Machine learning algorithms and biomarkers identification for pancreatic cancer diagnosis using multi-omics data...

Area of Science:

• Oncology
• Bioinformatics
• Artificial Intelligence

Background:

• Pancreatic cancer is a highly lethal malignancy often diagnosed at advanced stages, leading to poor prognoses.
• The development of novel diagnostic and prognostic markers is crucial for improving early detection and patient outcomes.
• Biomarkers can facilitate personalized and targeted treatment strategies, revolutionizing pancreatic cancer care.

Purpose of the Study:

• To review the application of machine learning algorithms in identifying biomarkers for pancreatic cancer.
• To assess the current landscape of AI-driven biomarker discovery in pancreatic cancer research.

Main Methods:

• A systematic literature search was conducted across PubMed, Scopus, EMBASE, and Web of Science up to September 2022, adhering to PRISMA guidelines.
• Study quality and bias were assessed using the Newcastle-Ottawa Scale (NOS).
• Descriptive statistics and quality assessments were compiled for included studies.

Main Results:

• Support Vector Machines (SVM) and Random Forests (RF) were identified as the most prevalent machine learning algorithms for biomarker discovery in pancreatic cancer.
• SVM, a supervised learning method, is utilized for classification and regression tasks by finding optimal hyperplanes in high-dimensional data.

Conclusions:

• Machine learning applications represent a significant advancement in the search for novel pancreatic cancer biomarkers.
• AI-powered biomarker discovery holds substantial promise for enhancing early detection and treatment efficacy.
• These advancements are expected to improve patient outcomes for pancreatic cancer.