, the software integrates seamlessly into the hospital or medical practice infrastructure. The relevant data is drawn from the healthcare facility IT system, processed and the reports are exported back to the facility system.

The NOVU artificial intelligence ecosystem

  • null

    Plug-and-Play AI Core

  • null

    On-Premise Integration

  • null

    GDPR Compliant Framework

  • null

    End-to-End Encryption

  • null

    Audit-proof Documentation

  • null

    Resource Monitorization and Auto-Scaling

We are

We are

Our purpose at NOVU is to join the international fight against cancer and enhance patients’ survival rates through the early detection of suspicious lesions. We want to implement innovative artificial intelligence and machine learning technologies to save lives. We want to create an ecosystem where doctors can trust AI technologies to face the current challenges in healthcare. Breast cancer and lung cancer have become the most common type of cancer among women and men, both for new cases and mortality rates. At NOVU, we are a team of medical specialists, engineers and business experts working to develop cutting-edge, efficiency-boosting solutions driven by AI to support doctors, governments and patients. 

Funded & supported by

Funded & supported by

*OECD (2021), Computed tomography (CT) exams (indicator). doi: 10.1787/3c994537-en
OECD (2021), Magnetic resonance imaging (MRI) exams (indicator). doi: 10.1787/1d89353f-en (accessed on June 7th, 2021)
Forsberg, D., Rosipko, B., & Sunshine, J.L., (2017).Radiologists’ variation of time to read across different procedure types. Journal of Digital Imaging 30(1), 86-94.
Close Bitnami banner
Bitnami