Yet Another Automated Gleason Grading System (YAAGGS) by weakly supervised deep learning

AbstractThe Gleason score contributes significantly in predicting prostate cancer outcomes and selecting the appropriate treatment option, which is affected by well-known inter-observer variations. We present a novel deep learning-based automated Gleason grading system that does not require extensive region-level manual annotations by experts and/or complex algorithms for the automatic generation of region-level annotations. A total of 6664 and 936 prostate needle biopsy single-core slides (689 and 99 cases) from two institutions were used for system discovery and validation, respectively. Pathological diagnoses were converted into grade groups and used as the reference standard. The grade group prediction accuracy of the system was 77.5% (95% confidence interval (CI): 72.3–82.7%), the Cohen’s kappa score (κ) was 0.650 (95% CI: 0.570–0.730), and the quadratic-weighted kappa score (κquad) was 0.897 (95% CI: 0.815–0.979). When trained on 621 cases from one institution and validated on 167 cases from the other institution, the system’s accuracy reached 67.4% (95% CI: 63.2–71.6%), κ 0.553 (95% CI: 0.495–0.610), and the κquad 0.880 (95% CI: 0.822–0.938). In order to evaluate the impact of the proposed method, performance comparison with several baseline methods was also performed. While limited by case volume and a few more factors, the results of this study can contribute to the potential development of an artificial intelligence system to diagnose other cancers without extensive region-level annotations.

Download Full-text

Impact of Video Compression and Multimodal Embedding on Scene Description

Electronics ◽

10.3390/electronics8090963 ◽

2019 ◽

Vol 8 (9) ◽

pp. 963

Author(s):

Jin Young Lee

Keyword(s):

Deep Learning ◽

Video Compression ◽

Automatic Generation ◽

Image Motion ◽

Label Information ◽

Scene Description ◽

The Impact ◽

Potential Issue ◽

Simple Network ◽

Compression Artifacts

Scene description refers to the automatic generation of natural language descriptions from videos. In general, deep learning-based scene description networks utilize multimodalities, such as image, motion, audio, and label information, to improve the description quality. In particular, image information plays an important role in scene description. However, scene description has a potential issue, because it may handle images with severe compression artifacts. Hence, this paper analyzes the impact of video compression on scene description, and then proposes a simple network that is robust to compression artifacts. In addition, a network cascading more encoding layers for efficient multimodal embedding is also proposed. Experimental results show that the proposed network is more efficient than conventional networks.

Download Full-text

Intraductal Carcinoma of Prostate (IDC-P), Grade Group, and Molecular Pathology: Recent Advances and Practical Implication

annals of urologic oncology ◽

10.32948/auo.2019.03.11 ◽

2019 ◽

pp. 1-10

Author(s):

Ashwyna Sunassee ◽

Ghadah Al Sannaa ◽

Jae Y. Ro

Keyword(s):

Prostate Cancer ◽

Molecular Pathology ◽

Invasive Carcinoma ◽

World Health ◽

Intraductal Carcinoma ◽

Grading System ◽

Grade Group ◽

Group System ◽

Gleason Grading ◽

Recent Advances

The Gleason grading system for prostatic carcinoma is widely used internationally and is based on microscopic architectural patterns of tumors. Over the years, there have been modifications to the original grading system established by Donald F Gleason in 1966 and refined in 1974 which have subsequently been established by the World Health Organization in its WHO Classification of Tumors of the Urinary System and Male Genital Organs book, published in 2016. There have been certain practical issues associated with the changes, of note, the addition of intraductal carcinoma of prostate (IDC-P), which unlike its breast counterpart rarely occurs in isolation without association with invasive carcinoma and tends to be associated with high-grade invasive carcinoma. In addition, the Grade group system has been introduced which categorizes tumors into prognostically relevant groups based on the histological grade scores. The grade group system brings to light the importance of making accurate scoring and subsequent grouping of the tumors as it affects the clinical treatment, prognostic implication and stage assignment. Molecular pathology of the prostate is not widely utilized in clinical practice, but is emerging. The most common genomic aberration in prostate cancer includes gene fusion, amplification, deletion, and mutation. In addition, up and down regulation of gene expression in critical cellular pathways is also at play. A series of long noncoding RNA expression changes have been also unveiled from transcriptome sequencing data. They play a regulatory role in prostate cancer and are promising diagnostic and potentially prognostic markers as well as molecular treatment strategy. In this review, we summarize recent advances in molecular pathology of prostate cancer and their emerging clinical utility with currently available molecular tests. In this review article, we discuss the followings: 1) Gleason grading system with its modification, 2) Grade group, 3) Intraductal carcinoma, and 4) molecular pathology. Additionally, we present that molecular studies continue to emerge, and there is significant opportunity for targeted therapeutic options that remains to be explored in depth.

