Tag Archive for: AI

Posts

Editorial: Can artificial intelligence optimize case selection for hemi‐gland ablation?

The victory of ‘AlphaGo’ over humans in Go, one of the most complex games with more than 10170 board configurations, has yielded tremendous attention worldwide [1]. The later version, ‘AlphaGo Zero’, has brought artificial intelligence (AI) to the next level by demonstrating an absolute superiority, winning 100‐0 against the champion‐defeating AlphaGo [2]. It is exciting, and perhaps shocking, to realize what AI can achieve.

In this issue of BJUI, the study by Zhou et al. [3] is the first to utilize AI to optimize case selection for hemi‐gland ablation. In this study, classification and regression tree (CART) analysis, which is a form of supervised machine‐learning algorithm, was used to identify laterality of prostate cancer. In the conventional approach, case selection was based on biopsy results and MRI findings. For the CART model, in addition to the common clinical variables (i.e. age, PSA, prostate volume, biopsy and MRI results), biopsy coordinate‐derived spatial features were also used as model inputs. The model output was the probability of unilateral clinically significant prostate cancer considered suitable for hemi‐gland ablation. Whole‐mount prostatectomy specimens were used as the standard of reference. The CART model correctly identified laterality in 80% of the cases, compared to 73% with the conventional approach. The positive predictive value of the CART model was 83%, compared to 53% with the conventional approach. The superiority of the CART model has been demonstrated, and the area under curve was 0.82.

Artifical intelligence has been widely adopted in the field of Urology [4]. For prostate cancer detection in particular, our group evaluated the diagnostic performances of four machine‐learning models based on clinical variables in a biopsy cohort of 1625 men [5]. The machine‐learning models achieved excellent performances in detecting clinically significant prostate cancer, with an accuracy of up to 95.3%. Algohary et al. [6] constructed three machine‐learning models to identify the presence of clinically significant prostate cancer based on MRI radiomic features in patients who underwent active surveillance. When compared with the Prostate Imaging–Reporting and Data System (PI‐RADS) scoring system, the machine‐learning models were able to improve overall accuracy by 30–80%.

Fehr et al. [7] developed an automated system to classify Gleason scores based on MRI images. The automated system could distinguish between Gleason scores of 6 and 7 or above cancers with an accuracy of up to 93%. The differentiation between Gleason score 3+4 and 4+3 disease also yielded an accuracy of up to 93%. Importantly, the performance of AI and machine‐learning models is highly dependent on the quality and accuracy of the data being input. In terms of prostate cancer detection, either mapping biopsy or whole‐mount prostatectomy specimens should be considered to represent the ‘ground truth’.

There are a number of challenges in implementing AI in clinical practice. First, decision‐making in healthcare requires logical deduction and explanation. The data processing in AI, however, is often described as a ‘black box’. Taking AlphaGo as an example, some ‘moves’ were considered incomprehensible even by world‐class players. Second, although results from AI are promising, there is in general a lack of regulations and standards to assess its safety, efficacy and validity. Liability issues can be problematic in case of medical mishaps. Third, doctors are human. Conflict of interest does exist, and how we can utilize AI in a complementary rather than a competitive manner is a challenging obstacle to overcome.

Nevertheless, AI has huge potential in improving healthcare. Collaborative effort is needed globally to develop and optimize AI systems, and to increase its acceptability and practicality upon implementation. Future studies answering clinically important questions using appropriate standards of reference will be of paramount importance in paving the way for the AI era in urology.

by Jeremy Yuen‐Chun Teoh, Edmund Chiong and Chi‐Fai Ng

References

  1. Silver DHuang AMaddison CJ et al. Mastering the game of Go with deep neural networks and tree search. Nature 2016529484– 9
  2. Silver DSchrittwieser JSimonyan K et al. Mastering the game of Go without human knowledge. Nature 2017550354– 9
  3. Zhou SRPriester AMJayadevan R et al. Using spatial tracking with magnetic resonance imaging/ultrasound‐guided biopsy to identify unilateral prostate cancer. BJU Int 2020125399– 406
  4. Chen JRemulla DNguyen JH et al. Current status of artificial intelligence applications in urology and their potential to influence clinical practiceBJU Int 2019124567– 77
  5. Wang GTeoh JYChoi KSDiagnosis of prostate cancer in a Chinese population by using machine learning methods. Conf Proc IEEE Eng Med Biol Soc 201820181– 4
  6. Algohary AViswanath SShiradkar R et al. Radiomic features on MRI enable risk categorization of prostate cancer patients on active surveillance: Preliminary findings. J Magn Reson Imaging 201848818– 28
  7. Fehr DVeeraraghavan HWibmer A et al. Automatic classification of prostate cancer Gleason scores from multiparametric magnetic resonance images. Proc Natl Acad Sci USA 2015112E6265– 73

