Commentary: Automated diagnosis and gleason grading of prostate cancer – are artificial intelligence systems ready for prime time?

Abstract Background Gleason grading of prostate cancer is an important prognostic factor, but suffers from poor reproducibility, particularly among non-subspecialist pathologists. Although artificial intelligence (A.I.) tools have demonstrated Gleason grading on-par with expert pathologists, it remains an open question whether and to what extent A.I. grading translates to better prognostication. Methods In this study, we developed a system to predict prostate cancer-specific mortality via A.I.-based Gleason grading and subsequently evaluated its ability to risk-stratify patients on an independent retrospective cohort of 2807 prostatectomy cases from a single European center with 5–25 years of follow-up (median: 13, interquartile range 9–17). Results Here, we show that the A.I.’s risk scores produced a C-index of 0.84 (95% CI 0.80–0.87) for prostate cancer-specific mortality. Upon discretizing these risk scores into risk groups analogous to pathologist Grade Groups (GG), the A.I. has a C-index of 0.82 (95% CI 0.78–0.85). On the subset of cases with a GG provided in the original pathology report (n = 1517), the A.I.’s C-indices are 0.87 and 0.85 for continuous and discrete grading, respectively, compared to 0.79 (95% CI 0.71–0.86) for GG obtained from the reports. These represent improvements of 0.08 (95% CI 0.01–0.15) and 0.07 (95% CI 0.00–0.14), respectively. Conclusions Our results suggest that A.I.-based Gleason grading can lead to effective risk stratification, and warrants further evaluation for improving disease management.

Download Full-text

On modern methods of automated diagnosis of skin tumors in clinical practice

Medical alphabet ◽

10.33667/2078-5631-2020-6-76-78 ◽

2020 ◽

pp. 76-78

Author(s):

V. Yu. Sergeev ◽

Yu. Yu. Sergeev ◽

O. B. Tamrazova ◽

V. G. Nikitaev ◽

A. N. Pronichev

Keyword(s):

Public Health ◽

Artificial Intelligence ◽

Clinical Practice ◽

Public Health Service ◽

Automated Diagnosis ◽

Skin Cancers ◽

Modern Methods ◽

Number Of Publications ◽

New Algorithms ◽

Artificial Intelligence Systems

Despite the existence of many algorithms for automated diagnosis of melanoma and other skin cancers, these remain almost inaccessible to public health service. A small number of publications on the efficacy of existing artificial intelligence systems marks the problems of their implementation into current examination routines in dermatology and oncology. New algorithms and software solutions as well as studies demonstrating their diagnostic accuracy on compatible and verifiable clinical material are still in demand.

Download Full-text

Artificial Intelligence for Diagnosis and Gleason Grading of Prostate Cancer in Biopsies—Current Status and Next Steps

European Urology Focus ◽

10.1016/j.euf.2021.07.002 ◽

2021 ◽

Author(s):

Kimmo Kartasalo ◽

Wouter Bulten ◽

Brett Delahunt ◽

Po-Hsuan Cameron Chen ◽

Hans Pinckaers ◽

...

Keyword(s):

Prostate Cancer ◽

Artificial Intelligence ◽

Current Status ◽

Gleason Grading

Download Full-text

MP20-04 AUTOMATED DIAGNOSIS OF PROSTATE CANCER LOCATION AND FORM BY ARTIFICIAL INTELLIGENCE IN MULTIPARAMETRIC MRI

The Journal of Urology ◽

10.1016/j.juro.2018.02.674 ◽

2018 ◽

Vol 199 (4S) ◽

Author(s):

Yuichiro Oishi ◽

Takeya Kitta ◽

Nobuo Shinohara ◽

Hirokazu Nosato ◽

Hidenori Sakanashi ◽

...

Keyword(s):

Prostate Cancer ◽

Artificial Intelligence ◽

Multiparametric Mri ◽

Automated Diagnosis

Download Full-text

Artificial intelligence for Gleason grading in prostate cancer

European Urology Supplements ◽

10.1016/s1569-9056(19)33329-9 ◽

2019 ◽

Vol 18 (8) ◽

pp. e3082

Author(s):

A. Krzyzanowska ◽

F.E. Marginean ◽

I. Arvidsson ◽

A. Simoulis ◽

N.C. Overgaard ◽

...

Keyword(s):

Prostate Cancer ◽

Artificial Intelligence ◽

Gleason Grading

Download Full-text

Artificial intelligence for diagnosis and Gleason grading of prostate cancer: the PANDA challenge

Nature Medicine ◽

10.1038/s41591-021-01620-2 ◽

2022 ◽

Author(s):

Wouter Bulten ◽

Kimmo Kartasalo ◽

Po-Hsuan Cameron Chen ◽

Peter Ström ◽

Hans Pinckaers ◽

...