Download Full-text

Contemporary Grading of Prostate Cancer: The Impact of Grading Criteria and the Significance of the Amount of Intraductal Carcinoma

Cancers ◽

10.3390/cancers13215454 ◽

2021 ◽

Vol 13 (21) ◽

pp. 5454

Author(s):

Vasiliki Tzelepi ◽

Ioanna Maria Grypari ◽

Souzana Logotheti ◽

Stavros Kontogiannis ◽

Panagiotis Kallidonis ◽

...

Keyword(s):

Prostate Cancer ◽

Prognostic Significance ◽

Intraductal Carcinoma ◽

Grading System ◽

Grade Group ◽

Prognostic Parameter ◽

Therapeutic Decisions ◽

Histopathologic Analysis ◽

The Impact ◽

Potential Area

(1) Background: Prognostic grade group (PGG) is an important prognostic parameter in prostate cancer that guides therapeutic decisions. The cribriform pattern and intraductal carcinoma (IDC) are two histological patterns, that have additional prognostic significance. However, discrepancies exist regarding the handling of IDC according to the guidelines published by two international genitourinary pathology societies. Furthermore, whether, in addition to its presence, the amount of IDC is also of importance has not been studied before. Lastly, the handling of tertiary patterns has also been a matter of debate in the literature. (2) Methods: A total of 129 prostatectomy cases were retrieved and a detailed histopathologic analysis was performed. (3) Results: Two cases (1.6%) upgraded their PGG, when IDC was incorporated in the grading system. The presence and the amount of IDC, as well as the presence of cribriform carcinoma were associated with adverse pathologic characteristics. Interestingly, in six cases (4.7%) there was a difference in PGG when using the different guidelines regarding the handling of tertiary patterns. In total, 6.2% of the cases would be assigned a different grade depending on the guidelines followed. (4) Conclusions: These findings highlight a potential area of confusion among pathologists and clinicians and underscore the need for a consensus grading system.

Download Full-text

Understanding the gleason grading system and its changes

Journal of Pathology of Nepal ◽

10.3126/jpn.v9i2.25723 ◽

2019 ◽

Vol 9 (2) ◽

pp. 1580-1585

Author(s):

Sujata Pudasaini ◽

Neeraj Subedi

Keyword(s):

Prostatic Carcinoma ◽

World Health ◽

Grading System ◽

Pathology Report ◽

Gleason Grade ◽

High Grade ◽

Grade Group ◽

Gleason Grading ◽

Gleason Pattern ◽

Health Organization

Gleason Grading System is the most widely used grading system used for prostatic carcinoma. The five basic grade patterns are used to generate a histologic score, which can range from 2 to 10 (including primary and secondary patterns). The original Gleason Grading System was used to grade acinar adenocarcinoma based on architectural features and it has been correlated with excellent clinical outcomes. Since 1960s, after the discovery of the original Gleason Grading System, a modified version of the Gleason Grading System was introduced in the International Society of Urological Pathology 2005 which came up with many changes including elimination of Gleason pattern 1. The ISUP 2005 was further updated in 2014 to provide more accurate stratification of prostatic carcinoma. The new Gleason Grade Group 1 to 5 has been introduced and it has little resemblance to the original Gleason system. This Gleason Grade Group has been accepted by the 2016 World Health Organization classification of tumors of the prostate. For a needle biopsy, high grade component of any quantity should be included in the Gleason score as it indicates a high probability of finding significant high grade tumor in the prostate. By understanding the principles and practice of this grading system, the pathology report has to clearly indicate which system is adopted in the reporting. This review discusses GGS and its recent development focusing on major changes over the years that led to the new Grade Group system proposed by the 2014 ISUP consensus.

