Heterogeneous data fusion for brain tumor classification

Assessment and prediction of glioblastoma therapy response: challenges and opportunities

Glioblastoma is the most aggressive type of primary adult brain tumour. The median survival of patients with glioblastoma remains approximately 15 months, and the 5-year survival rate is <10%. Current treatment options are limited, and the standard of care has remained relatively constant since 2011. Over the last decade, a range of different treatment regimens have been investigated with very limited success. Tumour recurrence is almost inevitable with the current treatment strategies, as glioblastoma tumours are highly heterogeneous and invasive. Additionally, another challenging issue facing patients with glioblastoma is how to distinguish between tumour progression and treatment effects, especially when relying on routine diagnostic imaging techniques in the clinic. The specificity of routine imaging for identifying tumour progression early or in a timely manner is poor due to the appearance similarity of post-treatment effects. Here, we concisely describe the current status and challenges in the assessment and early prediction of therapy response and the early detection of tumour progression or recurrence. We also summarize and discuss studies of advanced approaches such as quantitative imaging, liquid biomarker discovery and machine intelligence that hold exceptional potential to aid in the therapy monitoring of this malignancy and early prediction of therapy response, which may decisively transform the conventional detection methods in the era of precision medicine.

A Review on Data Fusion of Multidimensional Medical and Biomedical Data

Data fusion aims to provide a more accurate description of a sample than any one source of data alone. At the same time, data fusion minimizes the uncertainty of the results by combining data from multiple sources. Both aim to improve the characterization of samples and might improve clinical diagnosis and prognosis. In this paper, we present an overview of the advances achieved over the last decades in data fusion approaches in the context of the medical and biomedical fields. We collected approaches for interpreting multiple sources of data in different combinations: image to image, image to biomarker, spectra to image, spectra to spectra, spectra to biomarker, and others. We found that the most prevalent combination is the image-to-image fusion and that most data fusion approaches were applied together with deep learning or machine learning methods.

Research Progress of Gliomas in Machine Learning

In the field of gliomas research, the broad availability of genetic and image information originated by computer technologies and the booming of biomedical publications has led to the advent of the big-data era. Machine learning methods were applied as possible approaches to speed up the data mining processes. In this article, we reviewed the present situation and future orientations of machine learning application in gliomas within the context of workflows to integrate analysis for precision cancer care. Publicly available tools or algorithms for key machine learning technologies in the literature mining for glioma clinical research were reviewed and compared. Further, the existing solutions of machine learning methods and their limitations in glioma prediction and diagnostics, such as overfitting and class imbalanced, were critically analyzed.

Machine learning approaches to study glioblastoma: A review of the last decade of applications

BACKGROUND

Glioblastoma (GB, formally glioblastoma multiforme) is a malignant type of brain cancer that currently has no cure and is characterized by being highly heterogeneous with high rates of re-incidence and therapy resistance. Thus, it is urgent to characterize the mechanisms of GB pathogenesis to help researchers identify novel therapeutic targets to cure this devastating disease. Recently, a promising approach to identifying novel therapeutic targets is the integration of tumor omics data with clinical information using machine learning (ML) techniques.

RECENT FINDINGS

ML has become a valuable addition to the researcher's toolbox, thanks to its flexibility, multidimensional approach, and a growing community of users. The goal of this review is to introduce basic concepts and applications of ML for studying GB to clinicians and practitioners who are new to data science. ML applications include exploring large data sets, finding new relevant patterns, predicting outcomes, or merely understanding associations of the complex molecular networks presented within the tumor. Here, we review ML applications published between 2008 and 2018 and discuss ML strategies intending to identify new potential therapeutic targets to improve the management and treatment of GB.

CONCLUSIONS

ML applications to study GB vary in purpose and complexity, with positive results. In GB studies, ML is often used to analyze high-dimensional datasets with prediction or classification as a primary goal. Despite the strengths of ML techniques, they are not fail-safe and methodological issues can occur in GB studies that use them. This is why researchers need to be aware of these issues when planning and appraising studies that apply ML to the study of GB.