 

Science, technology and artificial intelligence

As the year comes to a close, it is time to reflect fondly on the revolutionary reports in the world of scientific publishing. To me, the most exciting were the findings from the Cassini spacecraft diving within Saturn’s rings before destroying itself in its upper atmosphere. This so‐called ‘Cassini Grand Finale’ had begun with the launch of the spacecraft over 20 years ago with the hope of finding subsurface water and potentially habitable environments on Saturn’s moons [1]. Our search for intelligent life continues, driven by advances in new technology. Back on earth, modern microscopy can allow single molecules to be observed and genomes can be precisely manipulated by Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)‐mediated gene editing. The handling of the large data that are generated is likely to be enhanced by the ever‐evolving role of artificial intelligence (AI) [2]. Our New York Dedicated Servers come wіth a 100% network uptime SLA tо dеlіvеr a rеlіаblе dedicated ѕеrvеr hоѕtіng experience fоr уоur buѕіnеѕѕ.

This is the year when we have heard more about AI within the surgical community than any other [3]. Most of us carry AI devices in our pockets in the form of our mobile phones. How can we use this to our benefit perhaps during the few minutes that we have between cases on a busy urological operating list? My usual trick is to ask ‘Siri’ (Speech Interpretation and Recognition Interface) on my iPhone® (Apple Inc., Cupertino, CA, USA) to play me a BJUI podcast, which provides me with a summary of a new paper without having to read any text. Many have told me that listening is becoming as fashionable as reading text, and this is one of our attempts at using AI to augment the BJUI experience.

We also set ourselves the target of becoming one of the first journals to embrace and embed AI. With this in mind, I requested Andrew Hung from California to join the BJUI as Consulting Editor for AI. Andrew has already been publishing novel and often paradoxical reports on surgical performance based on automated performance metrics. A team from Canada has found that machine‐learning (a subset of AI) algorithms can predict biochemical recurrence after radical prostatectomy more accurately than traditional statistical modelling [4]. While being excited by these results, Hung [5] reminds us that this needs to be validated externally in a larger patient population before it is ready for prime time. Next year we hope to report more from the world of AI and perhaps even surprise our readers with embedded technology within the BJUI itself.

With such rapid advances in science and technology comes the description of a new kind of education for our generation and the next. Joseph Aoun [6], who leads Northeastern University, describes this as ‘Humanics’ in his new book on higher education in the age of AI. It involves the fundamental difference between what machines and AI can do better than humans but equally what humans do better than machines. This book is a must‐read, as it describes the pillars of technological, data and human literacy. So much so that I have started advising my scientifically minded students and colleagues to consider participating in short boot camps on data science.

I wish you all, wherever you are and whatever the weather, much happiness and greetings of the season!

Prokar Dasgupta

Editor-in-Chief, BJUI

References

Dougherty MK, Cao H, Khurana KK et al. Saturn’s magnetic field revealed by the Cassini Grand Finale. Science 2018362: 5434

Mao S, Vinson V. Power couple: science and technologyScience 2018361: 864–5

Dasgupta P. New robots – cost, connectivity and artificial intelligenceBJU Int 2018122: 349–50

Wong NC, Lam C, Patterson L, Shayegan B. Use of machine learning to predict early biochemical recurrence after robot‐assisted prostatectomy. BJU Int 2018.

Hung A. Can machine learning algorithms replace conventional statistics? BJU Int 2019

Aoun JE. Robot‐Proof: Higher Education in the Age of Artificial Intelligence. Cambridge, MA: The MIT Press, 2017

© 2020 BJU International. All Rights Reserved.