Download Full-text

KORELASI ANTARA EKSPRESI Her-2 DAN Ki-67 DENGAN GLEASON GRADE GROUP PADA ADENOKARSINOMA ASINAR PROSTAT

Jurnal Kedokteran RAFLESIA ◽

10.33369/juke.v5i1.9125 ◽

2019 ◽

Vol 5 (1) ◽

pp. 39-52

Author(s):

Putri Ajeng ayu Larasati

Keyword(s):

Tumor Suppressor Genes ◽

Grading System ◽

Gleason Grade ◽

Grade Group ◽

Ki 67 ◽

Cross Sectional ◽

Gleason Grading ◽

Her 2 ◽

Group 5 ◽

Group 1

Latar Belakang : Menurut Database GLOBOCAN 2012, kanker prostat merupakan penyebab utama kelima kematian pada pria.(1) Pemahaman tentang peran onkogen dan tumor suppressor genes mendominasi penelitian tentang biologi kanker saat ini dan berpotensi menghasilkan target terapi kanker terbaru. Salah satu perannya pada adenokarsinoma prostat yang masih belum jelas ialah Her-2 dan Ki-67. Perbedaan ekspresi profil molekular Her-2 dan Ki-67 yang diklasifikasikan berdasarkan Gleason grading system terbaru yaitu Grade group diharapkan membantu penentuan prognosis dan manajemen terapi penyakit pada kanker prostat.Tujuan : Tujuan dari penelitian ini adalah untuk mengetahui korelasi antara ekspresi Her-2 dan Ki-67 dengan Gleason Grade group adenokarsinoma asinar prostat.Metode: Merupakan penelitian observatif analitik dengan desain cross-sectional menggunakan 31 blok parafin yang terfiksasi formalin dari laboratorium Patologi Anatomik RSUP dr. Kariadi, dengan diagnosis adenokarsinoma asinar prostat kemudian diklasifikasikan menurut kelompok prognostik WHO / ISUP Gleason (Gleason Grade group 1 - 5). Pemeriksaan imunohistokimia dilakukan dengan menggunakan antibodi Her-2 dan Ki-67. Analisa hasil menggunakan uji Spearmans dan uji Kruskal WallisHasil : 31 kasus yang termasuk dalam kriteria inklusi pada penelitian ini, hampir separuhnya (41.94%) termasuk ke dalam Gleason grade group 5. Ekspresi Her-2 positif (+1, +2, +3) hampir didapatkan pada seluruh sampel (90,22%), dengan Her-2 +3 sebagian besar didapatkan pada grade group 5 (71,43%). Ekspresi Ki-67 positif (+1, +2, +3) didapatkan pada seluruh sampel, dengan Ki-67 +3 (>5%) pada Gleason grade group 5 didapatkan sejumlah 5 sampel (38,46%). Pada analisa data dengan uji Spearmans disimpulkan bahwa terdapat korelasi positif antara ekspresi Her-2 dan Ki-67 dengan Gleason Grade group. Pada uji Kruskal Wallis didapatkan perbedaan bermakna antara ekspresi Ki-67 dengan Gleason Grade group.Kesimpulan: Analisa ekspresi Her-2 dan Ki-67 melalui pemeriksaan imunohistokimia dengan klasifikasi terbaru Gleason Grade group dapat menjadi salah satu parameter prognosis dan manajemen terapi pada adenokarsinoma asinar prostat.Kata kunci : Adenokarsinoma asinar prostat, Her-2, Ki-67, Gleason Grade group

Download Full-text

Prostate cancer grading in 2018: limitations, implementations, cribriform morphology, and biological markers

The International Journal of Biological Markers ◽

10.1177/1724600818781296 ◽

2018 ◽

Vol 33 (4) ◽

pp. 331-334 ◽

Cited By ~ 3

Author(s):

Rodolfo Montironi ◽

Alessia Cimadamore ◽

Liang Cheng ◽

Antonio Lopez-Beltran ◽

Marina Scarpelli

Keyword(s):

Prostate Cancer ◽

Point Of View ◽

The Other ◽

Grading System ◽

Grade Group ◽

Gleason Grading ◽

Gleason Pattern ◽

Consensus Conferences ◽

And Biological Markers ◽

Cancer Grading

The Gleason grading system is among the most important prognostic factors in patients with prostate cancer. From the 2005 to the 2014 consensus conferences, organized by the International Society of Urological Pathology, the morphologic criteria for the identification of the Gleason patterns were redefined, thus resulting in the shrinkage of the Gleason pattern 3. This led to the expansion of the Gleason pattern 4. The newly proposed grade group system reduces the Gleason scores of prostate cancer to the lowest number, each associated with a unique behavior from the prognostic point of view. The advantage is that the simplified system with five groups allows for a more accurate stratification of the patients in comparison with the Gleason system. Cribriform, fused, ill-defined and glomeruloid glands are part of the histologic spectrum of the Gleason pattern 4. Cribriform morphology has a prognosis that is worse in comparison with the other non-cribriform Gleason 4 patterns. One of the major implications of the cribriform growth is that it precludes a patient from choosing active surveillance.

Download Full-text