-
Frontiers in Oncology 2024The risk that a large polyp (≥10 mm) evolves into high-grade dysplasia (HGD) is relatively high compared with that of a small/diminutive polyp (<10 mm). Recently, the...
INTRODUCTION
The risk that a large polyp (≥10 mm) evolves into high-grade dysplasia (HGD) is relatively high compared with that of a small/diminutive polyp (<10 mm). Recently, the detection of small and diminutive polyps has been substantially improved with the advancement of endoscopy. However, further research is needed on the role of the incidence of HGD caused by the co-occurrence of small and diminutive polyps in the progression of HGD. In this study, we aim to investigate whether and how the small and diminutive polyps correlate with the incidence of HGD in the population.
METHODS
The pooled data were deeply analyzed from four published randomized controlled trials (RCTs) regarding colon polyp detection. All polyps detected were examined and confirmed by pathologists. The primary outcome was the composition ratio of the HGD polyps in each polyp size category.
RESULTS
Among a total of 3,179 patients with 2,730 polyps identified, there were 83 HGD polyps confirmed, and 68 patients had at least one polyp with HGD. The risk of development of HGD was lower for a single small and diminutive polyp than for one large polyp (2.18% vs. 22.22%, < 0.0001). On the contrary, the composition ratio for HGD from small and diminutive polyps was significantly higher than that from the large ones (68.67% vs. 31.33%, < 0.0001). The combined number of HGD presented a trend negatively correlated to size.
CONCLUSIONS
Our data demonstrated that the absolute number of HGD significantly derives more from small and diminutive polyps than from the large ones, and the collective number of small and diminutive polyps per patient is indicative of his/her HGD exposure. These findings positively provide novel perspectives on the management of polyps and may further optimize the prevention of colorectal cancer.
SYSTEMATIC REVIEW REGISTRATION
http://www.chictr.org.cn, identifier ChiCTR1900025235, ChiCTR1800017675, ChiCTR1800018058, and ChiCTR1900023086.
PubMed: 38410098
DOI: 10.3389/fonc.2024.1294745 -
Archives of Pathology & Laboratory... May 2024Automated prostate cancer detection using machine learning technology has led to speculation that pathologists will soon be replaced by algorithms. This review covers...
CONTEXT
Automated prostate cancer detection using machine learning technology has led to speculation that pathologists will soon be replaced by algorithms. This review covers the development of machine learning algorithms and their reported effectiveness specific to prostate cancer detection and Gleason grading.
OBJECTIVE
To examine current algorithms regarding their accuracy and classification abilities. We provide a general explanation of the technology and how it is being used in clinical practice. The challenges to the application of machine learning algorithms in clinical practice are also discussed.
DATA SOURCES
The literature for this review was identified and collected using a systematic search. Criteria were established prior to the sorting process to effectively direct the selection of studies. A 4-point system was implemented to rank the papers according to their relevancy. For papers accepted as relevant to our metrics, all cited and citing studies were also reviewed. Studies were then categorized based on whether they implemented binary or multi-class classification methods. Data were extracted from papers that contained accuracy, area under the curve (AUC), or κ values in the context of prostate cancer detection. The results were visually summarized to present accuracy trends between classification abilities.
CONCLUSIONS
It is more difficult to achieve high accuracy metrics for multiclassification tasks than for binary tasks. The clinical implementation of an algorithm that can assign a Gleason grade to clinical whole slide images (WSIs) remains elusive. Machine learning technology is currently not able to replace pathologists but can serve as an important safeguard against misdiagnosis.
Topics: Humans; Prostatic Neoplasms; Male; Machine Learning; Artificial Intelligence; Neoplasm Grading; Algorithms
PubMed: 37594900
DOI: 10.5858/arpa.2022-0460-RA