Unlike the mpMRI protocol, biparametric MRI (bpMRI) does not include dynamic contrast-enhanced imaging —thereby reducing costs, eliminating any risk of adverse effects from the use of contrast agents, and shortening examination times (Turkbey et al., 2019). Thus, despite providing less diagnostic information than mpMRI (de Rooij et al., 2020), bpMRI is more suitable within the scope of high-volume, population-based screening (Eklund et al., 2021).

Figure. The challenge of discriminating csPCa due to its morphological heterogeneity. (a-b) T2-weighted imaging (T2W), (c-d) high b-value diffusion-weighted imaging (DWI) and (e-f) apparent diffusion coefficient (ADC) maps constituting prostate bpMRI scans for two different patients are shown above, where yellow contours indicate csPCa lesions. While one of the patients had large, severe csPCa developing from both ends (top row), the other was afflicted by a single, relatively focal csPCa lesion surrounded by perceptually similar nodules of benign prostatic hyperplasia (BPH) (bottom row). Probability density functions (right) reveal a large overlap between the distributions of malignant and non-malignant prostatic tissue intensities across all three MRI channels. (excerpt from Saha et al., 2021)

Modern artificial intelligence (AI) algorithms have paved the way for powerful computer-aided detection and diagnosis (CAD) systems that rival human performance in medical image analysis (Esteva et al., 2017, McKinney et al., 2020). Clinical trials are the gold standard for assessing new medications and interventions in a controlled and comparative manner, and the equivalent for developing AI algorithms are international competitions or “grand challenges”. Grand challenges can address the lack of trust, scientific evidence and adequate validation among AI solutions (Leeuwen et al., 2021), by providing the means to compare algorithms against each other in a bias-free manner, using common training and testing data. Prior to the release of this challenge, the only public benchmark of csPCa detection/diagnosis was the ProstateX Challenge, which used a testing set of 140 mpMRI exams to evaluate and compare AI algorithms. However, its small sample size, limited diversity (all cases from the same center and MRI vendor) and weak evaluation format (with publicly available, as opposed to truly “unseen” testing images), limited the ability to reliably draw out definitive conclusions.

The PI-CAI Challenge 👩‍⚕️🧑‍💻

PI-CAI (Prostate Imaging: Cancer AI) is an all-new grand challenge, with over 10,000 carefully-curated prostate MRI exams to validate modern AI algorithms and estimate radiologists’ performance at csPCa detection and diagnosis. Key aspects of the study design have been established in conjunction with an international, multi-disciplinary scientific advisory board (16 experts in prostate AI, radiology and urology) ⁠—to unify and standardize present-day guidelines, and to ensure meaningful validation of prostate-AI towards clinical translation (Reinke et al., 2022).

The 2022 edition of PI-CAI will focus on validating AI at automated 3D detection and diagnosis of csPCa in bpMRI.

PI-CAI primarily consists of two sub-studies:

• AI Study (or Grand Challenge): An annotated multi-center, multi-vendor dataset of 1500 bpMRI exams (including their basic clinical and acquisition variables) is made publicly available for all participating teams and the research community at large. Teams can use this dataset to develop AI models, and submit their trained algorithms (in Docker containers) for evaluation. At the end of this open development phase, all algorithms are ranked, based on their performance on a hidden testing cohort of 1000 unseen scans. In the closed testing phase, organizers retrain the top-ranking 5 AI algorithms using a larger dataset of 9107 bpMRI scans (including additional training scans from a private dataset). Finally, their performance is re-evaluated on the hidden testing cohort (with rigorous statistical analyses).

• Reader Study: 50+ international prostate radiologists perform a reader study using a subset of 400 scans from the hidden testing cohort. For each case, radiologists complete their assessments in two rounds. At first, using basic clinical and acquisition variables + bpMRI sequences only, enabling head-to-head comparisons against AI trained on the same. And then, using basic clinical and acquisition variables + full mpMRI sequences, enabling comparisons between AI and current clinical practice (PI-RADS v2.1). Overall, the goal of this study is to estimate the performance of the average radiologist at detection and diagnosis of csPCa in MRI.

In the end, PI-CAI aims to benchmark state-of-the-art AI algorithms developed in the grand challenge, against prostate radiologists participating in the reader study —to evaluate the clinical viability of modern prostate-AI solutions at csPCa detection and diagnosis in MRI.

Prizes 🏆

Top 5 prostate-AI teams (participating in both phases of the challenge) will be invited to join the PI-CAI consortium, and in turn, they will be listed as consortium authors on an upcoming high-impact journal paper summarizing the findings of this challenge. Furthermore, they will receive the following cash prizes, as per their ranking:

🥇1st place: €1000
🥈2nd place: €500
🥉3rd place: €250
🏵️4th place: €150
🏵️5th place: €100