Metabolomic prostate cancer fields in HRMAS MRS-profiled histologically benign tissue vary with cancer status and distance from cancer

In this article, we review the state of the field of high resolution magic angle spinning MRS (HRMAS MRS)-based cancer metabolomics since its beginning in 2004; discuss the concept of cancer metabolomic fields, where metabolomic profiles measured from histologically benign tissues reflect patient cancer status; and report our HRMAS MRS metabolomic results, which characterize metabolomic fields in prostatectomy-removed cancerous prostates. Three-dimensional mapping of cancer lesions throughout each prostate enabled multiple benign tissue samples per organ to be classified based on distance from and extent of the closest cancer lesion as well as the Gleason score (GS) of the entire prostate. Cross-validated partial least squares-discriminant analysis separations were achieved between cancer and benign tissue, and between cancer tissue from prostates with high (≥4 + 3) and low (≤3 + 4) GS. Metabolomic field effects enabled histologically benign tissue adjacent to cancer to distinguish the GS and extent of the cancer lesion itself. Benign samples close to either low GS cancer or extensive cancer lesions could be distinguished from those far from cancer. Furthermore, a successfully cross-validated multivariate model for three benign tissue groups with varying distances from cancer lesions within one prostate indicates the scale of prostate cancer metabolomic fields. While these findings could, at present, be potentially useful in the prostate cancer clinic for analysis of biopsy or surgical specimens to complement current diagnostics, the confirmation of metabolomic fields should encourage further examination of cancer fields and can also enhance understanding of the metabolomic characteristics of cancer in myriad organ systems. Our results together with the success of HRMAS MRS-based cancer metabolomics presented in our literature review demonstrate the potential of cancer metabolomics to provide supplementary information for cancer diagnosis, staging, and patient prognostication.

Applications of high-resolution magic angle spinning MRS in biomedical studies II-Human diseases

High-resolution magic angle spinning (HRMAS) MRS is a powerful method for gaining insight into the physiological and pathological processes of cellular metabolism. Given its ability to obtain high-resolution spectra of non-liquid biological samples, while preserving tissue architecture for subsequent histopathological analysis, the technique has become invaluable for biochemical and biomedical studies. Using HRMAS MRS, alterations in measured metabolites, metabolic ratios, and metabolomic profiles present the possibility to improve identification and prognostication of various diseases and decipher the metabolomic impact of drug therapies. In this review, we evaluate HRMAS MRS results on human tissue specimens from malignancies and non-localized diseases reported in the literature since the inception of the technique in 1996. We present the diverse applications of the technique in understanding pathological processes of different anatomical origins, correlations with in vivo imaging, effectiveness of therapies, and progress in the HRMAS methodology.

Evaluation of Cancer Metabolomics Using ex vivo High Resolution Magic Angle Spinning (HRMAS) Magnetic Resonance Spectroscopy (MRS)

According to World Health Organization (WHO) estimates, cancer is responsible for more deaths than all coronary heart disease or stroke worldwide, serving as a major public health threat around the world. High resolution magic angle spinning (HRMAS) magnetic resonance spectroscopy (MRS) has demonstrated its usefulness in the identification of cancer metabolic markers with the potential to improve diagnosis and prognosis for the oncology clinic, due partially to its ability to preserve tissue architecture for subsequent histological and molecular pathology analysis. Capable of the quantification of individual metabolites, ratios of metabolites, and entire metabolomic profiles, HRMAS MRS is one of the major techniques now used in cancer metabolomic research. This article reviews and discusses literature reports of HRMAS MRS studies of cancer metabolomics published between 2010 and 2015 according to anatomical origins, including brain, breast, prostate, lung, gastrointestinal, and neuroendocrine cancers. These studies focused on improving diagnosis and understanding patient prognostication, monitoring treatment effects, as well as correlating with the use of in vivo MRS in cancer clinics.