Quantib Prostate: an example of how AI in radiology is the present, not the future

Pubblicato il admin

Back in 2019, the EAU (European Association of Urology) and the ACR (American College of Radiology) changed their guidelines and included a pre-biopsy MRI as a standard procedure for prostate cancer (PCa) diagnosis.

These changes in the guidelines were motivated by the promising results of the PROMIS study, published back in 2017, that showed that the negative predictive value (NPV) of multiparametric prostate MRI is 89% while the previous standard, transrectal ultrasound-guided prostate biopsies, had been shown to be only 74%1.

Since then, the number of prostate MRI that are being performed has increased rapidly, and with that, the interest in the application of AI to the process of prostate MRI reading in search of higher efficacy and efficiency in patient care.

Current challenges in the workflow and how AI implementation can help combat them

Reading prostate MRI is a complex task that is difficult to master. Besides that, the process includes a long list of steps which can be extremely time-consuming and demanding.

Reading software packages tend to not offer all functionalities needed for precise evaluation of prostate MRIs, making the process tedious as radiologists have to change back and forth between different programs. On the other hand, AI software packages can include most of the functionalities from the same user interface, for example by using AI algorithms to power PSA-related calculations or prostate volume measurements, making the calculations more precise compared to the traditional ellipsoid approach2,3 and less time-consuming.

AI software can streamline the process further by automatically analyzing larger amounts of data at once (imaging sequences) or provide PI-RADS scoring support through a step by step guide.

AI software can also automatically generate a standardized report complete with key images ad findings that can be an extremely useful asset for the preparation and communication during multidisciplinary team meetings (MDTs).

Lastly, since AI software incorporate knowledge obtained from a large amount of MRI scans, it provides expert level reading support that will be invaluable for starting radiologists as it can shorten their learning curve.

About Quantib® Prostate

Quantib® Prostate is an AI software for prostate MRI reading is designed and developed by Quantib. The AI software is FDA cleared and CE marked (Class IIb) thus complying with the European MDR 2017/745, and therefore, deemed suitable and safe for its use in the clinical setting.  Quantib® Prostate provides:

  1. Automatic prostate segmentation, prostate volumetry and PSA density calculation, done automatically, and in seconds, thanks to its AI algorithm.
  2. Fast, AI-supported ROI segmentation of suspicious areas.
  3. Visual representation of the kinetic curves of enhancement per ROI and of the full prostate gland to support PI-RADS assessment.
  4. Visual PI-RADS v.2.1 scoring support.
  5. A structured and easy-to understand report that includes key images and visual representation of PI-RADS v.2.1 results, useful for better communication with referring physicians and patients.

The clinical impact of Quantib® Prostate

Multiple studies have been performed and recently published that evaluated the performance of Quantib® Prostate. A prospective fusion-biopsy study done in Germany showed that Quantib® Prostate detected all significant PCa, as well as extra PCa not detected by initial radiology reads4. An Italian single-center study showed that an AI-assisted inexperienced radiologist was able to find all PCa positive lesions increasing diagnostic performance compared to an experienced radiologist5. Lastly, another Italian single-center study concluded that Quantib® Prostate can accelerate prostate MRI reading, making experienced radiologist read almost 10% faster, and inexperienced radiologists read 14% faster. Besides that, the study also showed that the use of Quantib® Prostate enabled a slightly higher inter-reader agreement in mpMRI 6.

In essence, Quantib® Prostate is an example of responsible integration of AI in the clinical practice since it is a clinically validated AI software that has a zero footprint user interface and allows easy integration with PACS.

References

  1. Ahmed, H. U. et al. Diagnostic accuracy of multi-parametric MRI and TRUS biopsy in prostate cancer (PROMIS): a paired validating confirmatory study. The Lancet 389, (2017).
  2. Aprikian, S. et al. Improving ultrasound-based prostate volume estimation. BMC Urology 19, (2019).
  3. Rodriguez, E., Skarecky, D., Narula, N. & Ahlering, T. E. Prostate Volume Estimation Using the Ellipsoid Formula Consistently Underestimates Actual Gland Size. Journal of Urology 179, (2008).
  4. Günzel, K. et al. “Artificial intelligence” (Quantib®Prostate) to improve cancer detection in transperineal MRI/ultrasound fusion biopsies of the prostate – The next game changer? ESUR CONFERENCE PRESENTATION (2022).
  5. Faiella, E. et al. Quantib Prostate Compared to an Expert Radiologist for the Diagnosis of Prostate Cancer on mpMRI: A Single-Center Preliminary Study. Tomography (2022)
  6. Cipollari, S. et al. Biparametric prostate MRI: impact of a deep learning-based software and of quantitative ADC values on the inter-reader agreement of experienced and inexperienced readers. La radiologia medica (2022).

Site: www.quantib.com