Announcements & News๐ฃ¶
๐ข19-22 September 2024:ย Primary outcomes of the PI-CAI challenge and preliminary work in future follow-up projects were discussed in oral presentations during theย 30th European Symposium on Urogenital Radiology (ESUR 2024) in Lisbon, Portugal.¶
๐ขย 13 September 2024: Several editorials and perspective pieces discussing the outcomes of the PI-CAI challenge, have been published inย Nature Reviews Urology,ย The Lancet Oncology,ย European Urology,ย American Journal of Roentgenology,ย Radiology, Radiology: Imaging Cancerย andย Radiology: Artificial Intelligence.¶
๐ข 27-29 June 2024: Primary outcomes of the PI-CAI challenge were presented at the 12th European Society of Urogenital Radiology (ESUR) Prostate MRI Workshop in Zeist, The Netherlands.¶
¶
๐ข 11 June 2024: Primary outcomes of the PI-CAI challenge have been published in The Lancet Oncology: https://www.thelancet.com/journals/lanonc/article/PIIS1470-2045(24)00220-1/abstract.ย 
¶
๐ข 1 March 2024: Primary outcomes of the PI-CAI challenge and future follow-up projects were announced in oral presentations during the 2024 European Congress of Radiology (ECR 2024) in Vienna, Austria.¶
๐ข27 November 2023: Primary outcomes of the PI-CAI: Reader Study were presented in an oral presentation on 27 November, 8-9 AM CST (Session ID: M1-SSGU03), at this year's 109th Scientific Assembly and Annual Meeting of the Radiological Society of North America (RSNA 2023) in Chicago, USA.¶
¶
๐ข9 November 2023: The PI-CAI Challenge was presented, as an example of how large-scale studies can benefit from crowdsourcing research products, at this year's Clinical Radiology Research Day of The Royal College of Radiologists, UK.¶
๐ข11 October 2023: Intermediate findings from the PI-CAI challenge, comparing the performance of an ensemble of the top five AI models from the grand challenge in comparison to radiologists participating in the reader study, were presented in an oral presentation at the Annual Meeting of the Society for Advanced Body Imaging (SABI 2023) in Dallas, USA, where our abstract was awarded the "Resoundant Innovation Award".¶
¶
๐ข10 July 2023: Diagnostic performance of all state-of-the-art AI solutions from the PI-CAI challenge, when trained using >9100 MRI exams and tested across 1000 multi-center, multi-vendor patient cases (including external data), was just announced in a plenary oral presentation at the Medical Imaging with Deep Learning conference (MIDL 2023) in Nashville, USA.¶
๐ข10 March 2023: Preliminary outcomes of the PI-CAI: Reader Study were presented at the 2023 European Congress of Radiology (ECR 2023) in Vienna, Austria, for which the abstract was awarded the "Best Research Presentation Abstract" in the topic of "Oncologic Imaging". Meanwhile, an overview of our study design, key outcomes reached thus far and the roadmap ahead, were presented in the plenary "Game Changing" session at the annual congress of the 38th Annual Congress of the European Association of Urology (EAU 23) in Milan, Italy. For a summary of this talk, you can check out its featured article on UroToday: http://bit.ly/3yuyql1.¶
¶
๐ข 8 December 2022: The PI-CAI challenge has been featured in the December 2022 issue of RSIP Vision - Computer Vision News and the RSNA 2022: Daily Bulletin. Thanks to Ralph Anzarouth and Melissa Silverberg for the shout-out!¶
¶
๐ข 29 November 2022: Open Development Phase of the PI-CAI challenge has now concluded! All teams can continue to make submissions to the Open Development Phase - Tuning Leaderboard, albeit at a reduced rate (once per week per verified user). Meanwhile, the Open Development Phase - Testing Leaderboard will reopen in early 2023 and accept submissions on a per-application basis to the PI-CAI consortium via e-mail. Please check out this forum thread for more details.¶
๐ข 20 October 2022: The PI-CAI challenge will be featured in an upcoming webinar from the MICCAI Special Interest Group for Challenges, titled "How to Run a Challenge?", on 25 October 2022. See you there!¶
¶
๐ข 5 September 2022: Participants can now submit a maximum of 1 AI algorithm per day per verified user to the Open Development Phase - Tuning Leaderboard, instead of 1 AI algorithm per week per team like before. All participants must still form teams (even if the team is composed of a single participant), and each participant can only be a member of a single team. For the full list of participation rules, visit: https://pi-cai.grand-challenge.org/AIPR/ .¶
๐ข 8 August 2022: The PI-CAI challenge is now preregistered in compliance with NLM-NIH and BIAS reporting guidelines! To learn more about our end-to-end study design, visit: - clinicaltrials.gov/NCT05489341 - zenodo.org/record/6667655 (DOI:10.5281/zenodo.6667655)¶
๐ข 16 June 2022: GitHub repos with our source code for preprocessing prostate MRI data archives, training baseline diagnostic AI models, and evaluating lesion detection and diagnosis performance are now live!¶
- Preprocessing: github.com/DIAGNijmegen/picai_prep
- Baseline AI models: github.com/DIAGNijmegen/picai_baseline
- Performance evaluation: github.com/DIAGNijmegen/picai_eval