Identification of Tumor-Specific MRI Biomarkers Using Machine Learning (ML)

T1-mapping characterization of two tumor types

T1 mapping is a quantitative method to characterize tissues with magnetic resonance imaging in a quick and efficient manner. It utilizes the relaxation rate of protons to depict the underlying structures within the imaging frame. While T1-mapping techniques are used with some frequency in areas such as cardiac imaging, their application for understanding malignancies and identifying tumor structures has yet to be thoroughly investigated. Utilizing a saturation recovery method to acquire T1 maps for two different tumor models has revealed that longitudinal relaxation mapping is sensitive enough to distinguish between normal and malignant tissue. This is seen even with decreased signal/noise ratios using small voxel sizes to obtain high-resolution images. In both tumor models, it was revealed that relaxation mapping recorded significantly different relaxation values between regions encapsulating the tumor, muscle, kidney, or spleen, as well as between the cell lines themselves. This indicates a potential future application of relaxation mapping as a method to fingerprint various stages of tumor development and may prove a useful measure to identify micro-metastases.

Diagnostic Performance Evaluation of Multiparametric Magnetic Resonance Imaging in the Detection of Prostate Cancer with Supervised Machine Learning Methods

Prostate cancer is the second leading cause of cancer-related death in men. Its early and correct diagnosis is of particular importance to controlling and preventing the disease from spreading to other tissues. Artificial intelligence and machine learning have effectively detected and graded several cancers, in particular prostate cancer. The purpose of this review is to show the diagnostic performance (accuracy and area under the curve) of supervised machine learning algorithms in detecting prostate cancer using multiparametric MRI. A comparison was made between the performances of different supervised machine-learning methods. This review study was performed on the recent literature sourced from scientific citation websites such as Google Scholar, PubMed, Scopus, and Web of Science up to the end of January 2023. The findings of this review reveal that supervised machine learning techniques have good performance with high accuracy and area under the curve for prostate cancer diagnosis and prediction using multiparametric MR imaging. Among supervised machine learning methods, deep learning, random forest, and logistic regression algorithms appear to have the best performance.

A Review of the Recent Advances in Alzheimer's Disease Research and the Utilization of Network Biology Approaches for Prioritizing Diagnostics and Therapeutics

Alzheimer's disease (AD) is a polygenic multifactorial neurodegenerative disease that, after decades of research and development, is still without a cure. There are some symptomatic treatments to manage the psychological symptoms but none of these drugs can halt disease progression. Additionally, over the last few years, many anti-AD drugs failed in late stages of clinical trials and many hypotheses surfaced to explain these failures, including the lack of clear understanding of disease pathways and processes. Recently, different epigenetic factors have been implicated in AD pathogenesis; thus, they could serve as promising AD diagnostic biomarkers. Additionally, network biology approaches have been suggested as effective tools to study AD on the systems level and discover multi-target-directed ligands as novel treatments for AD. Herein, we provide a comprehensive review on Alzheimer's disease pathophysiology to provide a better understanding of disease pathogenesis hypotheses and decipher the role of genetic and epigenetic factors in disease development and progression. We also provide an overview of disease biomarkers and drug targets and suggest network biology approaches as new tools for identifying novel biomarkers and drugs. We also posit that the application of machine learning and artificial intelligence to mining Alzheimer's disease multi-omics data will facilitate drug and biomarker discovery efforts and lead to effective individualized anti-Alzheimer treatments.

Unlocking the Potential of the Human Microbiome for Identifying Disease Diagnostic Biomarkers

The human microbiome encodes more than three million genes, outnumbering human genes by more than 100 times, while microbial cells in the human microbiota outnumber human cells by 10 times. Thus, the human microbiota and related microbiome constitute a vast source for identifying disease biomarkers and therapeutic drug targets. Herein, we review the evidence backing the exploitation of the human microbiome for identifying diagnostic biomarkers for human disease. We describe the importance of the human microbiome in health and disease and detail the use of the human microbiome and microbiota metabolites as potential diagnostic biomarkers for multiple diseases, including cancer, as well as inflammatory, neurological, and metabolic diseases. Thus, the human microbiota has enormous potential to pave the road for a new era in biomarker research for diagnostic and therapeutic purposes. The scientific community needs to collaborate to overcome current challenges in microbiome research concerning the lack of standardization of research methods and the lack of understanding of causal relationships between microbiota and human disease.

Recent Advances in Device Engineering and Computational Analysis for Characterization of Cell-Released Cancer Biomarkers

During cancer progression, tumors shed different biomarkers into the bloodstream, including circulating tumor cells (CTCs), extracellular vesicles (EVs), circulating cell-free DNA (cfDNA), and circulating tumor DNA (ctDNA). The analysis of these biomarkers in the blood, known as 'liquid biopsy' (LB), is a promising approach for early cancer detection and treatment monitoring, and more recently, as a means for cancer therapy. Previous reviews have discussed the role of CTCs and ctDNA in cancer progression; however, ctDNA and EVs are rapidly evolving with technological advancements and computational analysis and are the subject of enormous recent studies in cancer biomarkers. In this review, first, we introduce these cell-released cancer biomarkers and briefly discuss their clinical significance in cancer diagnosis and treatment monitoring. Second, we present conventional and novel approaches for the isolation, profiling, and characterization of these markers. We then investigate the mathematical and in silico models that are developed to investigate the function of ctDNA and EVs in cancer progression. We convey our views on what is needed to pave the way to translate the emerging technologies and models into the clinic and make the case that optimized next-generation techniques and models are needed to precisely evaluate the clinical relevance of these LB markers.

Transcriptome-based stemness indices analysis reveals platinum-based chemo-theraputic response indicators in advanced-stage serous ovarian cancer

Serous ovarian cancer (SOC) is a main histological subtype of ovarian cancer, in which cancer stem cells (CSC) are responsible for its chemoresistance. However, the underlying modulation mechanisms of chemoresistance led by cancer stemness are still undefined. We aimed to investigate potential drug-response indicators among stemness-associated biomarkers in advanced SOC samples. The mRNA expression-based stemness index (mRNAsi) of The Cancer Genome Atlas (TCGA) was evaluated and corrected by tumor purity. Weighted gene co-expression network analysis (WGCNA) was utilized to explore the gene modules and key genes involved in stemness characteristics. We found that mRNAsi and corrected mRNAsi scores were both greater in tumors of Grade 3 and 4 than that of Grade 1 and 2. Forty-two key genes were obtained from the most significant mRNAsi-related gene module. Functional annotation revealed that these key genes were mainly involved in the mitotic division. Thirteen potential platinum-response indicators were selected from the genes enriched to platinum-response associated pathways. Among them, we identified 11 genes with prognostic value of progression-free survival (PFS) in advanced SOC patients treated with platinum and 7 prognostic genes in patients treated with a combination of platinum and taxol. The expressions of the 13 key genes were also validated between platinum-resistant and -sensitive SOC samples of advanced stages in two Gene Expression Omnibus (GEO) datasets. The results revealed that CDC20 was a potential platinum-sensitivity indicator in advanced SOC. These findings may provide a new insight for chemotherapies in advanced SOC patients clinically.