News & media


8 August 2018

White Paper: Innovations in Cancer Screening

White Paper: Innovations in Cancer Screening

There has been great progress in the diagnosis and treatment of cancer in recent years, driven largely by the increased use of medical imaging technologies. As the world’s population ages, the number of new cases of cancer is set to rise — to 21.7 million new cases by 2030[i] —  which necessitates improved efficiencies in the use of imaging technologies.

A new White Paper, Innovations in Cancer Screening showcases how innovations in imaging technology and the use of artificial intelligence and analytics are driving efficiencies in cancer imaging, by improving the accuracy and cost of medical imaging for the early diagnosis of cancer as well as increasing patient confidence in crucial preventive screening programs.

Key highlights include:

The importance of early cancer screening

Screening can identify cancer in its early stages before symptoms become apparent. It’s well established that early detection leads to better outcomes, such as increased likelihood of a positive response to treatment, a greater probability of survival and less expensive treatment.

For example, 5-year survival rates for lung cancer are 70% with an early diagnosis but only 14% late stage diagnosis[ii]. This is mirrored for ovarian cancer: 90% with an early diagnosis versus 5% with a late diagnosis[ii], and liver cancer: 31% early diagnosis versus 3% late diagnosis[iii].

Image-based screening can also be used to monitor tumor progression (of both malignant and benign tumors) and treatment effects.

The hurdles to early cancer detection

Although the World Health Organization and other bodies encourage healthcare professionals to educate patients about the value of early diagnosis and screening, the White Paper highlights factors that can discourage people from taking advantage of early diagnostic tools. The high rate of inaccuracy in reading image-based screens is particularly problematic.

One example cited is that despite adequate evidence that mammography screening reduces breast cancer mortality in women aged 40–74 years, a US Preventive Services Task Force recommended that women in their 40s make their own decision as to whether to be screened for breast cancer. The Task Force cited the frequency of false-positive test results as a primary factor behind their decision[iv].

The number of false positives can be high —  one study showed that 23% patients experienced false positives on image readings for certain cancer screenings[v] — and often leads to unnecessary invasive procedures and follow-up scans that increase anxiety for patients. This high rate of false positives in cancer screenings is due in part to the fact that images are traditionally evaluated visually by radiologists, who are often overly cautious for fear of missing something.

How cutting-edge imaging technology can overcome hurdles

The White Paper highlights how advances in artificial intelligence and analytics related to medical imaging can overcome challenges in medical imaging in cancer such as false-positive test results. One influential technology is computer-aided detection, which uses deep learning algorithms to automatically identify tumors or suspicious lesions in images. Importantly it can deliver more consistent analyses than visual assessments alone, reducing the rate of false negatives.

Computer-aided detection can also be used to:

  • Assess patients for clinical trial enrollment criteria
  • Track population outcome trends in drug development trials and wide-scale healthcare programs
  • Enable earlier diagnosis through the identification of small variances that might not be easily spotted by humans; for example, through 3D digital breast tomosynthesis[vi]
  • Robustly track lesions and snapshots over time
 Moving into the future

 Implementing new technologies to aid image-based cancer screening is not without its challenges. For example, the criteria for regulatory approval needs to be established and there must be a supply of trained personnel to deliver deep learning algorithms. Yet in the coming years it is anticipated that computer-aided detection systems and data-driven analyses to enable more efficient cancer screening, thus allowing radiologists to accurately diagnose their patients.

To find out more about the topics highlighted here, download Innovations in Cancer Screening now.






[v] Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening N Engl J Med. 2011 August 4; 365(5): 395–409.