Keyword(s):

Prostate Cancer ◽

Artificial Intelligence ◽

United States ◽

Clinical Trials ◽

External Validation ◽

Gleason Grading ◽

Independent Validation ◽

Prostate Biopsies ◽

Algorithmic Approaches ◽

Level Performance

AbstractArtificial intelligence (AI) has shown promise for diagnosing prostate cancer in biopsies. However, results have been limited to individual studies, lacking validation in multinational settings. Competitions have been shown to be accelerators for medical imaging innovations, but their impact is hindered by lack of reproducibility and independent validation. With this in mind, we organized the PANDA challenge—the largest histopathology competition to date, joined by 1,290 developers—to catalyze development of reproducible AI algorithms for Gleason grading using 10,616 digitized prostate biopsies. We validated that a diverse set of submitted algorithms reached pathologist-level performance on independent cross-continental cohorts, fully blinded to the algorithm developers. On United States and European external validation sets, the algorithms achieved agreements of 0.862 (quadratically weighted κ, 95% confidence interval (CI), 0.840–0.884) and 0.868 (95% CI, 0.835–0.900) with expert uropathologists. Successful generalization across different patient populations, laboratories and reference standards, achieved by a variety of algorithmic approaches, warrants evaluating AI-based Gleason grading in prospective clinical trials.

Download Full-text

Diagnostic performance study on the melanoma automated diagnosis software powered by artificial intelligence technologies

Russian Journal of Skin and Venereal Diseases ◽

10.17816/dv52794 ◽

2020 ◽

Vol 23 (5) ◽

pp. 288-292

Author(s):

Vasiliy Yu. Sergeev ◽

Yu. Yu. Sergeev ◽

O. B. Tamrazova ◽

V. G. Nikitaev ◽

A. N. Pronichev ◽

...

Keyword(s):

Artificial Intelligence ◽

Diagnostic Performance ◽

High Sensitivity ◽

Skin Neoplasms ◽

Skin Tumors ◽

Published Data ◽

Performance Study ◽

Automated Diagnosis ◽

Accuracy Rate ◽

Artificial Intelligence Systems

INTRODUCTION: The research evaluates a series of publications on the machine recognition efficacy of cutaneous melanoma dermatoscopic images. Some authors report high sensitivity and specificity of automated diagnostics of skin tumors. Significant differences in the published data can be attributed to the use of different algorithms and groups of skin neoplasms to calculate the accuracy rate. MATERIALS AND METHODS: The diagnostic performance of two automated artificial intelligence systems is compared. RESULTS: The convolutional neural network algorithm improves the overall diagnostic accuracy by 7% compared to the algorithm without deep learning, while the overall accuracy rate was 78%. An initial set of 100 dermatoscopic images used in the study is published online for the assessment of the applicability of the obtained data when introducing existing artificial intelligence systems. CONCLUSION: The main limitations and possible ways to further improve the automated diagnosis of skin tumors based on digital dermatoscopy are outlined.

Download Full-text

Artificial intelligence as a moving force of improvement and innovative development in education and pedagogic

Informatics and Education ◽

10.32517/0234-0453-2018-33-10-21-30 ◽

2019 ◽

pp. 21-30

Author(s):

M. G. Koliada ◽

T. I. Bugayova

Keyword(s):

Artificial Intelligence ◽

Intelligent Systems ◽

Intelligent System ◽

Interaction Analysis ◽

Educational Data Mining ◽

Main Task ◽

Intellectual System ◽

Will Force ◽

To Come ◽

Artificial Intelligence Systems

The article discusses the history of the development of the problem of using artificial intelligence systems in education and pedagogic. Two directions of its development are shown: “Computational Pedagogic” and “Educational Data Mining”, in which poorly studied aspects of the internal mechanisms of functioning of artificial intelligence systems in this field of activity are revealed. The main task is a problem of interface of a kernel of the system with blocks of pedagogical and thematic databases, as well as with the blocks of pedagogical diagnostics of a student and a teacher. The role of the pedagogical diagnosis as evident reflection of the complex influence of factors and reasons is shown. It provides the intelligent system with operative and reliable information on how various reasons intertwine in the interaction, which of them are dangerous at present, where recession of characteristics of efficiency is planned. All components of the teaching and educational system are subject to diagnosis; without it, it is impossible to own any pedagogical situation optimum. The means in obtaining information about students, as well as the “mechanisms” of work of intelligent systems based on innovative ideas of advanced pedagogical experience in diagnostics of the professionalism of a teacher, are considered. Ways of realization of skill of the teacher on the basis of the ideas developed by the American scientists are shown. Among them, the approaches of researchers D. Rajonz and U. Bronfenbrenner who put at the forefront the teacher’s attitude towards students, their views, intellectual and emotional characteristics are allocated. An assessment of the teacher’s work according to N. Flanders’s system, in the form of the so-called “The Interaction Analysis”, through the mechanism of fixing such elements as: the verbal behavior of the teacher, events at the lesson and their sequence is also proposed. A system for assessing the professionalism of a teacher according to B. O. Smith and M. O. Meux is examined — through the study of the logic of teaching, using logical operations at the lesson. Samples of forms of external communication of the intellectual system with the learning environment are given. It is indicated that the conclusion of the found productive solutions can have the most acceptable and comfortable form both for students and for the teacher in the form of three approaches. The first shows that artificial intelligence in this area can be represented in the form of robotized being in the shape of a person; the second indicates that it is enough to confine oneself only to specially organized input-output systems for targeted transmission of effective methodological recommendations and instructions to both students and teachers; the third demonstrates that life will force one to come up with completely new hybrid forms of interaction between both sides in the form of interactive educational environments, to some extent resembling the educational spaces of virtual reality.

Download Full-text