Artificial Intelligence in Precision Medicine
Artificial intelligence in precision medicine is a revolutionary new approach advancing health and wellness, knowledge, and health care delivery to maximize the quality of life for all over a lifetime. The main concept of precision medicine is providing health care which is individually tailored on the basis of a person’s genes, lifestyle and environment. With the advances in genetics, artificial intelligence and the growing availability of health data, present an opportunity to make precise personalized patient care a clinical reality.
It is like cricket. No two cricket ball deliveries, players, — or patients — are exactly alike. No two games or diseases are exactly the same. To win the game every ball, every delivery needs unique strategy. Precision medicine is like that. No two diseases are same, so the treatments will be different and unique.
AI with precision medicine is a part of artificial intelligence in health care. It brings together innovations in genomics, metabolomics, mobile health, biomedical data sciences, imaging, social engagement and networking, communication, and environmental sciences to make diagnostics, therapeutics, and prevention more individualized, proactive, predictive, and precise.
Precision medicine often involves the application of panomic analysis and systems biology to analyze the cause of an individual patient’s disease at the molecular level and then to utilize targeted treatments (possibly in combination) to address that individual patient’s disease process. Here Panomicsrefers to the range of molecular biology technologies including genomics, proteomics, metabolomics, transcriptomics, and so forth, or the integration of their combined use. Systems biology approaches are often based upon the use of panomic analysis data.
The the National Research Council explains it as: “Precision Medicine refers to the tailoring of medical treatment to the individual characteristics of each patient. It does not literally mean the creation of drugs or medical devices that are unique to a patient, but rather the ability to classify individuals into sub-populations that differ in their susceptibility to a particular disease, in the biology or prognosis of those diseases they may develop, or in their response to a specific treatment.
Preventive or therapeutic interventions can then be concentrated on those who will benefit, sparing expense and side effects for those who will not. Although the term ‘personalized medicine’ is also used to convey this meaning, that term is sometimes misinterpreted as implying that unique treatments can be designed for each individual.”[1]
Artificial intelligence in precision medicine combines genetic, proteomic, structural, and computational methods to proceed from patient-based systems data, such as genome-wide association studies, to functional complexes, to pathways, and ultimately to predictive networks.
Availability of Large Data for Machine Learning
Large data sets are now available through international collaborative projects, such as the 1000 Genomes Project, the 100,000 Genomes Project, ENCODE, the Roadmap Epigenomics Project and the US National Institutes of Health’s 4D Nucleome Initiative. These large data offers an opportunity for machine learning to have a significant impact on biology and precision medicine.
Artificial Intelligence for Genome Sequencing
Genes are the building blocks of life. When a gene is expressed, it is first transcribed into an RNA sequence, and the RNA is then translated into a protein, a sequence of amino acids. Normally, for small scale, biologists measure the protein-production rate directly, but that is very difficult and impractical on a large scale. Alternatively, machine learning algorithms process can measure gene-expression levels more efficiently.
The genotype is the set of genes in our DNA which is responsible for a particular trait. The phenotype is the physical expression, or characteristics, of that trait. A clinical phenotype is the presentation of a disease in a given individual. A clinical phenotype may be result from complex, nonlinear, and often stochastic interactions among various factors that contribute to the phenotype. Even a single gene disorder can cause a disease. Common diseases such as coronary artery disease and hypertension are caused, at the genetic level, by a combination of common and rare variants. Predicting clinical phenotype requires knowledge of the genetic diversity of the humans, etiological complexity of the clinical phenotype, and phenotypic variability of the diseases and artificial intelligence has a crucial role to play.
Artificial Intelligence for Genomics sequencing help medical professionals to interpret how genetic variation affects metabolism, DNA repair, and cell growth. Machine learning algorithms are designed based on patterns identified in large genetic data sets. Generally, machine learning methods generally have one of two goals: prediction or interpretation. Gene-expression microarrays, commonly called “gene chips”, make it possible to measure the rate at which a cell or tissue is expressing itself in translating into a DNA-to-RNA to protein. Artificial intelligence gives the ability to measure at once the transcription of all the genes in an organism. For this, the amount of data that biologists need to examine is overwhelming. Finding some combination of genes whose expression levels can distinguish the groups of patients is a daunting task for a human, but a relatively natural one for a machine-learning algorithm.
Source
Compassionate Artificial Superintelligene AI-5.0, Dr. Amit Ray, 2018
References:
- Toward Precision Medicine: Building a Knowledge Network for Biomedical Research and a New Taxonomy of Disease
- Machine learning applications in genetics and genomics, Nature Reviews Genetics volume 16, pages 321–332 (2015).