How Machine Learning Will Transform Biomedicine

Abstract

This Perspective explores the application of machine learning toward improved diagnosis and treatment. We outline a vision for how machine learning can transform three broad areas of biomedicine: clinical diagnostics, precision treatments, and health monitoring, where the goal is to maintain health through a range of diseases and the normal aging process. For each area, early instances of successful machine learning applications are discussed, as well as opportunities and challenges for machine learning. When these challenges are met, machine learning promises a future of rigorous, outcomes-based medicine with detection, diagnosis, and treatment strategies that are continuously adapted to individual and environmental differences.

An innovative approach for HLA typing, molecular tumor testing and the validation of tumor exclusive antigens

“…However, with a larger database, we assume subjective evaluation will be dispensable soon. In future, machine learning might revolutionize and improve clinical diagnostics, precision treatments and health monitoring 52 . The immunopeptidome, with its substantial diversity and detailed immunological information, which can be structured and made comprehensible with the help of artificial intelligence, has the potential to become an important part of it.…”

Method development for cross-study microbiome data mining: Challenges and opportunities

“…Since microbial communities shape the dynamics of ecological systems, ranging from the human gut to the marine, one potential of microbiome is linking variation of microbial composition to phenotypic and physiological statuses, which can inspire the development of new techniques for disease diagnosis, ecological dysbiosis detection and treatment evaluation. Previous studies have demonstrated the feasibility of ML methods [18] , [76] in disease detection and classification with human-associated microbiome data for inflammatory bowel disease (IBD) [77] , colorectal cancer (CRC) [19] , caries [78] , etc., by extreme gradient boosting (XGBoost), random forest (RF), support vector machine (SVM), k-nearest neighbor (KNN) and other ML algorithms. As a quantitative approach, the ML-based indices are also designed to assess the risks for potential diseases and to evaluate the effects among different treatments [79] , [80] .…”

“…Newly developed bioinformatics tools are bringing opportunities in deciphering the microbiome data, from general-purpose algorithms such as sequence alignment and machine learning (ML), to microbiome-specific approaches like operational taxonomy unit (OTU) picking [12] and phylogeny-based distance metrics [13] , [14] . On the other hand, challenges have also already been placed by the vast volume of microbiome data, especially in integration of datasets produced by multiple studies and platforms [15] , comparison among samples [16] and status or disease classification and prediction by training on large-scale datasets [17] , [18] .…”

Biomarkers in Early Diagnosis and Early Stage Lung Cancer: The Clinician’s Point of View

“…In this way, comparisons can be drawn between data from other countries for the common purpose of creating a worldwide Big Data database [ 65 ]. Artificial intelligence [ 66 , 67 ] will be used to process, overlap, and integrate the molecular, clinical, and epidemiological data, and machine learning [ 68 , 69 ] will be employed to produce realistic multiple diagnostic algorithms which the clinician can readily use for diagnostic, preventive, and prognostic purposes.…”