Early Detection of Cancer Using Data Sciences
In 2014 we developed a Big-Data genomic algorithm called SPKMG (Sequence Per Kilobase of exon, per Megabase of the mappable Genome) to trace cancer evolution and discover the roots of cancer. In 2016 we used SPKMG algorithm on 11 non-BRCA1/BRCA2 familial breast Breast Cancer patients’ WES (Whole Exome Sequence) data and published our results in a scientific journals [Tracking Cancer Genetic Evolution using OncoTrack: https://www.nature.com/articles/srep29647]. These WES data of each cancer patient ranges from 12 to 15 Giga Bytes. We have since enhanced the SPKMG algorithm with other AI algorithms, which is now called DNA Dosages. We tested DNA Dosages suite of algorithms on 252 cancer patients from four cancer types viz., Esophageal Squamous Cell Carcinoma. Breast Cancer, Bladder Cancer. and Lung Adenocarcinoma. Our DNA dosages set of algorithms use Data Sciences extensively starting from Unsupervised Machine Learning to Knowledge Engineering – it is now able to sense “Concerization Field” hypothesized by DP Slaughter and team in 1953. Slaughter and team discovered that there exists a “Field” of cancerization agents, which extends beyond the tumor boundaries. Using Data Science and AI we are able to prove that Slaughter’s hypothesis is not only true but the “Field of Cancerization” extends beyond the primary cancer tissue and can be sensed even in common blood. Using these data science algorithms we can now find traces of cancer in peripheral whole blood (common blood) – this technique is named Next Generation Liquid Biopsy (NGLB). In this talk we will present how data science is used to detect cancer through blood test at a very early stage of cancer, which will facilitate cure for cancer.
Outline/Structure of the Talk
Using Data Science for Early Detection of Cancer
1) How data science can be used for Genomic data
2) How data sciences can be used on solving complex diseases like cancer
Intermediate knowledge of data science