Page 14 - AIH-1-1

Artificial Intelligence in Health AI in prostate cancer detection

et al. described how ML contributes to differentiating highlighted blue circles at each year correspond to the
[39]
between the transitional and peripheral zones. trending topics of that year. In 2016, the trending topics
included decision trees, histology, artificial neural networks
5. Keyword Analysis (ANNs), reproducibility, and image analysis. During this
This section entails a keyword analysis. An overlay period, decision trees and neural networks were applied in
network analysis of keywords, as depicted in Figure 5, prostate cancer classification and detection. The following
reveals the trends observed in recently published research. year, research focus shifted toward image interpretation,
Keywords such as “guided biopsy,” “radiomics,” and diagnosis, computer-assisted methods, and sensitivity and
“cancer grading” are emerging prominently in the recently specificity, undoubtedly contributing to improved cancer
diagnostics. Subsequently, the research evolved to encompass
published literature. To delve further into evolving MRI, algorithm, priority journal, prostatic neoplasms, and
topics, we conducted a detailed exploration of the trends humans. As new technologies and their application in the
reported in Figure 6. Emerging keywords include “feature medical field evolved, we observed ML, prostate cancer
selection,” “genomics,” “transrectal ultrasound-guided diagnostic accuracy, prostate biopsy, and controlled studies
biopsy,” “quality of life and metabolomics,” and “diagnostic gaining prominence in prostate cancer research.
accuracy,” among others.

Trending, in general, provides an understanding of 6. Historiographical analysis
a pattern based on the provided information. Figure 6 A historiographical analysis is presented through a direct
illustrates the trending topics from the years 2016 citation network . Figure 7 illustrates the historical direct
[28]
through 2020 in the prostate cancer research domain. The citation network from the year 2016 onward. Five paths
Table 3. Co‑citation network analysis

Cluster theme (s) Notable references in clusters Cluster
[13]
AI application in MRI Kasivisvanathan et al. ; Bonekamp et al. ; Schelb et al. [51] 1 (Red)
[35]
[53]
[52]
Prostate cancer grading systems Epstein et al. ; Fehr et al. ; Epstein et al. [54] 2 (Green)
AI applications in prostate cancer classification using MRI Wibmer et al. ; Weinreb et al. ; Mottet et al. [16] 3 (Blue)
[55]
[56]
AI applications in prostate cancer pathology Nir et al. ; Campanella et al. ; Matoso and Epstein [58] 4 (Yellow)
[57]
[36]
Deep learning in cancer diagnostics Wason ; Turkbey et al. (2011) ; Song et al. [61] 5 (Purple)
[60]
[59]
Abbreviations: AI: Artificial intelligence; MRI: Magnetic resonance imaging.
Table 4. Citation network analysis
Cluster theme (s) References in the citation network Cluster
AI applications of prostate cancer detection using MRI Litjens et al. ; Artan et al. ; Tiwari et al. [44] 1 (Red)
[62]
[30]
AI applications in prostate cancer classification Freedman et al. ; Ozer et al. ; Zhan et al. [64] 2 (Green)
[47]
[63]
[38]
[32]
Multi-feature prostate cancer detection Tabesh et al. ; Monaco et al. ; Gorelick et al. [37] 3 (Blue)
Machine learning in radiomics Wang et al. ; Bonekamp et al. ; Ginsburg et al. [39] 4 (Yellow)
[35]
[31]
Abbreviations: AI: Artificial intelligence; MRI: Magnetic resonance imaging.

Figure 4. Co-citation network.

Volume 1 Issue 1 (2024) 8 https://doi.org/10.36922/aih.1958